-
Artificial Intelligence-Based Opportunistic Coronary Calcium Screening in the Veterans Affairs National Healthcare System
Authors:
Raffi Hagopian,
Timothy Strebel,
Simon Bernatz,
Gregory A Myers,
Erik Offerman,
Eric Zuniga,
Cy Y Kim,
Angie T Ng,
James A Iwaz,
Sunny P Singh,
Evan P Carey,
Michael J Kim,
R Spencer Schaefer,
Jeannie Yu,
Amilcare Gentili,
Hugo JWL Aerts
Abstract:
Coronary artery calcium (CAC) is highly predictive of cardiovascular events. While millions of chest CT scans are performed annually in the United States, CAC is not routinely quantified from scans done for non-cardiac purposes. A deep learning algorithm was developed using 446 expert segmentations to automatically quantify CAC on non-contrast, non-gated CT scans (AI-CAC). Our study differs from p…
▽ More
Coronary artery calcium (CAC) is highly predictive of cardiovascular events. While millions of chest CT scans are performed annually in the United States, CAC is not routinely quantified from scans done for non-cardiac purposes. A deep learning algorithm was developed using 446 expert segmentations to automatically quantify CAC on non-contrast, non-gated CT scans (AI-CAC). Our study differs from prior works as we leverage imaging data across the Veterans Affairs national healthcare system, from 98 medical centers, capturing extensive heterogeneity in imaging protocols, scanners, and patients. AI-CAC performance on non-gated scans was compared against clinical standard ECG-gated CAC scoring. Non-gated AI-CAC differentiated zero vs. non-zero and less than 100 vs. 100 or greater Agatston scores with accuracies of 89.4% (F1 0.93) and 87.3% (F1 0.89), respectively, in 795 patients with paired gated scans within a year of a non-gated CT scan. Non-gated AI-CAC was predictive of 10-year all-cause mortality (CAC 0 vs. >400 group: 25.4% vs. 60.2%, Cox HR 3.49, p < 0.005), and composite first-time stroke, MI, or death (CAC 0 vs. >400 group: 33.5% vs. 63.8%, Cox HR 3.00, p < 0.005). In a screening dataset of 8,052 patients with low-dose lung cancer-screening CTs (LDCT), 3,091/8,052 (38.4%) individuals had AI-CAC >400. Four cardiologists qualitatively reviewed LDCT images from a random sample of >400 AI-CAC patients and verified that 527/531 (99.2%) would benefit from lipid-lowering therapy. To the best of our knowledge, this is the first non-gated CT CAC algorithm developed across a national healthcare system, on multiple imaging protocols, without filtering intra-cardiac hardware, and compared against a strong gated CT reference. We report superior performance relative to previous CAC algorithms evaluated against paired gated scans that included patients with intra-cardiac hardware.
△ Less
Submitted 15 September, 2024;
originally announced September 2024.
-
Lens functions for exploring UMAP Projections with Domain Knowledge
Authors:
Daniel M. Bot,
Jan Aerts
Abstract:
Dimensionality reduction algorithms are often used to visualise high-dimensional data. Previously, studies have used prior information to enhance or suppress expected patterns in projections. In this paper, we adapt such techniques for domain knowledge guided interactive exploration. Inspired by Mapper and STAD, we present three types of lens functions for UMAP, a state-of-the-art dimensionality r…
▽ More
Dimensionality reduction algorithms are often used to visualise high-dimensional data. Previously, studies have used prior information to enhance or suppress expected patterns in projections. In this paper, we adapt such techniques for domain knowledge guided interactive exploration. Inspired by Mapper and STAD, we present three types of lens functions for UMAP, a state-of-the-art dimensionality reduction algorithm. Lens functions enable analysts to adapt projections to their questions, revealing otherwise hidden patterns. They filter the modelled connectivity to explore the interaction between manually selected features and the data's structure, creating configurable perspectives each potentially revealing new insights. The effectiveness of the lens functions is demonstrated in two use cases and their computational cost is analysed in a synthetic benchmark. Our implementation is available in an open-source Python package: https://github.com/vda-lab/lensed_umap.
△ Less
Submitted 15 May, 2024;
originally announced May 2024.
-
Incorporating Anatomical Awareness for Enhanced Generalizability and Progression Prediction in Deep Learning-Based Radiographic Sacroiliitis Detection
Authors:
Felix J. Dorfner,
Janis L. Vahldiek,
Leonhard Donle,
Andrei Zhukov,
Lina Xu,
Hartmut Häntze,
Marcus R. Makowski,
Hugo J. W. L. Aerts,
Fabian Proft,
Valeria Rios Rodriguez,
Judith Rademacher,
Mikhail Protopopov,
Hildrun Haibel,
Torsten Diekhoff,
Murat Torgutalp,
Lisa C. Adams,
Denis Poddubnyy,
Keno K. Bressem
Abstract:
Purpose: To examine whether incorporating anatomical awareness into a deep learning model can improve generalizability and enable prediction of disease progression.
Methods: This retrospective multicenter study included conventional pelvic radiographs of 4 different patient cohorts focusing on axial spondyloarthritis (axSpA) collected at university and community hospitals. The first cohort, whic…
▽ More
Purpose: To examine whether incorporating anatomical awareness into a deep learning model can improve generalizability and enable prediction of disease progression.
Methods: This retrospective multicenter study included conventional pelvic radiographs of 4 different patient cohorts focusing on axial spondyloarthritis (axSpA) collected at university and community hospitals. The first cohort, which consisted of 1483 radiographs, was split into training (n=1261) and validation (n=222) sets. The other cohorts comprising 436, 340, and 163 patients, respectively, were used as independent test datasets. For the second cohort, follow-up data of 311 patients was used to examine progression prediction capabilities. Two neural networks were trained, one on images cropped to the bounding box of the sacroiliac joints (anatomy-aware) and the other one on full radiographs. The performance of the models was compared using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity.
Results: On the three test datasets, the standard model achieved AUC scores of 0.853, 0.817, 0.947, with an accuracy of 0.770, 0.724, 0.850. Whereas the anatomy-aware model achieved AUC scores of 0.899, 0.846, 0.957, with an accuracy of 0.821, 0.744, 0.906, respectively. The patients who were identified as high risk by the anatomy aware model had an odds ratio of 2.16 (95% CI: 1.19, 3.86) for having progression of radiographic sacroiliitis within 2 years.
Conclusion: Anatomical awareness can improve the generalizability of a deep learning model in detecting radiographic sacroiliitis. The model is published as fully open source alongside this study.
△ Less
Submitted 12 May, 2024;
originally announced May 2024.
-
SynthBrainGrow: Synthetic Diffusion Brain Aging for Longitudinal MRI Data Generation in Young People
Authors:
Anna Zapaishchykova,
Benjamin H. Kann,
Divyanshu Tak,
Zezhong Ye,
Daphne A. Haas-Kogan,
Hugo J. W. L. Aerts
Abstract:
Synthetic longitudinal brain MRI simulates brain aging and would enable more efficient research on neurodevelopmental and neurodegenerative conditions. Synthetically generated, age-adjusted brain images could serve as valuable alternatives to costly longitudinal imaging acquisitions, serve as internal controls for studies looking at the effects of environmental or therapeutic modifiers on brain de…
▽ More
Synthetic longitudinal brain MRI simulates brain aging and would enable more efficient research on neurodevelopmental and neurodegenerative conditions. Synthetically generated, age-adjusted brain images could serve as valuable alternatives to costly longitudinal imaging acquisitions, serve as internal controls for studies looking at the effects of environmental or therapeutic modifiers on brain development, and allow data augmentation for diverse populations. In this paper, we present a diffusion-based approach called SynthBrainGrow for synthetic brain aging with a two-year step. To validate the feasibility of using synthetically-generated data on downstream tasks, we compared structural volumetrics of two-year-aged brains against synthetically-aged brain MRI. Results show that SynthBrainGrow can accurately capture substructure volumetrics and simulate structural changes such as ventricle enlargement and cortical thinning. Our approach provides a novel way to generate longitudinal brain datasets from cross-sectional data to enable augmented training and benchmarking of computational tools for analyzing lifespan trajectories. This work signifies an important advance in generative modeling to synthesize realistic longitudinal data with limited lifelong MRI scans. The code is available at XXX.
△ Less
Submitted 22 February, 2024;
originally announced May 2024.
-
Fabrication of planar halfmetallic ferromagnetic Josephson junctions with long range coupling
Authors:
Junxiang Yao,
Jan Aarts
Abstract:
Superconducting junctions with a ferromagnet as the weak link, where triplet correlations can transport supercurrents over a substantial distance, have been of long-standing interest. In this work, we study the triplet transport in planar La$_{0.7}$Sr$_{0.3}$MnO$_3$ (LSMO) nanowire Josephson junctions with NbTi superconducting contacts. By meticulous ion etching with an artificial Pt hard mask, th…
▽ More
Superconducting junctions with a ferromagnet as the weak link, where triplet correlations can transport supercurrents over a substantial distance, have been of long-standing interest. In this work, we study the triplet transport in planar La$_{0.7}$Sr$_{0.3}$MnO$_3$ (LSMO) nanowire Josephson junctions with NbTi superconducting contacts. By meticulous ion etching with an artificial Pt hard mask, the NbTi/LSMO bilayer is structured to form an LSMO bridge without damaging its top layer. We observe superconducting (critical) currents of the order of 10$^{9}$ A/m$^2$ in a junction with a length of 1.3 $μ$m, and distinguishing superconducting quantum interference (SQI) patterns when sweeping a magnetic field perpendicular ($B_\perp$) to the plane of the wire or parallel ($B_\parallel$) to the plane and along the wire. The observed Gaussian-shaped SQI pattern is attributed to the diffusive transport of triplet pairs in the LSMO. Our work demonstrates that combinations of oxide magnets with conventional ($s$-wave) alloy superconductors can be a promising new route to realizing superconducting spintronics.
△ Less
Submitted 24 April, 2024;
originally announced April 2024.
-
Magnetic resonance delta radiomics to track radiation response in lung tumors receiving stereotactic MRI-guided radiotherapy
Authors:
Yining Zha,
Benjamin H. Kann,
Zezhong Ye,
Anna Zapaishchykova,
John He,
Shu-Hui Hsu,
Jonathan E. Leeman,
Kelly J. Fitzgerald,
David E. Kozono,
Raymond H. Mak,
Hugo J. W. L. Aerts
Abstract:
Introduction: Lung cancer is a leading cause of cancer-related mortality, and stereotactic body radiotherapy (SBRT) has become a standard treatment for early-stage lung cancer. However, the heterogeneous response to radiation at the tumor level poses challenges. Currently, standardized dosage regimens lack adaptation based on individual patient or tumor characteristics. Thus, we explore the potent…
▽ More
Introduction: Lung cancer is a leading cause of cancer-related mortality, and stereotactic body radiotherapy (SBRT) has become a standard treatment for early-stage lung cancer. However, the heterogeneous response to radiation at the tumor level poses challenges. Currently, standardized dosage regimens lack adaptation based on individual patient or tumor characteristics. Thus, we explore the potential of delta radiomics from on-treatment magnetic resonance (MR) imaging to track radiation dose response, inform personalized radiotherapy dosing, and predict outcomes. Methods: A retrospective study of 47 MR-guided lung SBRT treatments for 39 patients was conducted. Radiomic features were extracted using Pyradiomics, and stability was evaluated temporally and spatially. Delta radiomics were correlated with radiation dose delivery and assessed for associations with tumor control and survival with Cox regressions. Results: Among 107 features, 49 demonstrated temporal stability, and 57 showed spatial stability. Fifteen stable and non-collinear features were analyzed. Median Skewness and surface to volume ratio decreased with radiation dose fraction delivery, while coarseness and 90th percentile values increased. Skewness had the largest relative median absolute changes (22%-45%) per fraction from baseline and was associated with locoregional failure (p=0.012) by analysis of covariance. Skewness, Elongation, and Flatness were significantly associated with local recurrence-free survival, while tumor diameter and volume were not. Conclusions: Our study establishes the feasibility and stability of delta radiomics analysis for MR-guided lung SBRT. Findings suggest that MR delta radiomics can capture short-term radiographic manifestations of intra-tumoral radiation effect.
