About
Experience & Education
Publications
-
Adaptive Graph Convolutional Neural Networks
The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18)
Graph Convolutional Neural Networks (Graph CNNs) are
generalizations of classical CNNs to handle graph data such as
molecular data, point could and social networks. Current filters
in graph CNNs are built for fixed and shared graph structure.
However, for most real data, the graph structures varies in both
size and connectivity. The paper proposes a generalized and
flexible graph CNN taking data of arbitrary graph structure as
input. In that way a task-driven adaptive graph…Graph Convolutional Neural Networks (Graph CNNs) are
generalizations of classical CNNs to handle graph data such as
molecular data, point could and social networks. Current filters
in graph CNNs are built for fixed and shared graph structure.
However, for most real data, the graph structures varies in both
size and connectivity. The paper proposes a generalized and
flexible graph CNN taking data of arbitrary graph structure as
input. In that way a task-driven adaptive graph is learned for
each graph data while training. To efficiently learn the graph, a
distance metric learning is proposed. Extensive experiments on
nine graph-structured datasets have demonstrated the superior
performance improvement on both convergence speed and
predictive accuracy. -
Fast Preconditioning for Accelerated Multi-Contrast MRI Reconstruction
18th Annual International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI'15, Munich, Germany, October 2015
Real-time reconstruction in multi-contrast magnetic reso-
nance imaging (MC-MRI) is very challenging due to the slow scanning
and reconstruction process. In this study, we propose a novel algorith-
m to accelerate the MC-MRI reconstruction in the framework of com-
pressed sensing. The problem is formulated as the minimization of the
least square data fitting with joint total variation (JTV) regularization
term. We first utilized the iterative reweighted least square (IRLS)…Real-time reconstruction in multi-contrast magnetic reso-
nance imaging (MC-MRI) is very challenging due to the slow scanning
and reconstruction process. In this study, we propose a novel algorith-
m to accelerate the MC-MRI reconstruction in the framework of com-
pressed sensing. The problem is formulated as the minimization of the
least square data fitting with joint total variation (JTV) regularization
term. We first utilized the iterative reweighted least square (IRLS) frame-
work to reformulate the problem. A joint preconditioner is dexterously
designed to efficiently compute the inverse of large transform matrix
at each iteration. We compared our algorithm with eight cutting-edge
compressive sensing MRI algorithms on real MC-MRI dataset. Exten-
sive experiments demonstrate that the proposed algorithm can achieve
far better reconstruction performance than all other eight cutting-edge
methods.Other authors -
Fast Regions-of-Interest Detection in Whole Slide Histopathology Images
1st International Workshop on Patch-based Techniques in Medical Imaging, PMI'15, Munich, Germany, October 2015
In this paper, we present a novel superpixel based Region
of Interest (ROI) search and segmentation algorithm. The proposed su-
perpixel generation method differs from pioneer works due to its combi-
nation of boundary update and coarse-to-fine refinement for superpixel
clustering. The former maintains the accuracy of segmentation, mean-
while, avoids much of unnecessary revisit to the ‘non-boundary’ pixels.
The latter reduces the complexity by faster localizing those…In this paper, we present a novel superpixel based Region
of Interest (ROI) search and segmentation algorithm. The proposed su-
perpixel generation method differs from pioneer works due to its combi-
nation of boundary update and coarse-to-fine refinement for superpixel
clustering. The former maintains the accuracy of segmentation, mean-
while, avoids much of unnecessary revisit to the ‘non-boundary’ pixels.
The latter reduces the complexity by faster localizing those boundary
blocks. The paper introduces the novel superpixel algorithm [10] to the
problem of ROI detection and segmentation along with a coarse-to-fine
refinement scheme over a set of image of different magnification. Exten-
sive experiments indicates that the proposed method gives better accu-
racy and efficiency than other superpixel-based methods for lung can-
cer cell images. Moreover, the block-wise coarse-to-fine scheme enables
a quick search and segmentation of ROIs in whole slide images, while,
other methods still cannot.Other authors
Courses
-
Algorithms
-
-
Compiler
-
-
Data Modeling and Analysis
-
-
Distributed System
-
-
Game Theory
-
-
Machine Learning
-
-
Medical Image Processing
-
-
Statistical Signal Processing
-
Languages
-
Chinese
Native or bilingual proficiency
-
English
Professional working proficiency
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Ruoyu Li in United States
-
Ruoyu Li
MSIS Student at Northeastern University | CMU 23' | Aspiring Software Engineer | Seeking Summer 2025 Internship
-
Ruoyu Li
Available for booking inquiries
-
Ruoyu Li
Quantum Computing Scientist, Office of the CTO, Applied Materials
-
Ruoyu LI
MS Biostatistics @ UCLA
33 others named Ruoyu Li in United States are on LinkedIn
See others named Ruoyu Li