Pub Date : 2020-09-07DOI: 10.59275/j.melba.2021-bbf2
Jinwei Zhang, Hang Zhang, M. Sabuncu, P. Spincemaille, Thanh D. Nguyen, Yi Wang
A learning-based posterior distribution estimation method, Probabilistic Dipole Inversion (PDI), is proposed to solve the quantitative susceptibility mapping (QSM) inverse problem in MRI with uncertainty estimation. In PDI, a deep convolutional neural network (CNN) is used to represent the multivariate Gaussian distribution as the approximate posterior distribution of susceptibility given the input measured field. Such CNN is first trained on healthy subjects via posterior density estimation, where the training dataset contains samples from the true posterior distribution. Domain adaptations are then deployed on patient datasets with new pathologies not included in pre-training, where PDI updates the pre-trained CNN’s weights in an unsupervised fashion by minimizing the Kullback-Leibler divergence between the approximate posterior distribution represented by CNN and the true posterior distribution from the likelihood distribution of a known physical model and pre-defined prior distribution. Based on our experiments, PDI provides additional uncertainty estimation compared to the conventional MAP approach, while addressing the potential issue of the pre-trained CNN when test data deviates from training. Our code is available at https://github.com/Jinwei1209/Bayesian_QSM.
{"title":"Probabilistic dipole inversion for adaptive quantitative susceptibility mapping","authors":"Jinwei Zhang, Hang Zhang, M. Sabuncu, P. Spincemaille, Thanh D. Nguyen, Yi Wang","doi":"10.59275/j.melba.2021-bbf2","DOIUrl":"https://doi.org/10.59275/j.melba.2021-bbf2","url":null,"abstract":"A learning-based posterior distribution estimation method, Probabilistic Dipole Inversion (PDI), is proposed to solve the quantitative susceptibility mapping (QSM) inverse problem in MRI with uncertainty estimation. In PDI, a deep convolutional neural network (CNN) is used to represent the multivariate Gaussian distribution as the approximate posterior distribution of susceptibility given the input measured field. Such CNN is first trained on healthy subjects via posterior density estimation, where the training dataset contains samples from the true posterior distribution. Domain adaptations are then deployed on patient datasets with new pathologies not included in pre-training, where PDI updates the pre-trained CNN’s weights in an unsupervised fashion by minimizing the Kullback-Leibler divergence between the approximate posterior distribution represented by CNN and the true posterior distribution from the likelihood distribution of a known physical model and pre-defined prior distribution. Based on our experiments, PDI provides additional uncertainty estimation compared to the conventional MAP approach, while addressing the potential issue of the pre-trained CNN when test data deviates from training. Our code is available at https://github.com/Jinwei1209/Bayesian_QSM.","PeriodicalId":75083,"journal":{"name":"The journal of machine learning for biomedical imaging","volume":"745 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76841908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-08-04DOI: 10.59275/j.melba.2023-3462
Mara Graziani, Sebastian Otálora, S. Marchand-Maillet, H. Muller, V. Andrearczyk
Adopting Convolutional Neural Networks (CNNs) in the daily routine of primary diagnosis requires not only near-perfect precision, but also a sufficient degree of generalization to data acquisition shifts and transparency. Existing CNN models act as black boxes, not ensuring to the physicians that important diagnostic features are used by the model. Building on top of successfully existing techniques such as multi-task learning, domain adversarial training and concept-based interpretability, this paper addresses the challenge of introducing diagnostic factors in the training objectives. Here we show that our architecture, by learning end-to-end an uncertainty-based weighting combination of multi-task and adversarial losses, is encouraged to focus on pathology features such as density and pleomorphism of nuclei, e.g. variations in size and appearance, while discarding misleading features such as staining differences. Our results on breast lymph node tissue show significantly improved generalization in the detection of tumorous tissue, with best average AUC 0.89 (0.01) against the baseline AUC 0.86 (0.005). By applying the interpretability technique of linearly probing intermediate representations, we also demonstrate that interpretable pathology features such as nuclei density are learned by the proposed CNN architecture, confirming the increased transparency of this model. This result is a starting point towards building interpretable multi-task architectures that are robust to data heterogeneity. Our code is available at https://github.com/maragraziani/multitask_adversarial
{"title":"Learning Interpretable Microscopic Features of Tumor by Multi-task Adversarial CNNs To Improve Generalization","authors":"Mara Graziani, Sebastian Otálora, S. Marchand-Maillet, H. Muller, V. Andrearczyk","doi":"10.59275/j.melba.2023-3462","DOIUrl":"https://doi.org/10.59275/j.melba.2023-3462","url":null,"abstract":"Adopting Convolutional Neural Networks (CNNs) in the daily routine of primary diagnosis requires not only near-perfect precision, but also a sufficient degree of generalization to data acquisition shifts and transparency. Existing CNN models act as black boxes, not ensuring to the physicians that important diagnostic features are used by the model. Building on top of successfully existing techniques such as multi-task learning, domain adversarial training and concept-based interpretability, this paper addresses the challenge of introducing diagnostic factors in the training objectives. Here we show that our architecture, by learning end-to-end an uncertainty-based weighting combination of multi-task and adversarial losses, is encouraged to focus on pathology features such as density and pleomorphism of nuclei, e.g. variations in size and appearance, while discarding misleading features such as staining differences. Our results on breast lymph node tissue show significantly improved generalization in the detection of tumorous tissue, with best average AUC 0.89 (0.01) against the baseline AUC 0.86 (0.005). By applying the interpretability technique of linearly probing intermediate representations, we also demonstrate that interpretable pathology features such as nuclei density are learned by the proposed CNN architecture, confirming the increased transparency of this model. This result is a starting point towards building interpretable multi-task architectures that are robust to data heterogeneity. Our code is available at https://github.com/maragraziani/multitask_adversarial","PeriodicalId":75083,"journal":{"name":"The journal of machine learning for biomedical imaging","volume":"4 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88793048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-02DOI: 10.59275/j.melba.2022-16cc
Nalini Singh, J. E. Iglesias, E. Adalsteinsson, Adrian V. Dalca, P. Golland
We propose neural network layers that explicitly combine frequency and image feature representations and show that they can be used as a versatile building block for reconstruction from frequency space data. Our work is motivated by the challenges arising in MRI acquisition where the signal is a corrupted Fourier transform of the desired image. The proposed joint learning schemes enable both correction of artifacts native to the frequency space and manipulation of image space representations to reconstruct coherent image structures at every layer of the network. This is in contrast to most current deep learning approaches for image reconstruction that treat frequency and image space features separately and often operate exclusively in one of the two spaces. We demonstrate the advantages of joint convolutional learning for a variety of tasks, including motion correction, denoising, reconstruction from undersampled acquisitions, and combined undersampling and motion correction on simulated and real world multicoil MRI data. The joint models produce consistently high quality output images across all tasks and datasets. When integrated into a state of the art unrolled optimization network with physics-inspired data consistency constraints for undersampled reconstruction, the proposed architectures significantly improve the optimization landscape, which yields an order of magnitude reduction of training time. This result suggests that joint representations are particularly well suited for MRI signals in deep learning networks. Our code and pretrained models are publicly available at https://github.com/nalinimsingh/interlacer.