△ Less
Submitted 23 February, 2024;
originally announced February 2024.
-
LongHealth: A Question Answering Benchmark with Long Clinical Documents
Authors:
Lisa Adams,
Felix Busch,
Tianyu Han,
Jean-Baptiste Excoffier,
Matthieu Ortala,
Alexander Löser,
Hugo JWL. Aerts,
Jakob Nikolas Kather,
Daniel Truhn,
Keno Bressem
Abstract:
Background: Recent advancements in large language models (LLMs) offer potential benefits in healthcare, particularly in processing extensive patient records. However, existing benchmarks do not fully assess LLMs' capability in handling real-world, lengthy clinical data.
Methods: We present the LongHealth benchmark, comprising 20 detailed fictional patient cases across various diseases, with each…
▽ More
Background: Recent advancements in large language models (LLMs) offer potential benefits in healthcare, particularly in processing extensive patient records. However, existing benchmarks do not fully assess LLMs' capability in handling real-world, lengthy clinical data.
Methods: We present the LongHealth benchmark, comprising 20 detailed fictional patient cases across various diseases, with each case containing 5,090 to 6,754 words. The benchmark challenges LLMs with 400 multiple-choice questions in three categories: information extraction, negation, and sorting, challenging LLMs to extract and interpret information from large clinical documents.
Results: We evaluated nine open-source LLMs with a minimum of 16,000 tokens and also included OpenAI's proprietary and cost-efficient GPT-3.5 Turbo for comparison. The highest accuracy was observed for Mixtral-8x7B-Instruct-v0.1, particularly in tasks focused on information retrieval from single and multiple patient documents. However, all models struggled significantly in tasks requiring the identification of missing information, highlighting a critical area for improvement in clinical data interpretation.
Conclusion: While LLMs show considerable potential for processing long clinical documents, their current accuracy levels are insufficient for reliable clinical use, especially in scenarios requiring the identification of missing information. The LongHealth benchmark provides a more realistic assessment of LLMs in a healthcare setting and highlights the need for further model refinement for safe and effective clinical application.
We make the benchmark and evaluation code publicly available.
△ Less
Submitted 25 January, 2024;
originally announced January 2024.
-
FLASC: A Flare-Sensitive Clustering Algorithm
Authors:
D. M. Bot,
J. Peeters,
J. Liesenborgs,
J. Aerts
Abstract:
Clustering algorithms are often used to find subpopulations in exploratory data analysis workflows. Not only the clusters themselves, but also their shape can represent meaningful subpopulations. In this paper, we present FLASC, an algorithm that detects branches within clusters to identify such subpopulations. FLASC builds upon HDBSCAN*, a state-of-the-art density-based clustering algorithm, and…
▽ More
Clustering algorithms are often used to find subpopulations in exploratory data analysis workflows. Not only the clusters themselves, but also their shape can represent meaningful subpopulations. In this paper, we present FLASC, an algorithm that detects branches within clusters to identify such subpopulations. FLASC builds upon HDBSCAN*, a state-of-the-art density-based clustering algorithm, and detects branches in a post-processing step that describes within-cluster connectivity. Two variants of the algorithm are presented, which trade computational cost for noise robustness. We show that both variants scale similarly to HDBSCAN* in terms of computational cost and provide stable outputs using synthetic data sets, resulting in an efficient flare-sensitive clustering algorithm. In addition, we demonstrate the benefit of branch-detection on two real-world data sets.
△ Less
Submitted 8 October, 2024; v1 submitted 27 November, 2023;
originally announced November 2023.
-
The impact of responding to patient messages with large language model assistance
Authors:
Shan Chen,
Marco Guevara,
Shalini Moningi,
Frank Hoebers,
Hesham Elhalawani,
Benjamin H. Kann,
Fallon E. Chipidza,
Jonathan Leeman,
Hugo J. W. L. Aerts,
Timothy Miller,
Guergana K. Savova,
Raymond H. Mak,
Maryam Lustberg,
Majid Afshar,
Danielle S. Bitterman
Abstract:
Documentation burden is a major contributor to clinician burnout, which is rising nationally and is an urgent threat to our ability to care for patients. Artificial intelligence (AI) chatbots, such as ChatGPT, could reduce clinician burden by assisting with documentation. Although many hospitals are actively integrating such systems into electronic medical record systems, AI chatbots utility and i…
▽ More
Documentation burden is a major contributor to clinician burnout, which is rising nationally and is an urgent threat to our ability to care for patients. Artificial intelligence (AI) chatbots, such as ChatGPT, could reduce clinician burden by assisting with documentation. Although many hospitals are actively integrating such systems into electronic medical record systems, AI chatbots utility and impact on clinical decision-making have not been studied for this intended use. We are the first to examine the utility of large language models in assisting clinicians draft responses to patient questions. In our two-stage cross-sectional study, 6 oncologists responded to 100 realistic synthetic cancer patient scenarios and portal messages developed to reflect common medical situations, first manually, then with AI assistance.
We find AI-assisted responses were longer, less readable, but provided acceptable drafts without edits 58% of time. AI assistance improved efficiency 77% of time, with low harm risk (82% safe). However, 7.7% unedited AI responses could severely harm. In 31% cases, physicians thought AI drafts were human-written. AI assistance led to more patient education recommendations, fewer clinical actions than manual responses. Results show promise for AI to improve clinician efficiency and patient care through assisting documentation, if used judiciously. Monitoring model outputs and human-AI interaction remains crucial for safe implementation.
△ Less
Submitted 29 November, 2023; v1 submitted 26 October, 2023;
originally announced October 2023.
-
FUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcare
Authors:
Karim Lekadir,
Aasa Feragen,
Abdul Joseph Fofanah,
Alejandro F Frangi,
Alena Buyx,
Anais Emelie,
Andrea Lara,
Antonio R Porras,
An-Wen Chan,
Arcadi Navarro,
Ben Glocker,
Benard O Botwe,
Bishesh Khanal,
Brigit Beger,
Carol C Wu,
Celia Cintas,
Curtis P Langlotz,
Daniel Rueckert,
Deogratias Mzurikwao,
Dimitrios I Fotiadis,
Doszhan Zhussupov,
Enzo Ferrante,
Erik Meijering,
Eva Weicken,
Fabio A González
, et al. (95 additional authors not shown)
Abstract:
Despite major advances in artificial intelligence (AI) for medicine and healthcare, the deployment and adoption of AI technologies remain limited in real-world clinical practice. In recent years, concerns have been raised about the technical, clinical, ethical and legal risks associated with medical AI. To increase real world adoption, it is essential that medical AI tools are trusted and accepted…
▽ More
Despite major advances in artificial intelligence (AI) for medicine and healthcare, the deployment and adoption of AI technologies remain limited in real-world clinical practice. In recent years, concerns have been raised about the technical, clinical, ethical and legal risks associated with medical AI. To increase real world adoption, it is essential that medical AI tools are trusted and accepted by patients, clinicians, health organisations and authorities. This work describes the FUTURE-AI guideline as the first international consensus framework for guiding the development and deployment of trustworthy AI tools in healthcare. The FUTURE-AI consortium was founded in 2021 and currently comprises 118 inter-disciplinary experts from 51 countries representing all continents, including AI scientists, clinicians, ethicists, and social scientists. Over a two-year period, the consortium defined guiding principles and best practices for trustworthy AI through an iterative process comprising an in-depth literature review, a modified Delphi survey, and online consensus meetings. The FUTURE-AI framework was established based on 6 guiding principles for trustworthy AI in healthcare, i.e. Fairness, Universality, Traceability, Usability, Robustness and Explainability. Through consensus, a set of 28 best practices were defined, addressing technical, clinical, legal and socio-ethical dimensions. The recommendations cover the entire lifecycle of medical AI, from design, development and validation to regulation, deployment, and monitoring. FUTURE-AI is a risk-informed, assumption-free guideline which provides a structured approach for constructing medical AI tools that will be trusted, deployed and adopted in real-world practice. Researchers are encouraged to take the recommendations into account in proof-of-concept stages to facilitate future translation towards clinical practice of medical AI.
△ Less
Submitted 8 July, 2024; v1 submitted 11 August, 2023;
originally announced September 2023.
-
Challenges and Opportunities in Data Visualization Education: A Call to Action
Authors:
Benjamin Bach,
Mandy Keck,
Fateme Rajabiyazdi,
Tatiana Losev,
Isabel Meirelles,
Jason Dykes,
Robert S. Laramee,
Mashael AlKadi,
Christina Stoiber,
Samuel Huron,
Charles Perin,
Luiz Morais,
Wolfgang Aigner,
Doris Kosminsky,
Magdalena Boucher,
Søren Knudsen,
Areti Manataki,
Jan Aerts,
Uta Hinrichs,
Jonathan C. Roberts,
Sheelagh Carpendale
Abstract:
This paper is a call to action for research and discussion on data visualization education. As visualization evolves and spreads through our professional and personal lives, we need to understand how to support and empower a broad and diverse community of learners in visualization. Data Visualization is a diverse and dynamic discipline that combines knowledge from different fields, is tailored to…
▽ More
This paper is a call to action for research and discussion on data visualization education. As visualization evolves and spreads through our professional and personal lives, we need to understand how to support and empower a broad and diverse community of learners in visualization. Data Visualization is a diverse and dynamic discipline that combines knowledge from different fields, is tailored to suit diverse audiences and contexts, and frequently incorporates tacit knowledge. This complex nature leads to a series of interrelated challenges for data visualization education. Driven by a lack of consolidated knowledge, overview, and orientation for visualization education, the 21 authors of this paper-educators and researchers in data visualization-identify and describe 19 challenges informed by our collective practical experience. We organize these challenges around seven themes People, Goals & Assessment, Environment, Motivation, Methods, Materials, and Change. Across these themes, we formulate 43 research questions to address these challenges. As part of our call to action, we then conclude with 5 cross-cutting opportunities and respective action items: embrace DIVERSITY+INCLUSION, build COMMUNITIES, conduct RESEARCH, act AGILE, and relish RESPONSIBILITY. We aim to inspire researchers, educators and learners to drive visualization education forward and discuss why, how, who and where we educate, as we learn to use visualization to address challenges across many scales and many domains in a rapidly changing world: viseducationchallenges.github.io.
