Lymphoma segmentation from 3D PET-CT images using a deep evidential network. (English) Zbl 07581228
Summary: An automatic evidential segmentation method based on Dempster-Shafer theory and deep learning is proposed to segment lymphomas from three-dimensional Positron Emission Tomography (PET) and Computed Tomography (CT) images. The architecture is composed of a deep feature-extraction module and an evidential layer. The feature extraction module uses an encoder-decoder framework to extract semantic feature vectors from 3D inputs. The evidential layer then uses prototypes in the feature space to compute a belief function at each voxel quantifying the uncertainty about the presence or absence of a lymphoma at this location. Two evidential layers are compared, based on different ways of using distances to prototypes for computing mass functions. The whole model is trained end-to-end by minimizing the Dice loss function. The proposed combination of deep feature extraction and evidential segmentation is shown to outperform the baseline UNet model as well as three other state-of-the-art models on a dataset of 173 patients.
MSC:
| 68T37 | Reasoning under uncertainty in the context of artificial intelligence |
Keywords:
medical image analysis; Dempster-Shafer theory; evidence theory; belief functions; uncertainty quantification; deep learningReferences:
| [1] | Jhanwar, Y. S.; Straus, D. J., The role of PET in lymphoma, J. Nucl. Med., 47, 8, 1326-1334 (2006) |
| [2] | Zaidi, H.; El Naqa, I., PET-guided delineation of radiation therapy treatment volumes: a survey of image segmentation techniques, Eur. J. Nucl. Med. Mol. Imaging, 37, 11, 2165-2187 (2010) |
| [3] | Ilyas, H.; Mikhaeel, N. G.; Dunn, J. T.; Rahman, F.; Møller, H.; Smith, D.; Barrington, S. F., Defining the optimal method for measuring baseline metabolic tumour volume in diffuse large B cell lymphoma, Eur. J. Nucl. Med. Mol. Imaging, 45, 7, 1142-1154 (2018) |
| [4] | Eude, F.; Toledano, M. N.; Vera, P.; Tilly, H.; Mihailescu, S.-D.; Becker, S., Reproducibility of baseline tumour metabolic volume measurements in diffuse large B-cell lymphoma: is there a superior method?, Metabolites, 11, 2, 72 (2021) |
| [5] | Onoma, D.; Ruan, S.; Thureau, S., Segmentation of heterogeneous or small FDG PET positive tissue based on a 3d-locally adaptive random walk algorithm, Comput. Med. Imaging Graph., 38, 8, 753-763 (2014) |
| [6] | Hu, H.; Decazes, P.; Vera, P.; Li, H.; Ruan, S., Detection and segmentation of lymphomas in 3D PET images via clustering with entropy-based optimization strategy, Int. J. Comput. Assisted Radiol. Surg., 14, 10, 1715-1724 (2019) |
| [7] | Li, H.; Jiang, H.; Li, S., DenseX-net: an end-to-end model for lymphoma segmentation in whole-body PET/CT images, IEEE Access, 8, 8004-8018 (2019) |
| [8] | Hu, H.; Shen, L.; Zhou, T., Lymphoma segmentation in PET images based on multi-view and conv3d fusion strategy, (2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI) (2020), IEEE), 1197-1200 |
| [9] | Long, J.; Shelhamer, E.; Darrell, T., Fully convolutional networks for semantic segmentation, (Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, USA (2015)), 3431-3440 |
| [10] | Ronneberger, O.; Fischer, P.; Brox, T., U-Net: convolutional networks for biomedical image segmentation, (Navab, N.; Hornegger, J.; Wells, W. M.; Frangi, A. F., Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015 (2015), Springer International Publishing: Springer International Publishing Cham), 234-241 |
| [11] | Milletari, F.; Navab, N.; Ahmadi, S.-A., V-net: fully convolutional neural networks for volumetric medical image segmentation, (2016 Fourth International Conference on 3D Vision (2016), IEEE), 565-571 |
| [12] | Myronenko, A., 3D MRI brain tumor segmentation using autoencoder regularization, (International MICCAI Brain Lesion Workshop (2018), Springer), 311-320 |
| [13] | Isensee, F.; Petersen, J.; Klein, A.; Zimmerer, D., Nnu-net: self-adapting framework for u-net-based medical image segmentation, arXiv preprint |
| [14] | Blanc-Durand, P.; Jégou, S.; Kanoun, S., Fully automatic segmentation of diffuse large B cell lymphoma lesions on 3D FDG-PET/CT for total metabolic tumour volume prediction using a convolutional neural network, Eur. J. Nucl. Med. Mol. Imaging, 1-9 (2020) |
| [15] | Huang, L.; Denœux, T.; Tonnelet, D.; Decazes, P.; Ruan, S., Deep pet/ct fusion with Dempster-Shafer theory for lymphoma segmentation, (Lian, C.; Cao, X.; Rekik, I.; Xu, X.; Yan, P., Machine Learning in Medical Imaging (2021), Springer International Publishing: Springer International Publishing Cham), 30-39 |