△ Less
Submitted 15 August, 2023;
originally announced August 2023.
-
Condition-Based Production for Stochastically Deteriorating Systems: Optimal Policies and Learning
Authors:
Collin Drent,
Melvin Drent,
Joachim Arts
Abstract:
Production systems deteriorate stochastically due to usage and may eventually break down, resulting in high maintenance costs at scheduled maintenance moments. This deterioration behavior is affected by the system's production rate. While producing at a higher rate generates more revenue, the system may also deteriorate faster. Production should thus be controlled dynamically to trade-off deterior…
▽ More
Production systems deteriorate stochastically due to usage and may eventually break down, resulting in high maintenance costs at scheduled maintenance moments. This deterioration behavior is affected by the system's production rate. While producing at a higher rate generates more revenue, the system may also deteriorate faster. Production should thus be controlled dynamically to trade-off deterioration and revenue accumulation in between maintenance moments. We study systems for which the relation between production and deterioration is known and the same for each system as well as systems for which this relation differs from system to system and needs to be learned on-the-fly. The decision problem is to find the optimal production policy given planned maintenance moments (operational) and the optimal interval length between such maintenance moments (tactical). For systems with a known production-deterioration relation, we cast the operational decision problem as a continuous-time Markov decision process and prove that the optimal policy has intuitive monotonic properties. We also present sufficient conditions for the optimality of bang-bang policies and we partially characterize the structure of the optimal interval length, thereby enabling efficient joint optimization of the operational and tactical decision problem. For systems that exhibit variability in their production-deterioration relations, we propose a Bayesian procedure to learn the unknown deterioration rate under any production policy. Our extensive numerical study indicates significant profit increases of our approaches compared to the state-of-the-art.
△ Less
Submitted 6 December, 2023; v1 submitted 14 August, 2023;
originally announced August 2023.
-
Large Language Models to Identify Social Determinants of Health in Electronic Health Records
Authors:
Marco Guevara,
Shan Chen,
Spencer Thomas,
Tafadzwa L. Chaunzwa,
Idalid Franco,
Benjamin Kann,
Shalini Moningi,
Jack Qian,
Madeleine Goldstein,
Susan Harper,
Hugo JWL Aerts,
Guergana K. Savova,
Raymond H. Mak,
Danielle S. Bitterman
Abstract:
Social determinants of health (SDoH) have an important impact on patient outcomes but are incompletely collected from the electronic health records (EHR). This study researched the ability of large language models to extract SDoH from free text in EHRs, where they are most commonly documented, and explored the role of synthetic clinical text for improving the extraction of these scarcely documente…
▽ More
Social determinants of health (SDoH) have an important impact on patient outcomes but are incompletely collected from the electronic health records (EHR). This study researched the ability of large language models to extract SDoH from free text in EHRs, where they are most commonly documented, and explored the role of synthetic clinical text for improving the extraction of these scarcely documented, yet extremely valuable, clinical data. 800 patient notes were annotated for SDoH categories, and several transformer-based models were evaluated. The study also experimented with synthetic data generation and assessed for algorithmic bias. Our best-performing models were fine-tuned Flan-T5 XL (macro-F1 0.71) for any SDoH, and Flan-T5 XXL (macro-F1 0.70). The benefit of augmenting fine-tuning with synthetic data varied across model architecture and size, with smaller Flan-T5 models (base and large) showing the greatest improvements in performance (delta F1 +0.12 to +0.23). Model performance was similar on the in-hospital system dataset but worse on the MIMIC-III dataset. Our best-performing fine-tuned models outperformed zero- and few-shot performance of ChatGPT-family models for both tasks. These fine-tuned models were less likely than ChatGPT to change their prediction when race/ethnicity and gender descriptors were added to the text, suggesting less algorithmic bias (p<0.05). At the patient-level, our models identified 93.8% of patients with adverse SDoH, while ICD-10 codes captured 2.0%. Our method can effectively extracted SDoH information from clinic notes, performing better compare to GPT zero- and few-shot settings. These models could enhance real-world evidence on SDoH and aid in identifying patients needing social support.
△ Less
Submitted 5 March, 2024; v1 submitted 11 August, 2023;
originally announced August 2023.
-
Optimal Design of Lines Replaceable Units
Authors:
Joni Driessen,
Joost de Kruijf,
Joachim Arts,
Geert-Jan van Houtum
Abstract:
A Line Replaceable Unit (LRU) is a collection of connected parts in a system that is replaced when any part of the LRU fails. Companies use LRUs as a mechanism to reduce downtime of systems following a failure. The design of LRUs determines how fast a replacement is performed, so a smart design reduces replacement and downtime cost. A firm must purchase/repair a LRU upon failure, and large LRUs ar…
▽ More
A Line Replaceable Unit (LRU) is a collection of connected parts in a system that is replaced when any part of the LRU fails. Companies use LRUs as a mechanism to reduce downtime of systems following a failure. The design of LRUs determines how fast a replacement is performed, so a smart design reduces replacement and downtime cost. A firm must purchase/repair a LRU upon failure, and large LRUs are more expensive to purchase/repair. Hence, a firm seeks to design LRUs such that the average costs per time unit are minimized. We formalize this problem in a new model that captures how parts in a system are connected, and how they are disassembled from the system. Our model optimizes the design of LRUs such that the replacement (and downtime) costs and LRU purchase/repair costs are minimized. We present a set partitioning formulation for which we prove a rare result: the optimal solution is integer, despite a non--integral feasible polyhedron. Secondly, we formulate our problem as a binary linear program. The paper concludes by numerically comparing the computation times of both formulations and illustrates the effects of various parameters on the model's outcome.
△ Less
Submitted 7 August, 2023;
originally announced August 2023.
-
Load Asymptotics and Dynamic Speed Optimization for the Greenest Path Problem: A Comprehensive Analysis
Authors:
Poulad Moradi,
Joachim Arts,
Josué Velázquez-Martínez
Abstract:
We study the effect of using high-resolution elevation data on the selection of the most fuel-efficient (greenest) path for different trucks in various urban environments. We adapt a variant of the Comprehensive Modal Emission Model (CMEM) to show that the optimal speed and the greenest path are slope dependent (dynamic). When there are no elevation changes in a road network, the most fuel-efficie…
▽ More
We study the effect of using high-resolution elevation data on the selection of the most fuel-efficient (greenest) path for different trucks in various urban environments. We adapt a variant of the Comprehensive Modal Emission Model (CMEM) to show that the optimal speed and the greenest path are slope dependent (dynamic). When there are no elevation changes in a road network, the most fuel-efficient path is the shortest path with a constant (static) optimal speed throughout. However, if the network is not flat, then the shortest path is not necessarily the greenest path, and the optimal driving speed is dynamic. We prove that the greenest path converges to an asymptotic greenest path as the payload approaches infinity and that this limiting path is attained for a finite load. In a set of extensive numerical experiments, we benchmark the CO2 emissions reduction of our dynamic speed and the greenest path policies against policies that ignore elevation data. We use the geo-spatial data of 25 major cities across 6 continents, such as Los Angeles, Mexico City, Johannesburg, Athens, Ankara, and Canberra. Our results show that, on average, traversing the greenest path with a dynamic optimal speed policy can reduce the CO2 emissions by 1.19% to 10.15% depending on the city and truck type for a moderate payload. They also demonstrate that the average CO2 reduction of the optimal dynamic speed policy is between 2% to 4% for most of the cities, regardless of the truck type. We confirm that disregarding elevation data yields sub-optimal paths that are significantly less CO2 efficient than the greenest paths.
△ Less
Submitted 2 June, 2023;
originally announced June 2023.
-
Enrichment of the NLST and NSCLC-Radiomics computed tomography collections with AI-derived annotations
Authors:
Deepa Krishnaswamy,
Dennis Bontempi,
Vamsi Thiriveedhi,
Davide Punzo,
David Clunie,
Christopher P Bridge,
Hugo JWL Aerts,
Ron Kikinis,
Andrey Fedorov
Abstract:
Public imaging datasets are critical for the development and evaluation of automated tools in cancer imaging. Unfortunately, many do not include annotations or image-derived features, complicating their downstream analysis. Artificial intelligence-based annotation tools have been shown to achieve acceptable performance and thus can be used to automatically annotate large datasets. As part of the e…
▽ More
Public imaging datasets are critical for the development and evaluation of automated tools in cancer imaging. Unfortunately, many do not include annotations or image-derived features, complicating their downstream analysis. Artificial intelligence-based annotation tools have been shown to achieve acceptable performance and thus can be used to automatically annotate large datasets. As part of the effort to enrich public data available within NCI Imaging Data Commons (IDC), here we introduce AI-generated annotations for two collections of computed tomography images of the chest, NSCLC-Radiomics, and the National Lung Screening Trial. Using publicly available AI algorithms we derived volumetric annotations of thoracic organs at risk, their corresponding radiomics features, and slice-level annotations of anatomical landmarks and regions. The resulting annotations are publicly available within IDC, where the DICOM format is used to harmonize the data and achieve FAIR principles. The annotations are accompanied by cloud-enabled notebooks demonstrating their use. This study reinforces the need for large, publicly accessible curated datasets and demonstrates how AI can be used to aid in cancer imaging.
△ Less
Submitted 31 May, 2023;
originally announced June 2023.
-
Evaluation of ChatGPT Family of Models for Biomedical Reasoning and Classification
Authors:
Shan Chen,
Yingya Li,
Sheng Lu,
Hoang Van,
Hugo JWL Aerts,
Guergana K. Savova,
Danielle S. Bitterman
Abstract:
Recent advances in large language models (LLMs) have shown impressive ability in biomedical question-answering, but have not been adequately investigated for more specific biomedical applications. This study investigates the performance of LLMs such as the ChatGPT family of models (GPT-3.5s, GPT-4) in biomedical tasks beyond question-answering. Because no patient data can be passed to the OpenAI A…
▽ More
Recent advances in large language models (LLMs) have shown impressive ability in biomedical question-answering, but have not been adequately investigated for more specific biomedical applications. This study investigates the performance of LLMs such as the ChatGPT family of models (GPT-3.5s, GPT-4) in biomedical tasks beyond question-answering. Because no patient data can be passed to the OpenAI API public interface, we evaluated model performance with over 10000 samples as proxies for two fundamental tasks in the clinical domain - classification and reasoning. The first task is classifying whether statements of clinical and policy recommendations in scientific literature constitute health advice. The second task is causal relation detection from the biomedical literature. We compared LLMs with simpler models, such as bag-of-words (BoW) with logistic regression, and fine-tuned BioBERT models. Despite the excitement around viral ChatGPT, we found that fine-tuning for two fundamental NLP tasks remained the best strategy. The simple BoW model performed on par with the most complex LLM prompting. Prompt engineering required significant investment.
△ Less
Submitted 5 April, 2023;
originally announced April 2023.
-
Triplet supercurrents in lateral Josephson junctions with a half-metallic ferromagnet
Authors:
Yao Junxiang,
Remko Fermin,
Mariona Cabero,
Kaveh Lahabi,
Jan Aarts
Abstract:
In the area of superconducting spintronics, spin triplet supercurrents in half-metallic ferromagnets (HMFs) could yield dissipationless spin transport over large distances, and high current density. Promising among the HMFs is the perovskite oxide La0.7Sr0.3MnO3 (LSMO), and recent studies in combination with the high-Tc superconductor YBa2Cu3O7, or the conventional superconductor NbTi, showed long…
▽ More
In the area of superconducting spintronics, spin triplet supercurrents in half-metallic ferromagnets (HMFs) could yield dissipationless spin transport over large distances, and high current density. Promising among the HMFs is the perovskite oxide La0.7Sr0.3MnO3 (LSMO), and recent studies in combination with the high-Tc superconductor YBa2Cu3O7, or the conventional superconductor NbTi, showed long range effects. Here we focus on two issues that as yet received less attention: the value of the critical current in the HMF in the limit of a very small electrode distance (20 nm), and the nature of the spin triplet generator. We use lateral junctions shaped as bar, square, and disk, and find high supercurrent densities, of order 10^11 A/m2, pointing to an efficient triplet generation mechanism. This is surprising in the sense that no magnetic inhomogeneity is purposely built in, as is done in conventional metal triplet junctions. Furthermore, from the magnetic field dependence of the critical current interference patterns we find a uniform supercurrent distribution in bar-shaped devices, but one more constricted to the rim in disk devices, which is an expected consequence of the geometry. We also analyze the temperature dependence of the critical current and find the quadratic dependence that was predicted in the limit of small junction lengths. From studying the NbTi/LSMO interface with scanning electron transmission microscopy, we conclude that the magnetic inhomogeneity required for triplet generation resides in the LSMO layer adjacent to the interface.
△ Less
Submitted 9 October, 2024; v1 submitted 24 March, 2023;
originally announced March 2023.
-
Natural language processing to automatically extract the presence and severity of esophagitis in notes of patients undergoing radiotherapy
Authors:
Shan Chen,
Marco Guevara,
Nicolas Ramirez,
Arpi Murray,
Jeremy L. Warner,
Hugo JWL Aerts,
Timothy A. Miller,
Guergana K. Savova,
Raymond H. Mak,
Danielle S. Bitterman
Abstract:
Radiotherapy (RT) toxicities can impair survival and quality-of-life, yet remain under-studied. Real-world evidence holds potential to improve our understanding of toxicities, but toxicity information is often only in clinical notes. We developed natural language processing (NLP) models to identify the presence and severity of esophagitis from notes of patients treated with thoracic RT. We fine-tu…
▽ More
Radiotherapy (RT) toxicities can impair survival and quality-of-life, yet remain under-studied. Real-world evidence holds potential to improve our understanding of toxicities, but toxicity information is often only in clinical notes. We developed natural language processing (NLP) models to identify the presence and severity of esophagitis from notes of patients treated with thoracic RT. We fine-tuned statistical and pre-trained BERT-based models for three esophagitis classification tasks: Task 1) presence of esophagitis, Task 2) severe esophagitis or not, and Task 3) no esophagitis vs. grade 1 vs. grade 2-3. Transferability was tested on 345 notes from patients with esophageal cancer undergoing RT.
Fine-tuning PubmedBERT yielded the best performance. The best macro-F1 was 0.92, 0.82, and 0.74 for Task 1, 2, and 3, respectively. Selecting the most informative note sections during fine-tuning improved macro-F1 by over 2% for all tasks. Silver-labeled data improved the macro-F1 by over 3% across all tasks. For the esophageal cancer notes, the best macro-F1 was 0.73, 0.74, and 0.65 for Task 1, 2, and 3, respectively, without additional fine-tuning.
To our knowledge, this is the first effort to automatically extract esophagitis toxicity severity according to CTCAE guidelines from clinic notes. The promising performance provides proof-of-concept for NLP-based automated detailed toxicity monitoring in expanded domains.
△ Less
Submitted 23 March, 2023;
originally announced March 2023.
-
MEDBERT.de: A Comprehensive German BERT Model for the Medical Domain
Authors:
Keno K. Bressem,
Jens-Michalis Papaioannou,
Paul Grundmann,
Florian Borchert,
Lisa C. Adams,
Leonhard Liu,
Felix Busch,
Lina Xu,
Jan P. Loyen,
Stefan M. Niehues,
Moritz Augustin,
Lennart Grosser,
Marcus R. Makowski,
Hugo JWL. Aerts,
Alexander Löser
Abstract:
This paper presents medBERTde, a pre-trained German BERT model specifically designed for the German medical domain. The model has been trained on a large corpus of 4.7 Million German medical documents and has been shown to achieve new state-of-the-art performance on eight different medical benchmarks covering a wide range of disciplines and medical document types. In addition to evaluating the ove…
▽ More
This paper presents medBERTde, a pre-trained German BERT model specifically designed for the German medical domain. The model has been trained on a large corpus of 4.7 Million German medical documents and has been shown to achieve new state-of-the-art performance on eight different medical benchmarks covering a wide range of disciplines and medical document types. In addition to evaluating the overall performance of the model, this paper also conducts a more in-depth analysis of its capabilities. We investigate the impact of data deduplication on the model's performance, as well as the potential benefits of using more efficient tokenization methods. Our results indicate that domain-specific models such as medBERTde are particularly useful for longer texts, and that deduplication of training data does not necessarily lead to improved performance. Furthermore, we found that efficient tokenization plays only a minor role in improving model performance, and attribute most of the improved performance to the large amount of training data. To encourage further research, the pre-trained model weights and new benchmarks based on radiological data are made publicly available for use by the scientific community.
△ Less
Submitted 24 March, 2023; v1 submitted 14 March, 2023;
originally announced March 2023.
-
The unusual distribution of spin-triplet supercurrents in disk-shaped Josephson junctions
Authors:
Remko Fermin,
Junxiang Yao,
Kaveh Lahabi,
Jan Aarts
Abstract:
The phenomenon of s-wave spin triplet Cooper pairs induced in ferromagnetic metals has been researched now for more than a decade, and its main aspects are well understood. Crucial in converting s-wave singlet pairs in the superconductor to s-wave triplets in the ferromagnet is the engineering of well-defined magnetic inhomogeneity (the 'generator') at the interface with the superconductor. Vertic…
▽ More
The phenomenon of s-wave spin triplet Cooper pairs induced in ferromagnetic metals has been researched now for more than a decade, and its main aspects are well understood. Crucial in converting s-wave singlet pairs in the superconductor to s-wave triplets in the ferromagnet is the engineering of well-defined magnetic inhomogeneity (the 'generator') at the interface with the superconductor. Vertical layer stacks are typically used as such, where two separate thin ferromagnetic layers with homogeneous but non-collinear magnetizations, provide the inhomogeneity. Alternatively, magnetic textures, like ferromagnetic domain walls and vortices, are possible triplet generators, although they are far less studied. In this paper we review our experiments on lateral disk-shaped Josephson junctions where a ferromagnetic bottom layer provides a weak link with a vortex magnetization imposed by the shape of the disk. We present three different junction configurations, exhibiting their own generator mechanism. In the first, we utilize the non-collinearity with a second ferromagnetic layer to produce the triplet correlations. The second configuration consists of only the bottom ferromagnet and the superconducting contacts; it relies on the vortex magnetization itself to generate the spin-polarized supercurrents. In the third case we exploit an intrinsic generator by combining a conventional superconductor (NbTi) and a half-metallic ferromagnetic oxide (La$_{0.7}$Sr$_{0.3}$MnO$_3$). We find strong supercurrents in all cases. A particularly interesting finding is that the supercurrents are strongly confined at the rims of the device, independent of the generating mechanism, but directly related to their triplet nature. What causes these rim currents remains an open question.
△ Less
Submitted 14 March, 2023;
originally announced March 2023.
-
Beyond the effective length: How to analyze magnetic interference patterns of thin-film planar Josephson junctions with finite lateral dimensions
Authors:
Remko Fermin,
Bob de Wit and,
Jan Aarts
Abstract:
The magnetic field dependent critical current $I_{\text{c}}(B)$ of a Josephson junction is determined by the screening currents in its electrodes. In macroscopic junctions, a local vector potential drives the currents, however, in thin film planar junctions, with electrodes of finite size and various shapes, they are governed by non-local electrodynamics. This complicates the extraction of paramet…
▽ More
The magnetic field dependent critical current $I_{\text{c}}(B)$ of a Josephson junction is determined by the screening currents in its electrodes. In macroscopic junctions, a local vector potential drives the currents, however, in thin film planar junctions, with electrodes of finite size and various shapes, they are governed by non-local electrodynamics. This complicates the extraction of parameters such as the geometry of the effective junction area, the effective junction length and, the critical current density distribution from the $I_{\text{c}}(B)$ interference patterns. Here we provide a method to tackle this problem by simulating the phase differences that drive the shielding currents and use those to find $I_{\text{c}}(B)$. To this end, we extend the technique proposed by John Clem [Phys. Rev. B, \textbf{81}, 144515 (2010)] to find $I_{\text{c}}(B)$ for Josephson junctions separating a superconducting strip of length $L$ and width $W$ with rectangular, ellipsoid and rhomboid geometries. We find the periodicity of the interference pattern ($ΔB$) to have geometry independent limits for $L \gg W$ and $L \ll W$. By fabricating elliptically shaped S$-$N$-$S junctions with various aspect ratios, we experimentally verify the $L/W$ dependence of $ΔB$. Finally, we incorporate these results to correctly extract the distribution of critical currents in the junction by the Fourier analysis of $I_{\text{c}}(B)$, which makes these results essential for the correct analysis of topological channels in thin film planar Josephson junctions.
△ Less
Submitted 10 February, 2023; v1 submitted 11 October, 2022;
originally announced October 2022.
-
What Does DALL-E 2 Know About Radiology?
Authors:
Lisa C. Adams,
Felix Busch,
Daniel Truhn,
Marcus R. Makowski,
Hugo JWL. Aerts,
Keno K. Bressem
Abstract:
Generative models such as DALL-E 2 could represent a promising future tool for image generation, augmentation, and manipulation for artificial intelligence research in radiology provided that these models have sufficient medical domain knowledge. Here we show that DALL-E 2 has learned relevant representations of X-ray images with promising capabilities in terms of zero-shot text-to-image generatio…
▽ More
Generative models such as DALL-E 2 could represent a promising future tool for image generation, augmentation, and manipulation for artificial intelligence research in radiology provided that these models have sufficient medical domain knowledge. Here we show that DALL-E 2 has learned relevant representations of X-ray images with promising capabilities in terms of zero-shot text-to-image generation of new images, continuation of an image beyond its original boundaries, or removal of elements, while pathology generation or CT, MRI, and ultrasound images are still limited. The use of generative models for augmenting and generating radiological data thus seems feasible, even if further fine-tuning and adaptation of these models to the respective domain is required beforehand.
△ Less
Submitted 27 September, 2022;
originally announced September 2022.