| [16] | Shafer, G., A Mathematical Theory of Evidence, vol. 42 (1976), Princeton University Press · Zbl 0359.62002 |
| [17] | Hüllermeier, E.; Waegeman, W., Aleatoric and epistemic uncertainty in machine learning: an introduction to concepts and methods, Mach. Learn., 110, 3, 457-506 (2021) · Zbl 07432810 |
| [18] | Quiñonero-Candela, J.; Sugiyama, M.; Lawrence, N. D.; Schwaighofer, A., Dataset Shift in Machine Learning (2009), MIT Press |
| [19] | Mehta, R.; Christinck, T.; Nair, T.; Lemaitre, P.; Arnold, D.; Arbel, T., Propagating uncertainty across cascaded medical imaging tasks for improved deep learning inference, (Uncertainty for Safe Utilization of Machine Learning in Medical Imaging and Clinical Image-Based Procedures (2019), Springer), 23-32 |
| [20] | Maddox, W. J.; Izmailov, P.; Garipov, T.; Vetrov, D. P.; Wilson, A. G., A simple baseline for Bayesian uncertainty in deep learning, Adv. Neural Inf. Process. Syst., 32, 13153-13164 (2019) |
| [21] | Yu, L.; Wang, S.; Li, X.; Fu, C.-W.; Heng, P.-A., Uncertainty-aware self-ensembling model for semi-supervised 3d left atrium segmentation, (International Conference on Medical Image Computing and Computer-Assisted Intervention (2019), Springer), 605-613 |
| [22] | Ghesu, F. C.; Georgescu, B.; Mansoor, A.; Yoo, Y.; Gibson, E.; Vishwanath, R.; Balachandran, A.; Balter, J. M.; Cao, Y.; Singh, R., Quantifying and leveraging predictive uncertainty for medical image assessment, Med. Image Anal., 68, Article 101855 pp. (2021) |
| [23] | Hinton, G. E.; van Camp, D., Keeping the neural networks simple by minimizing the description length of the weights, (Proceedings of the Sixth Annual Conference on Computational Learning Theory, COLT ’93 (1993), Association for Computing Machinery: Association for Computing Machinery New York, NY, USA), 5-13 |
| [24] | MacKay, D. J., A practical Bayesian framework for backpropagation networks, Neural Comput., 4, 3, 448-472 (1992) |
| [25] | Gal, Y.; Ghahramani, Z., Dropout as a Bayesian approximation: representing model uncertainty in deep learning, (International Conference on Machine Learning (2016), PMLR), 1050-1059 |
| [26] | Tran, D.; Dusenberry, M. W.; van der Wilk, M.; Hafner, D., Bayesian layers: a module for neural network uncertainty, arXiv preprint |
| [27] | Dempster, A. P., Upper and lower probability inferences based on a sample from a finite univariate population, Biometrika, 54, 3-4, 515-528 (1967) |
| [28] | Denœux, T.; Dubois, D.; Prade, H., Representations of uncertainty in artificial intelligence: beyond probability and possibility, (Marquis, P.; Papini, O.; Prade, H., A Guided Tour of Artificial Intelligence Research, vol. 1 (2020), Springer Verlag), 119-150, Ch. 4 |
| [29] | Smets, P., The combination of evidence in the transferable belief model, IEEE Trans. Pattern Anal. Mach. Intell., 12, 5, 447-458 (1990) |
| [30] | Denoeux, T., A k-nearest neighbor classification rule based on Dempster-Shafer theory, IEEE Trans. Syst. Man Cybern., 25, 5, 804-813 (1995) |
| [31] | Denœux, T., A neural network classifier based on Dempster-Shafer theory, IEEE Trans. Syst. Man Cybern., Part A, Syst. Hum., 30, 2, 131-150 (2000) |
| [32] | Denœux, T.; Masson, M.-H., Evclus: evidential clustering of proximity data, IEEE Trans. Syst. Man Cybern., Part B, Cybern., 34, 1, 95-109 (2004) |
| [33] | Pichon, F.; Mercier, D.; Lefèvre, E.; Delmotte, F., Proposition and learning of some belief function contextual correction mechanisms, Int. J. Approx. Reason., 72, 4-42 (2016) · Zbl 1352.68250 |
| [34] | Pichon, F.; Dubois, D.; Denœux, T., Quality of information sources in information fusion, (Bossé, É.; Rogova, G. L., Information Quality in Information Fusion and Decision Making (2019), Springer International Publishing: Springer International Publishing Cham), 31-49 |
| [35] | Chen, H.; Le Hégarat-Mascle, S.; Aldea, E., Belief functions clustering for epipole localization, Int. J. Approx. Reason., 137, 146-165 (2021) · Zbl 1520.68198 |
| [36] | Denœux, T.; Kanjanatarakul, O.; Sriboonchitta, S., A new evidential k-nearest neighbor rule based on contextual discounting with partially supervised learning, Int. J. Approx. Reason., 113, 287-302 (2019) · Zbl 1468.68151 |