-
Effective Dual-Sourcing Through Inventory Projection
Authors:
Melvin Drent,
Joachim Arts
Abstract:
We consider a single-echelon inventory system under periodic review with two suppliers facing stochastic demand, where excess demand is backlogged. The expedited supplier has a shorter lead time than the regular supplier but charges a higher unit price. We introduce the Projected Expedited Inventory Position (PEIP) policy, and we show that the relative difference between the long run average cost…
▽ More
We consider a single-echelon inventory system under periodic review with two suppliers facing stochastic demand, where excess demand is backlogged. The expedited supplier has a shorter lead time than the regular supplier but charges a higher unit price. We introduce the Projected Expedited Inventory Position (PEIP) policy, and we show that the relative difference between the long run average cost per period of this policy and the optimal policy converges to zero when both the shortage cost and the cost premium for expedited units become large, with their ratio held constant. A corollary of this result is that several existing heuristics are also asymptotically optimal in this non-trivial regime. We show through an extensive numerical investigation that the PEIP policy outperforms the current best performing heuristic policies in literature.
△ Less
Submitted 22 July, 2022;
originally announced July 2022.
-
Efficient Emission Reduction Through Dynamic Supply Mode Selection
Authors:
Melvin Drent,
Poulad Moradi,
Joachim Arts
Abstract:
We study the inbound supply mode and inventory management decision making for a company that sells an assortment of products. Stochastic demand for each product arrives periodically and unmet demand is backlogged. Each product has two distinct supply modes that may be different suppliers or different transport modes from the same supplier. These supply modes differ in terms of their carbon emissio…
▽ More
We study the inbound supply mode and inventory management decision making for a company that sells an assortment of products. Stochastic demand for each product arrives periodically and unmet demand is backlogged. Each product has two distinct supply modes that may be different suppliers or different transport modes from the same supplier. These supply modes differ in terms of their carbon emissions, speed, and costs. The company needs to decide when to ship how much using which supply mode such that total holding, backlog, and procurement costs are minimized while the emissions associated with different supply modes across the assortment remains below a certain target level. Since the optimal policy for this inventory system is highly complex, we assume that shipment decisions for each product are governed by a dual-index policy. This policy dynamically prescribes shipment quantities with both supply modes based on the on-hand inventory, the backlog, and the products that are still in-transit. We formulate this decision problem as a mixed integer linear program that we solve through Dantzig-wolfe decomposition. We benchmark our decision model against two state-of-the-art approaches in a large test-bed based on real-life carbon emissions data. Relative to our decision model, the first benchmark lacks the flexibility to dynamically ship products with two supply modes while the second benchmark makes supply mode decisions for each product individually rather than holistically for the entire assortment. Our computational experiment shows that our decision model can outperform the first and second benchmark by up to 15 and 40 percent, respectively, for realistic targets for carbon emission reduction.
△ Less
Submitted 24 March, 2023; v1 submitted 22 July, 2022;
originally announced July 2022.
-
Mesoscopic superconducting memory based on bistable magnetic textures
Authors:
Remko Fermin,
Naor Scheinowitz,
Jan Aarts,
Kaveh Lahabi
Abstract:
With the ever-increasing energy need to process big data, the realization of low-power computing technologies, such as superconducting logic and memories, has become a pressing issue. Developing fast and non-volatile superconducting memory elements, however, remains a challenge. Superconductor-ferromagnet hybrid devices offer a promising solution, as they combine ultra-fast manipulation of spins w…
▽ More
With the ever-increasing energy need to process big data, the realization of low-power computing technologies, such as superconducting logic and memories, has become a pressing issue. Developing fast and non-volatile superconducting memory elements, however, remains a challenge. Superconductor-ferromagnet hybrid devices offer a promising solution, as they combine ultra-fast manipulation of spins with dissipationless readout. Here, we present a new type of non-volatile Josephson junction memory that utilizes the bistable magnetic texture of a single mesoscopic ferromagnet. We use micromagnetic simulations to design an ellipse-shaped planar junction structured from a Nb/Co bilayer. The ellipse can be prepared as uniformly magnetized or as a pair of vortices at zero applied field. The two states yield considerably different critical currents, enabling reliable electrical readout of the element. We describe the mechanism used to control the critical current by applying numerical calculations to quantify the local stray field from the ferromagnet, which shifts the superconducting interference pattern. By combining micromagnetic modeling with bistable spin-textured junctions, our approach presents a novel route towards realizing superconducting memory applications.
△ Less
Submitted 14 June, 2022;
originally announced June 2022.
-
Deep Learning-based Assessment of Hepatic Steatosis on chest CT
Authors:
Zhongyi Zhang,
Jakob Weiss,
Jana Taron,
Roman Zeleznik,
Michael T. Lu,
Hugo J. W. L. Aerts
Abstract:
Purpose: Automatic methods are required for the early detection of hepatic steatosis to avoid progression to cirrhosis and cancer. Here, we developed a fully automated deep learning pipeline to quantify hepatic steatosis on non-contrast enhanced chest computed tomography (CT) scans. Materials and Methods: We developed and evaluated our pipeline on chest CT images of 1,431 randomly selected Nationa…
▽ More
Purpose: Automatic methods are required for the early detection of hepatic steatosis to avoid progression to cirrhosis and cancer. Here, we developed a fully automated deep learning pipeline to quantify hepatic steatosis on non-contrast enhanced chest computed tomography (CT) scans. Materials and Methods: We developed and evaluated our pipeline on chest CT images of 1,431 randomly selected National Lung Screening Trial (NLST) participants. A dataset of 451 CT scans with volumetric liver segmentations of expert readers was used for training a deep learning model. For testing, in an independent dataset of 980 CT scans hepatic attenuation was manually measured by an expert reader on three cross-sectional images at different hepatic levels by selecting three circular regions of interest. Additionally, 100 randomly selected cases of the test set were volumetrically segmented by expert readers. Hepatic steatosis on the test set was defined as mean hepatic attenuation of < 40 Hounsfield unit. Spearman correlation was conducted to analyze liver fat quantification accuracy and the Cohen's Kappa coefficient was calculated for hepatic steatosis prediction reliability. Results: Our pipeline demonstrated strong performance and achieved a mean dice score of 0.970 for the volumetric liver segmentation. The spearman correlation of the liver fat quantification was 0.954 (P <0.0001) between the automated and expert reader measurements. The cohen's kappa coefficient was 0.875 for automatic assessment of hepatic steatosis. Conclusion: We developed a fully automatic deep learning-based pipeline for the assessment of hepatic steatosis in chest CT images. With the fast and cheap screening of hepatic steatosis, our pipeline has the potential to help initiate preventive measures to avoid progression to cirrhosis and cancer.
△ Less
Submitted 4 February, 2022;
originally announced February 2022.
-
Superconducting triplet rim currents in a spin-textured ferromagnetic disk
Authors:
Remko Fermin,
Dyon van Dinter,
Michel Hubert,
Bart Woltjes,
Mikhail Silaev,
Jan Aarts,
Kaveh Lahabi
Abstract:
Since the discovery of the long-range superconducting proximity effect, the interaction between spin-triplet Cooper pairs and magnetic structures such as domain walls and vortices has been the subject of intense theoretical discussions, while the relevant experiments remain scarce. We have developed nanostructured Josephson junctions with highly controllable spin texture, based on a disk-shaped Nb…
▽ More
Since the discovery of the long-range superconducting proximity effect, the interaction between spin-triplet Cooper pairs and magnetic structures such as domain walls and vortices has been the subject of intense theoretical discussions, while the relevant experiments remain scarce. We have developed nanostructured Josephson junctions with highly controllable spin texture, based on a disk-shaped Nb/Co bilayer. Here, the vortex magnetization of Co and the Cooper pairs of Nb conspire to induce long-range triplet (LRT) superconductivity in the ferromagnet. Surprisingly, the LRT correlations emerge in highly localized (sub-80 nm) channels at the rim of the ferromagnet, despite its trivial band structure. We show that these robust rim currents arise from the magnetization texture acting as an effective spin-orbit coupling, which results in spin accumulation at the bilayer-vacuum boundary. Lastly, we demonstrate that by altering the spin texture of a single ferromagnet, both $0$ and $π$-channels can be realized in the same device.
△ Less
Submitted 2 May, 2022; v1 submitted 25 October, 2021;
originally announced October 2021.
-
Deep learning-based detection of intravenous contrast in computed tomography scans
Authors:
Zezhong Ye,
Jack M. Qian,
Ahmed Hosny,
Roman Zeleznik,
Deborah Plana,
Jirapat Likitlersuang,
Zhongyi Zhang,
Raymond H. Mak,
Hugo J. W. L. Aerts,
Benjamin H. Kann
Abstract:
Purpose: Identifying intravenous (IV) contrast use within CT scans is a key component of data curation for model development and testing. Currently, IV contrast is poorly documented in imaging metadata and necessitates manual correction and annotation by clinician experts, presenting a major barrier to imaging analyses and algorithm deployment. We sought to develop and validate a convolutional neu…
▽ More
Purpose: Identifying intravenous (IV) contrast use within CT scans is a key component of data curation for model development and testing. Currently, IV contrast is poorly documented in imaging metadata and necessitates manual correction and annotation by clinician experts, presenting a major barrier to imaging analyses and algorithm deployment. We sought to develop and validate a convolutional neural network (CNN)-based deep learning (DL) platform to identify IV contrast within CT scans. Methods: For model development and evaluation, we used independent datasets of CT scans of head, neck (HN) and lung cancer patients, totaling 133,480 axial 2D scan slices from 1,979 CT scans manually annotated for contrast presence by clinical experts. Five different DL models were adopted and trained in HN training datasets for slice-level contrast detection. Model performances were evaluated on a hold-out set and on an independent validation set from another institution. DL models was then fine-tuned on chest CT data and externally validated on a separate chest CT dataset. Results: Initial DICOM metadata tags for IV contrast were missing or erroneous in 1,496 scans (75.6%). The EfficientNetB4-based model showed the best overall detection performance. For HN scans, AUC was 0.996 in the internal validation set (n = 216) and 1.0 in the external validation set (n = 595). The fine-tuned model on chest CTs yielded an AUC: 1.0 for the internal validation set (n = 53), and AUC: 0.980 for the external validation set (n = 402). Conclusion: The DL model could accurately detect IV contrast in both HN and chest CT scans with near-perfect performance.
△ Less
Submitted 19 October, 2021; v1 submitted 15 October, 2021;
originally announced October 2021.
-
Apertif, Phased Array Feeds for the Westerbork Synthesis Radio Telescope
Authors:
W. A. van Cappellen,
T. A. Oosterloo,
M. A. W. Verheijen,
E. A. K. Adams,
B. Adebahr,
R. Braun,
K. M. Hess,
H. Holties,
J. M. van der Hulst,
B. Hut,
E. Kooistra,
J. van Leeuwen,
G. M. Loose,
R. Morganti,
V. A. Moss,
E. Orrú,
M. Ruiter,
A. P. Schoenmakers,
N. J. Vermaas,
S. J. Wijnholds,
A. S. van Amesfoort,
M. J. Arts,
J. J. Attema,
L. Bakker,
C. G. Bassa
, et al. (65 additional authors not shown)
Abstract:
We describe the APERture Tile In Focus (Apertif) system, a phased array feed (PAF) upgrade of the Westerbork Synthesis Radio Telescope which has transformed this telescope into a high-sensitivity, wide field-of-view L-band imaging and transient survey instrument. Using novel PAF technology, up to 40 partially overlapping beams can be formed on the sky simultaneously, significantly increasing the s…
▽ More
We describe the APERture Tile In Focus (Apertif) system, a phased array feed (PAF) upgrade of the Westerbork Synthesis Radio Telescope which has transformed this telescope into a high-sensitivity, wide field-of-view L-band imaging and transient survey instrument. Using novel PAF technology, up to 40 partially overlapping beams can be formed on the sky simultaneously, significantly increasing the survey speed of the telescope. With this upgraded instrument, an imaging survey covering an area of 2300 deg2 is being performed which will deliver both continuum and spectral line data sets, of which the first data has been publicly released. In addition, a time domain transient and pulsar survey covering 15,000 deg2 is in progress. An overview of the Apertif science drivers, hardware and software of the upgraded telescope is presented, along with its key performance characteristics.
△ Less
Submitted 30 September, 2021; v1 submitted 29 September, 2021;
originally announced September 2021.
-
On the resistance minimum in LaAlO$_3$/Eu$_{1-x}$La$_x$TiO$_3$/SrTiO$_3$ heterostructures
Authors:
N. Lebedev,
Y. Huang,
A. Rana,
D. Jannis,
N. Gauquelin,
J. Verbeeck,
J. Aarts
Abstract:
In this paper we study LaAlO$_3$/Eu$_{1-x}$La$_x$TiO$_3$/SrTiO$_3$ structures with nominally x = 0, 0.1 and different thicknesses of the Eu$_{1-x}$La$_x$TiO$_3$ layer. We observe that both systems have many properties similar to previously studied LaAlO$_3$/EuTiO$_3$/SrTiO$_3$ and other oxide interfaces, such as the formation of a 2D electron liquid for 1 or 2 unit cells of Eu$_{1-x}$La$_x$TiO…
▽ More
In this paper we study LaAlO$_3$/Eu$_{1-x}$La$_x$TiO$_3$/SrTiO$_3$ structures with nominally x = 0, 0.1 and different thicknesses of the Eu$_{1-x}$La$_x$TiO$_3$ layer. We observe that both systems have many properties similar to previously studied LaAlO$_3$/EuTiO$_3$/SrTiO$_3$ and other oxide interfaces, such as the formation of a 2D electron liquid for 1 or 2 unit cells of Eu$_{1-x}$La$_x$TiO$_3$; a metal-insulator transition driven by the thickness increase of Eu$_{1-x}$La$_x$TiO$_3$ layer; the presence of an Anomalous Hall effect (AHE) when driving the systems above the Lifshitz point with a backgate voltage; and a minimum in the temperature dependence of the sheet resistance below the Lifshitz point in the one-band regime, which becomes more pronounced with increasing gate voltage. However, and notwithstanding the likely presence of magnetism in the system, we do not attribute that minimum to the Kondo effect, but rather to the properties of SrTiO$_3$ crystal and the inevitable effects of charge trapping when using back gates.
△ Less
Submitted 1 September, 2021;
originally announced September 2021.
-
Imaging spin-wave damping underneath metals using electron spins in diamond
Authors:
Iacopo Bertelli,
Brecht G. Simon,
Tao Yu,
Jan Aarts,
Gerrit E. W. Bauer,
Yaroslav M. Blanter,
Toeno van der Sar
Abstract:
Spin waves in magnetic insulators are low-damping signal carriers that could enable a new generation of spintronic devices. The excitation, control, and detection of spin waves by metal electrodes is crucial for interfacing these devices to electrical circuits. It is therefore important to understand metal-induced damping of spin-wave transport, but characterizing this process requires access to t…
▽ More
Spin waves in magnetic insulators are low-damping signal carriers that could enable a new generation of spintronic devices. The excitation, control, and detection of spin waves by metal electrodes is crucial for interfacing these devices to electrical circuits. It is therefore important to understand metal-induced damping of spin-wave transport, but characterizing this process requires access to the underlying magnetic films. Here we show that spins in diamond enable imaging of spin waves that propagate underneath metals in magnetic insulators, and then use this capability to reveal a 100-fold increase in spin-wave damping. By analyzing spin-wave-induced currents in the metal, we derive an effective damping parameter that matches these observations well. We furthermore detect buried scattering centers, highlighting the technique's power for assessing spintronic device quality. Our results open new avenues for studying metal - spin-wave interaction and provide access to interfacial processes such as spin-wave injection via the spin-Hall effect.
△ Less
Submitted 16 June, 2021; v1 submitted 4 June, 2021;
originally announced June 2021.
-
Universal size-dependent nonlinear charge transport in single crystals of the Mott insulator Ca$_2$RuO$_4$
Authors:
Guerino Avallone,
Remko Fermin,
Kaveh Lahabi,
Veronica Granata,
Rosalba Fittipaldi,
Carla Cirillo,
Carmine Attanasio,
Antonio Vecchione,
Jan Aarts
Abstract:
The surprisingly low current density required for inducing the insulator to metal transition has made Ca$_2$RuO$_4$ an attractive candidate material for developing Mott-based electronics devices. The mechanism driving the resistive switching, however, remains a controversial topic in the field of strongly correlated electron systems. Here we probe an uncovered region of phase space by studying hig…
▽ More
The surprisingly low current density required for inducing the insulator to metal transition has made Ca$_2$RuO$_4$ an attractive candidate material for developing Mott-based electronics devices. The mechanism driving the resistive switching, however, remains a controversial topic in the field of strongly correlated electron systems. Here we probe an uncovered region of phase space by studying high-purity Ca$_2$RuO$_4$ single crystals, using the sample size as principal tuning parameter. Upon reducing the crystal size, we find a four orders of magnitude increase in the current density required for driving Ca$_2$RuO$_4$ out of the insulating state into a non-equilibrium (also called metastable) phase which is the precursor to the fully metallic phase. By integrating a microscopic platinum thermometer and performing thermal simulations, we gain insight into the local temperature during simultaneous application of current and establish that the size dependence is not a result of Joule heating. The findings suggest an inhomogeneous current distribution in the nominally homogeneous crystal. Our study calls for a reexamination of the interplay between sample size, charge current, and temperature in driving Ca$_2$RuO$_4$ towards the Mott insulator to metal transition.
△ Less
Submitted 12 November, 2021; v1 submitted 12 February, 2021;
originally announced February 2021.
-
Projected Inventory Level Policies for Lost Sales Inventory Systems: Asymptotic Optimality in Two Regimes
Authors:
Willem van Jaarsveld,
Joachim Arts
Abstract:
We consider the canonical periodic review lost sales inventory system with positive lead-times and stochastic i.i.d. demand under the average cost criterion. We introduce a new policy that places orders such that the expected inventory level at the time of arrival of an order is at a fixed level and call it the Projected Inventory Level (PIL) policy. We prove that this policy has a cost-rate super…
▽ More
We consider the canonical periodic review lost sales inventory system with positive lead-times and stochastic i.i.d. demand under the average cost criterion. We introduce a new policy that places orders such that the expected inventory level at the time of arrival of an order is at a fixed level and call it the Projected Inventory Level (PIL) policy. We prove that this policy has a cost-rate superior to the equivalent system where excess demand is back-ordered instead of lost and is therefore asymptotically optimal as the cost of losing a sale approaches infinity under mild distributional assumptions. We further show that this policy dominates the constant order policy for any finite lead-time and is therefore asymptotically optimal as the lead-time approaches infinity for the case of exponentially distributed demand per period. Numerical results show this policy also performs superior relative to other policies.
△ Less
Submitted 16 January, 2024; v1 submitted 19 January, 2021;
originally announced January 2021.
-
Inhomogeneous superconductivity and quasilinear magnetoresistance at amorphous LaTiO3/SrTiO3 interfaces
Authors:
N. Lebedev,
M. Stehno,
A. Rana,
N. Gauquelin,
J. Verbeeck,
A. Brinkman,
J. Aarts
Abstract:
We have studied the transport properties of LaTiO3/SrTiO3 (LTO/STO) heterostructures. In spite of 2D growth observed in reflection high energy electron diffraction, Transmission Electron Microscopy images revealed that the samples tend to amorphize. Still, we observe that the structures are conducting, and some of them exhibit high conductance and/or superconductivity. We established that conducti…
▽ More
We have studied the transport properties of LaTiO3/SrTiO3 (LTO/STO) heterostructures. In spite of 2D growth observed in reflection high energy electron diffraction, Transmission Electron Microscopy images revealed that the samples tend to amorphize. Still, we observe that the structures are conducting, and some of them exhibit high conductance and/or superconductivity. We established that conductivity arises mainly on the STO side of the interface, and shows all the signs of the 2-dimensional electron gas usually observed at interfaces between SrTiO3 and LaTiO3 or LaAlO3, including the presence of two electron bands and tunability with a gate voltage. Analysis of magnetoresistance (MR) and superconductivity indicates presence of a spatial fluctuations of the electronic properties in our samples. That can explain the observed quasilinear out-of-plane MR, as well as various features of the in-plane MR and the observed superconductivity.
△ Less
Submitted 7 August, 2020;
originally announced August 2020.
-
Magnetic resonance imaging of spin-wave transport and interference in a magnetic insulator
Authors:
Iacopo Bertelli,
Joris J. Carmiggelt,
Tao Yu,
Brecht G. Simon,
Coosje C. Pothoven,
Gerrit E. W. Bauer,
Yaroslav M. Blanter,
Jan Aarts,
Toeno van der Sar
Abstract:
Spin waves - the elementary excitations of magnetic materials - are prime candidate signal carriers for low dissipation information processing. Being able to image coherent spin-wave transport is crucial for developing interference-based spin-wave devices. We introduce a platform for probing coherent spin waves based on magnetic resonance imaging with electron spins in diamond. Focusing on a thin-…
▽ More
Spin waves - the elementary excitations of magnetic materials - are prime candidate signal carriers for low dissipation information processing. Being able to image coherent spin-wave transport is crucial for developing interference-based spin-wave devices. We introduce a platform for probing coherent spin waves based on magnetic resonance imaging with electron spins in diamond. Focusing on a thin-film magnetic insulator, we quantify spin-wave amplitudes, visualize the dispersion, and demonstrate time-domain measurements of spin-wave packets. We use our platform to study spin-wave interference, revealing uni-directional, autofocused spin-wave patterns with frequency-controlled numerical apertures. A theoretical analysis explains the patterns in terms of chiral spin-wave excitation and stray-field coupling to the sensor spins. These results pave the way for probing spin waves in atomically thin magnets, even when embedded between opaque materials.
△ Less
Submitted 16 April, 2020;
originally announced April 2020.
-
The importance of transparency and reproducibility in artificial intelligence research
Authors:
Benjamin Haibe-Kains,
George Alexandru Adam,
Ahmed Hosny,
Farnoosh Khodakarami,
MAQC Society Board,
Levi Waldron,
Bo Wang,
Chris McIntosh,
Anshul Kundaje,
Casey S. Greene,
Michael M. Hoffman,
Jeffrey T. Leek,
Wolfgang Huber,
Alvis Brazma,
Joelle Pineau,
Robert Tibshirani,
Trevor Hastie,
John P. A. Ioannidis,
John Quackenbush,
Hugo J. W. L. Aerts
Abstract:
In their study, McKinney et al. showed the high potential of artificial intelligence for breast cancer screening. However, the lack of detailed methods and computer code undermines its scientific value. We identify obstacles hindering transparent and reproducible AI research as faced by McKinney et al and provide solutions with implications for the broader field.