| [37] | Gong, C.; Gang Su, Z.; Hong Wang, P.; Wang, Q.; You, Y., Evidential instance selection for k-nearest neighbor classification of big data, Int. J. Approx. Reason., 138, 123-144 (2021) · Zbl 1520.68173 |
| [38] | Imoussaten, A.; Jacquin, L., Cautious classification based on belief functions theory and imprecise relabelling, Int. J. Approx. Reason., 142, 130-146 (2022) · Zbl 07478942 |
| [39] | Denoeux, T., NN-EVCLUS: neural network-based evidential clustering, Inf. Sci., 572, 297-330 (2021) · Zbl 1528.68372 |
| [40] | Antoine, V.; Guerrero, J. A.; Xie, J., Fast semi-supervised evidential clustering, Int. J. Approx. Reason., 133, 116-132 (2021) · Zbl 1522.68547 |
| [41] | Lian, C.; Ruan, S.; Denœux, T.; Li, H.; Vera, P., Joint tumor segmentation in PET-CT images using co-clustering and fusion based on belief functions, IEEE Trans. Image Process., 28, 2, 755-766 (2018) · Zbl 1409.94329 |
| [42] | Huang, L.; Ruan, S.; Denœux, T., Belief function-based semi-supervised learning for brain tumor segmentation, arXiv preprint |
| [43] | Tong, Z.; Xu, P.; Denœux, T., Evidential fully convolutional network for semantic segmentation, Appl. Intell., 51, 6376-6399 (2021) |
| [44] | Huang, L.; Ruan, S.; Decazes, P.; Denœux, T., Evidential segmentation of 3D PET/CT images, (Denœux, T.; Lefèvre, E.; Liu, Z.; Pichon, F., Belief Functions: Theory and Applications (2021), Springer International Publishing: Springer International Publishing Cham), 159-167 |
| [45] | Smets, P.; Kennes, R., The transferable belief model, Artif. Intell., 66, 191-243 (1994) · Zbl 0807.68087 |
| [46] | Denœux, T., Analysis of evidence-theoretic decision rules for pattern classification, Pattern Recognit., 30, 7, 1095-1107 (1997) |
| [47] | Ma, L.; Denœux, T., Partial classification in the belief function framework, Knowl.-Based Syst., 214, Article 106742 pp. (2021) |
| [48] | Denoeux, T., Decision-making with belief functions: a review, Int. J. Approx. Reason., 109, 87-110 (2019) · Zbl 1465.91038 |
| [49] | Denœux, T., Logistic regression, neural networks and Dempster-Shafer theory: a new perspective, Knowl.-Based Syst., 176, 54-67 (2019) |
| [50] | Tong, Z.; Xu, P.; Denœux, T., An evidential classifier based on Dempster-Shafer theory and deep learning, Neurocomputing, 450, 275-293 (2021) |
| [51] | Kerfoot, E.; Clough, J.; Oksuz, I.; Lee, J.; King, A. P.; Schnabel, J. A., Left-ventricle quantification using residual u-net, (International Workshop on Statistical Atlases and Computational Models of the Heart (2018), Springer), 371-380 |
| [52] | Ulyanov, D.; Vedaldi, A.; Lempitsky, V., Improved texture networks: maximizing quality and diversity in feed-forward stylization and texture synthesis, (Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2017)), 6924-6932 |
| [53] | He, K.; Zhang, X.; Ren, S.; Sun, J., Delving deep into rectifiers: surpassing human-level performance on imagenet classification, (Proceedings of the IEEE International Conference on Computer Vision (2015)), 1026-1034 |
| [54] | Lowekamp, B.; Chen, D.; Ibáñez, L.; Blezek, D., The design of SimpleITK, Front. Neuroinform., 7, 45 (2013) |
| [55] | Yaniv, Z.; Lowekamp, B. C.; Johnson, H. J.; Beare, R., SimpleITK image-analysis notebooks: a collaborative environment for education and reproducible research, J. Digit. Imaging, 31, 3, 290-303 (2018) |
| [56] | Guo, C.; Pleiss, G.; Sun, Y.; Weinberger, K. Q., On calibration of modern neural networks, (International Conference on Machine Learning (2017), PMLR), 1321-1330 |
| [57] | Jungo, A.; Balsiger, F.; Reyes, M., Analyzing the quality and challenges of uncertainty estimations for brain tumor segmentation, Front. Neurosci., 14, 282 (2020) |
| [58] | Rousseau, A.-J.; Becker, T.; Bertels, J.; Blaschko, M. B.; Valkenborg, D., Post training uncertainty calibration of deep networks for medical image segmentation, (2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI) (2021), IEEE), 1052-1056 |
| [59] | Conover, W. J.; Iman, R. L., On multiple-comparisons procedures (1979), Los Alamos Scientific Laboratory, Tech. Rep. LA-7677-MS |
| [60] | Dinno, A., Conover.test: conover-iman test of multiple comparisons using rank sums (2017), r package version 1.1.5 |
| [61] | Benjamini, Y.; Yekutieli, D., The control of the false discovery rate in multiple testing under dependency, Ann. Stat., 29, 4, 1165-1188 (2001) · Zbl 1041.62061 |
| [62] | Hryniowski, A.; Wong, A., Deeplabnet: end-to-end learning of deep radial basis networks, J. Comput. Vis. Imag. Syst., 5, 1, 1 (2020) |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