In their study, McKinney et al. showed the high potential of artificial intelligence for breast cancer screening. However, the lack of detailed methods and computer code undermines its scientific value. We identify obstacles hindering transparent and reproducible AI research as faced by McKinney et al and provide solutions with implications for the broader field.
△ Less
Submitted 7 March, 2020; v1 submitted 28 February, 2020;
originally announced March 2020.
-
Gate-tuned Anomalous Hall Effect Driven by Rashba Splitting in Intermixed LaAlO3/GdTiO3/SrTiO3
Authors:
N. Lebedev,
M. Stehno,
A. Rana,
P. Reith,
N. Gauquelin,
J. Verbeeck,
H. Hilgenkamp,
A. Brinkman,
J. Aarts
Abstract:
The Anomalous Hall Effect (AHE) is an important quantity in determining the properties and understanding the behavior of the two-dimensional electron system forming at the interface of SrTiO3-based oxide heterostructures. The occurrence of AHE is often interpreted as a signature of ferromagnetism, but it is becoming more and more clear that also paramagnets may contribute to AHE. We studied the in…
▽ More
The Anomalous Hall Effect (AHE) is an important quantity in determining the properties and understanding the behavior of the two-dimensional electron system forming at the interface of SrTiO3-based oxide heterostructures. The occurrence of AHE is often interpreted as a signature of ferromagnetism, but it is becoming more and more clear that also paramagnets may contribute to AHE. We studied the influence of magnetic ions by measuring intermixed LaAlO3/GdTiO3/SrTiO3 at temperatures below 10 K. We find that, as function of gate voltage, the system undergoes a Lifshitz transition, while at the same time an onset of AHE is observed. However, we do not observe clear signs of ferromagnetism. We argue the AHE to be due to the change in Rashba spin-orbit coupling at the Lifshitz transition and conclude that also paramagnetic moments which are easily polarizable at low temperatures and high magnetic filds lead to the presence of AHE, which needs to be taken into account when extracting carrier densities and mobilities.
△ Less
Submitted 26 February, 2020;
originally announced February 2020.
-
Spontaneous emergence of Josephson junctions in homogeneous rings of single-crystal Sr2RuO4
Authors:
Yuuki Yasui,
Kaveh Lahabi,
Victor Fernández Becerra,
Remko Fermin,
Muhammad Shahbaz Anwar,
Shingo Yonezawa,
Takahito Terashima,
Milorad V. Milošević,
Jan Aarts,
Yoshiteru Maeno
Abstract:
The chiral p-wave order parameter in Sr$_\mathsf{2}$RuO$_\mathsf{4}$ would make it a special case amongst the unconventional superconductors. A consequence of this symmetry is the possible existence of superconducting domains of opposite chirality. At the boundary of such domains, the locally supressed condensate can produce an intrinsic Josephson junction. Here, we provide evidence of such juncti…
▽ More
The chiral p-wave order parameter in Sr$_\mathsf{2}$RuO$_\mathsf{4}$ would make it a special case amongst the unconventional superconductors. A consequence of this symmetry is the possible existence of superconducting domains of opposite chirality. At the boundary of such domains, the locally supressed condensate can produce an intrinsic Josephson junction. Here, we provide evidence of such junctions using mesoscopic rings, structured from Sr$_\mathsf{2}$RuO$_\mathsf{4}$ single crystals. Our order parameter simulations predict such rings to host stable domain walls across their arms. Thisis verified with transport experiments on loops, with a sharp transition at 1.5 K, which show distinct critical current oscillations with periodicity corresponding to the flux quantum. In contrast, loops with broadened transitions at around 3 K are void of such junctions and show standard Little-Parks oscillations. Our analysis demonstrates the junctions are of intrinsic origin and makes a compelling case for the existence of superconducting domains.
△ Less
Submitted 16 February, 2020;
originally announced February 2020.
-
ModelHub.AI: Dissemination Platform for Deep Learning Models
Authors:
Ahmed Hosny,
Michael Schwier,
Christoph Berger,
Evin P Örnek,
Mehmet Turan,
Phi V Tran,
Leon Weninger,
Fabian Isensee,
Klaus H Maier-Hein,
Richard McKinley,
Michael T Lu,
Udo Hoffmann,
Bjoern Menze,
Spyridon Bakas,
Andriy Fedorov,
Hugo JWL Aerts
Abstract:
Recent advances in artificial intelligence research have led to a profusion of studies that apply deep learning to problems in image analysis and natural language processing among others. Additionally, the availability of open-source computational frameworks has lowered the barriers to implementing state-of-the-art methods across multiple domains. Albeit leading to major performance breakthroughs…
▽ More
Recent advances in artificial intelligence research have led to a profusion of studies that apply deep learning to problems in image analysis and natural language processing among others. Additionally, the availability of open-source computational frameworks has lowered the barriers to implementing state-of-the-art methods across multiple domains. Albeit leading to major performance breakthroughs in some tasks, effective dissemination of deep learning algorithms remains challenging, inhibiting reproducibility and benchmarking studies, impeding further validation, and ultimately hindering their effectiveness in the cumulative scientific progress. In developing a platform for sharing research outputs, we present ModelHub.AI (www.modelhub.ai), a community-driven container-based software engine and platform for the structured dissemination of deep learning models. For contributors, the engine controls data flow throughout the inference cycle, while the contributor-facing standard template exposes model-specific functions including inference, as well as pre- and post-processing. Python and RESTful Application programming interfaces (APIs) enable users to interact with models hosted on ModelHub.AI and allows both researchers and developers to utilize models out-of-the-box. ModelHub.AI is domain-, data-, and framework-agnostic, catering to different workflows and contributors' preferences.
△ Less
Submitted 26 November, 2019;
originally announced November 2019.
-
Growing a LaAlO3/SrTiO3 heterostructure on Ca2Nb3O10 nanosheets
Authors:
Alexander J. H. van der Torren,
Huiyu Yuan,
Zhaoliang Liao,
Johan E. ten Elshof,
Gertjan Koster,
Mark Huijben,
Guus J. H. M. Rijnders,
Marcel B. S. Hesselberth,
Johannes Jobst,
Sense van der Molen,
Jan Aarts
Abstract:
The two-dimensional electron liquid which forms between the band insulators LaAlO3 (LAO) and SrTiO3 (STO) is a promising component for oxide electronics, but the requirement of using single crystal SrTiO3 substrates for the growth limits its applications in terms of device fabrication. It is therefore important to find ways to deposit these materials on other substrates, preferably Si, or Si-based…
▽ More
The two-dimensional electron liquid which forms between the band insulators LaAlO3 (LAO) and SrTiO3 (STO) is a promising component for oxide electronics, but the requirement of using single crystal SrTiO3 substrates for the growth limits its applications in terms of device fabrication. It is therefore important to find ways to deposit these materials on other substrates, preferably Si, or Si-based, in order to facilitate integration with existing technology. Interesting candidates are micron-sized nanosheets of Ca2Nb3O10 which can be used as seed layers for perovskite materials on any substrate. We have used low-energy electron microscopy (LEEM) with in-situ pulsed laser deposition to study the subsequent growth of STO and LAO on such flakes which were deposited on Si. We can follow the morphology and crystallinity of the layers during growth, as well as fingerprint their electronic properties with angle resolved reflected electron spectroscopy. We find that STO layers, deposited on the nanosheets, can be made crystalline and flat; that LAO can be grown in a layer-by-layer fashion; and that the full heterostructure shows the signature of the formation of a conducting interface.
△ Less
Submitted 13 October, 2019;
originally announced October 2019.
-
Improving the Projection of Global Structures in Data through Spanning Trees
Authors:
Daniel Alcaide,
Jan Aerts
Abstract:
The connection of edges in a graph generates a structure that is independent of a coordinate system. This visual metaphor allows creating a more flexible representation of data than a two-dimensional scatterplot. In this work, we present STAD (Spanning Trees as Approximation of Data), a dimensionality reduction method to approximate the high-dimensional structure into a graph with or without formu…
▽ More
The connection of edges in a graph generates a structure that is independent of a coordinate system. This visual metaphor allows creating a more flexible representation of data than a two-dimensional scatterplot. In this work, we present STAD (Spanning Trees as Approximation of Data), a dimensionality reduction method to approximate the high-dimensional structure into a graph with or without formulating prior hypotheses. STAD generates an abstract representation of high-dimensional data by giving each data point a location in a graph which preserves the distances in the original high-dimensional space. The STAD graph is built upon the Minimum Spanning Tree (MST) to which new edges are added until the correlation between the distances from the graph and the original dataset is maximized. Additionally, STAD supports the inclusion of additional functions to focus the exploration and allow the analysis of data from new perspectives, emphasizing traits in data which otherwise would remain hidden. We demonstrate the effectiveness of our method by applying it to two real-world datasets: traffic density in Barcelona and temporal measurements of air quality in Castile and León in Spain.
△ Less
Submitted 12 July, 2019;
originally announced July 2019.
-
Tunable magnetic interactions in LaAlO3/SrTiO3 heterostructures by ionic liquid gating
Authors:
Chunhai Yin,
Kumar Prateek,
Wouter Gelling,
Jan Aarts
Abstract:
The gating effect achieved by an ionic liquid and its electric double layer allows for charge transfer which can be an order of magnitude larger than with conventional dielectrics. However, the large charged ions also causes inevitable Coulomb scattering in the conducting channel formed at the interface, which can limit the carrier mobility enhancement. In this work, we study the effect of the LaA…
▽ More
The gating effect achieved by an ionic liquid and its electric double layer allows for charge transfer which can be an order of magnitude larger than with conventional dielectrics. However, the large charged ions also causes inevitable Coulomb scattering in the conducting channel formed at the interface, which can limit the carrier mobility enhancement. In this work, we study the effect of the LaAlO3 thickness on the transport properties in LaAlO3/SrTiO3 heterostructures by ionic liquid gating. We find that the transport properties of the LaAlO3/SrTiO3 interface are dominated by the intrinsic interactions rather than the LaAlO3 thickness and possible effects from the ions in the liquid. We observe a Kondo effect, which is enhanced while increasing the gate voltage. We also observe a gate-tunable and temperature-dependent anomalous Hall effect, which always emerges near the Kondo temperature. Our experiments pave the way to manipulate the various magnetic interactions in LaAlO3/SrTiO3 heterostructures.
△ Less
Submitted 25 April, 2019;
originally announced April 2019.
-
Tuning Rashba spin-orbit coupling at LaAlO3/SrTiO3 interfaces by band filling
Authors:
Chunhai Yin,
Patrick Seiler,
Lucas M. K. Tang,
Inge Leermakers,
Nikita Lebedev,
Uli Zeitler,
Jan Aarts
Abstract:
The electric-field tunable Rashba spin-orbit coupling at the LaAlO3/SrTiO3 interface shows potential applications in spintronic devices. However, different gate dependence of the coupling strength has been reported in experiments. On the theoretical side, it has been predicted that the largest Rashba effect appears at the crossing point of the $d_{xy}$ and $d_{xz,yz}$ bands. In this work, we study…
▽ More
The electric-field tunable Rashba spin-orbit coupling at the LaAlO3/SrTiO3 interface shows potential applications in spintronic devices. However, different gate dependence of the coupling strength has been reported in experiments. On the theoretical side, it has been predicted that the largest Rashba effect appears at the crossing point of the $d_{xy}$ and $d_{xz,yz}$ bands. In this work, we study the tuneability of the Rashba effect in LaAlO3/SrTiO3 by means of back-gating. The Lifshitz transition was crossed multiple times by tuning the gate voltage so that the Fermi energy is tuned to approach or depart from the band crossing. By analyzing the weak antilocalization behavior in the magnetoresistance, we find that the maximum spin-orbit coupling effect occurs when the Fermi energy is near the Lifshitz point. Moreover, we find strong evidence for a single spin winding at the Fermi surface.
△ Less
Submitted 7 April, 2019;
originally announced April 2019.
-
New insights into the electron trapping mechanism in LaAlO_3 / SrTiO3 heterostructures
Authors:
Chunhai Yin,
Alexander E. M. Smink,
Inge Leermakers,
Lucas M. K. Tang,
Nikita Lebedev,
Uli Zeitler,
Wilfred G. van der Wiel,
Hans Hilgenkamp,
Jan Aarts
Abstract:
In LaAlO3/SrTiO3 heterostructures, a commonly observed but poorly understood phenomenon is that of electron trapping in back-gating experiments. In this work, by combining magnetotransport measurements and self-consistent Schroedinger-Poisson calculations, we obtain an empirical relation between the amount of trapped electrons and the gate voltage. We find that the trapped electrons follow an expo…
▽ More
In LaAlO3/SrTiO3 heterostructures, a commonly observed but poorly understood phenomenon is that of electron trapping in back-gating experiments. In this work, by combining magnetotransport measurements and self-consistent Schroedinger-Poisson calculations, we obtain an empirical relation between the amount of trapped electrons and the gate voltage. We find that the trapped electrons follow an exponentially decaying spatial distribution away from the interface. However, contrary to earlier observations, we find that the Fermi level remains well within the quantum well. The enhanced trapping of electrons induced by the gate voltage can therefore not be explained by a thermal escape mechanism. Further gate sweeping experiments strengthen our conclusion that the thermal escape mechanism is not valid. We propose a new mechanism which involves the electromigration and clustering of oxygen vacancies in SrTiO3. Our work indicates that electron trapping is a universal phenomenon in SrTiO3-based two-dimensional electron systems.
△ Less
Submitted 20 February, 2019;
originally announced February 2019.
-
Quantifying work function differences using low-energy electron microscopy: the case of mixed-terminated strontium titanate
Authors:
Johannes Jobst,
Laurens M. Boers,
Chunhai Yin,
Jan Aarts,
Rudolf M. Tromp,
Sense Jan van der Molen
Abstract:
For many applications, it is important to measure the local work function of a surface with high lateral resolution. Low-energy electron microscopy is regularly employed to this end since it is, in principle, very well suited as it combines high-resolution imaging with high sensitivity to local electrostatic potentials. For surfaces with areas of different work function, however, lateral electrost…
▽ More
For many applications, it is important to measure the local work function of a surface with high lateral resolution. Low-energy electron microscopy is regularly employed to this end since it is, in principle, very well suited as it combines high-resolution imaging with high sensitivity to local electrostatic potentials. For surfaces with areas of different work function, however, lateral electrostatic fields inevitably associated with work function discontinuities deflect the low-energy electrons and thereby cause artifacts near these discontinuities. We use ray-tracing simulations to show that these artifacts extend over hundreds of nanometers and cause an overestimation of the true work function difference near the discontinuity by a factor of 1.6 if the standard image analysis methods are used. We demonstrate on a mixed-terminated strontium titanate surface that comparing LEEM data with detailed ray-tracing simulations leads to much a more robust estimate of the work function difference.
△ Less
Submitted 14 December, 2018;
originally announced December 2018.
-
Influence of stoichiometry on interfacial conductance in LaAlO$_3$/SrTiO$_3$ grown by 90$^o$ off-axis sputtering
Authors:
Chunhai Yin,
Dileep Krishnan,
Nicolas Gauquelin,
Jo Verbeeck,
Jan Aarts
Abstract:
We report on the fabrication of conducting interfaces between LaAlO$_3$ and SrTiO$_3$ by 90$^o$ off-axis sputtering in an Ar atmosphere. At a growth pressure of 0.04 mbar the interface is metallic, with a carrier density of the order of $10^{13}$ cm$^{-2}$ at 3 K. By increasing the growth pressure, we observe an increase of the out-of-plane lattice constants of the LaAlO$_3$ films while the in-pla…
▽ More
We report on the fabrication of conducting interfaces between LaAlO$_3$ and SrTiO$_3$ by 90$^o$ off-axis sputtering in an Ar atmosphere. At a growth pressure of 0.04 mbar the interface is metallic, with a carrier density of the order of $10^{13}$ cm$^{-2}$ at 3 K. By increasing the growth pressure, we observe an increase of the out-of-plane lattice constants of the LaAlO$_3$ films while the in-plane lattice constants do not change. Also, the low-temperature sheet resistance increases with increasing growth pressure, leading to an insulating interface when the growth pressure reaches 0.10 mbar. We attribute the structural variations to an increase of the La/Al ratio, which also explains the transition from metallic behavior to insulating behavior of the interfaces. Our research emphasizes the key role of the cation stoichiometry of LaAlO$_3$ in the formation of the conducting interface, and also the control which is furnished by the Ar pressure in the growth process.
△ Less
Submitted 2 November, 2018;
originally announced November 2018.
-
Repeatability of Multiparametric Prostate MRI Radiomics Features
Authors:
Michael Schwier,
Joost van Griethuysen,
Mark G Vangel,
Steve Pieper,
Sharon Peled,
Clare M Tempany,
Hugo JWL Aerts,
Ron Kikinis,
Fiona M Fennessy,
Andrey Fedorov
Abstract:
In this study we assessed the repeatability of the values of radiomics features for small prostate tumors using test-retest Multiparametric Magnetic Resonance Imaging (mpMRI) images. The premise of radiomics is that quantitative image features can serve as biomarkers characterizing disease. For such biomarkers to be useful, repeatability is a basic requirement, meaning its value must remain stable…
▽ More
In this study we assessed the repeatability of the values of radiomics features for small prostate tumors using test-retest Multiparametric Magnetic Resonance Imaging (mpMRI) images. The premise of radiomics is that quantitative image features can serve as biomarkers characterizing disease. For such biomarkers to be useful, repeatability is a basic requirement, meaning its value must remain stable between two scans, if the conditions remain stable. We investigated repeatability of radiomics features under various preprocessing and extraction configurations including various image normalization schemes, different image pre-filtering, 2D vs 3D texture computation, and different bin widths for image discretization. Image registration as means to re-identify regions of interest across time points was evaluated against human-expert segmented regions in both time points. Even though we found many radiomics features and preprocessing combinations with a high repeatability (Intraclass Correlation Coefficient (ICC) > 0.85), our results indicate that overall the repeatability is highly sensitive to the processing parameters (under certain configurations, it can be below 0.0). Image normalization, using a variety of approaches considered, did not result in consistent improvements in repeatability. There was also no consistent improvement of repeatability through the use of pre-filtering options, or by using image registration between timepoints to improve consistency of the region of interest localization. Based on these results we urge caution when interpreting radiomics features and advise paying close attention to the processing configuration details of reported results. Furthermore, we advocate reporting all processing details in radiomics studies and strongly recommend making the implementation available.
△ Less
Submitted 15 November, 2018; v1 submitted 16 July, 2018;
originally announced July 2018.
-
Little-Parks oscillations with half-quantum fluxoid features in Sr2RuO4 micro rings
Authors:
Yuuki Yasui,
Kaveh Lahabi,
Muhammad Shahbaz Anwar,
Yuji Nakamura,
Shingo Yonezawa,
Takahito Terashima,
Jan Aarts,
Yoshiteru Maeno
Abstract:
In a micro ring of a superconductor with a spin-triplet equal-spin pairing state, a fluxoid, a combined object of magnetic flux and circulating supercurrent, can penetrate as half-integer multiples of the flux quantum. A candidate material to investigate such half-quantum fluxoids is Sr$_\mathsf{2}$RuO$_\mathsf{4}$. We fabricated Sr$_\mathsf{2}$RuO$_\mathsf{4}$ micro rings using single crystals an…
▽ More
In a micro ring of a superconductor with a spin-triplet equal-spin pairing state, a fluxoid, a combined object of magnetic flux and circulating supercurrent, can penetrate as half-integer multiples of the flux quantum. A candidate material to investigate such half-quantum fluxoids is Sr$_\mathsf{2}$RuO$_\mathsf{4}$. We fabricated Sr$_\mathsf{2}$RuO$_\mathsf{4}$ micro rings using single crystals and measured their resistance behavior under magnetic fields controlled with a three-axes vector magnet. Proper Little-Parks oscillations in the magnetovoltage as a function of an axially applied field, associated with fluxoid quantization are clearly observed, for the first time using bulk single crystalline superconductors. We then performed magnetovoltage measurements with additional in-plane magnetic fields. By carefully analyzing both the voltages $V_+$ ($V_-$) measured at positive (negative) current, we find that, above an in-plane threshold field of about 10 mT, the magnetovoltage maxima convert to minima. We interpret this behavior as the peak splitting expected for the half-quantum fluxoid states.
△ Less
Submitted 31 October, 2017;
originally announced October 2017.
-
Formation of a conducting LaAlO$_3$ / SrTiO$_3$ interface studied by low energy electron reflection during growth
Authors:
A. J. H. van der Torren,
Z. Liao,
C. Xu,
N. Gauquelin,
C. Yin,
J. Aarts,
S. J. van der Molen
Abstract:
The two-dimensional electron gas occurring between the band insulators SrTiO$_3$ and LaAlO$_3$ continues to attract considerable interest, due to the possibility of dynamic control over the carrier density, and the ensuing phenomena such as magnetism and superconductivity. The formation of this conducting interface is sensitive to the growth conditions, but despite numerous investigations, there a…
▽ More
The two-dimensional electron gas occurring between the band insulators SrTiO$_3$ and LaAlO$_3$ continues to attract considerable interest, due to the possibility of dynamic control over the carrier density, and the ensuing phenomena such as magnetism and superconductivity. The formation of this conducting interface is sensitive to the growth conditions, but despite numerous investigations, there are still questions about the details of the physics involved. In particular, not much is known about the electronic structure of the growing LaAlO$_3$ layer at the growth temperature (around 800 $^o$C) in oxygen (pressure around $5\times 10^{-5}$ mbar), since analysis techniques at these conditions are not readily available. We developed a pulsed laser deposition system inside a low-energy electron microscope in order to study this issue. The setup allows for layer-by-layer growth control and in-situ measurements of the angle-dependent electron reflection intensity, which can be used as a fingerprint of the electronic structure of the surface layers during growth. By using different substrate terminations and growth conditions we observe two families of reflectivity maps, which we can connect either to samples with an AlO$_2$-rich surface and a conducting interface; or to samples with a LaO-rich surface and an insulating interface. Our observations emphasize that substrate termination and stoichiometry determine the electronic structure of the growing layer, and thereby the conductance of the interface.
△ Less
Submitted 8 September, 2017;
originally announced September 2017.