{"title":"Joint Frequency and Image Space Learning for MRI Reconstruction and Analysis","authors":"Nalini Singh, J. E. Iglesias, E. Adalsteinsson, Adrian V. Dalca, P. Golland","doi":"10.59275/j.melba.2022-16cc","DOIUrl":"https://doi.org/10.59275/j.melba.2022-16cc","url":null,"abstract":"We propose neural network layers that explicitly combine frequency and image feature representations and show that they can be used as a versatile building block for reconstruction from frequency space data. Our work is motivated by the challenges arising in MRI acquisition where the signal is a corrupted Fourier transform of the desired image. The proposed joint learning schemes enable both correction of artifacts native to the frequency space and manipulation of image space representations to reconstruct coherent image structures at every layer of the network. This is in contrast to most current deep learning approaches for image reconstruction that treat frequency and image space features separately and often operate exclusively in one of the two spaces. We demonstrate the advantages of joint convolutional learning for a variety of tasks, including motion correction, denoising, reconstruction from undersampled acquisitions, and combined undersampling and motion correction on simulated and real world multicoil MRI data. The joint models produce consistently high quality output images across all tasks and datasets. When integrated into a state of the art unrolled optimization network with physics-inspired data consistency constraints for undersampled reconstruction, the proposed architectures significantly improve the optimization landscape, which yields an order of magnitude reduction of training time. This result suggests that joint representations are particularly well suited for MRI signals in deep learning networks. Our code and pretrained models are publicly available at https://github.com/nalinimsingh/interlacer.","PeriodicalId":75083,"journal":{"name":"The journal of machine learning for biomedical imaging","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86189684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-02-09DOI: 10.59275/j.melba.2021-2dcc
Razvan V. Marinescu, N. Oxtoby, A. Young, E. Bron, A. Toga, M. Weiner, F. Barkhof, Nick C Fox, A. Eshaghi, Tina Toni, Marcin Salaterski, V. Lunina, M. Ansart, S. Durrleman, Pascal Lu, S. Iddi, Dan Li, W. Thompson, M. Donohue, A. Nahon, Yarden Levy, Dan Halbersberg, M. Cohen, Huiling Liao, Tengfei Li, Kaixian Yu, Hongtu Zhu, Jose Gerardo Tamez-Peña, A. Ismail, Timothy Wood, H. C. Bravo, Minh Nguyen, Nanbo Sun, Jiashi Feng, B. Yeo, Gan Chen, Kexin Qi, Shi-Yu Chen, D. Qiu, I. Buciuman, A. Kelner, R. Pop, Denisa Rimocea, M. Ghazi, M. Nielsen, S. Ourselin, Lauge Sørensen, Vikram Venkatraghavan, Keli Liu, C. Rabe, P. Manser, S. Hill, J. Howlett, Zhiyue Huang, S. Kiddle, S. Mukherjee, Anaïs Rouanet, B. Taschler, B. Tom, S. White, N. Faux, S. Sedai, Javier de Velasco Oriol, Edgar E. V. Clemente, K. Estrada, Leon M. Aksman, A. Altmann, C. Stonnington, Yalin Wang, Jianfeng Wu, Vivek Devadas, Clémentine Fourrier, L. L. Rakêt, Aristeidis Sotiras, G. Erus, J. Doshi, C. Davatzikos, J. Vogel, Andrew Doyle, Angela Tam, A
Accurate prediction of progression in subjects at risk of Alzheimer's disease is crucial for enrolling the right subjects in clinical trials. However, a prospective comparison of state-of-the-art algorithms for predicting disease onset and progression is currently lacking. We present the findings of "The Alzheimer's Disease Prediction Of Longitudinal Evolution" (TADPOLE) Challenge, which compared the performance of 92 algorithms from 33 international teams at predicting the future trajectory of 219 individuals at risk of Alzheimer's disease. Challenge participants were required to make a prediction, for each month of a 5-year future time period, of three key outcomes: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog13), and total volume of the ventricles. The methods used by challenge participants included multivariate linear regression, machine learning methods such as support vector machines and deep neural networks, as well as disease progression models. No single submission was best at predicting all three outcomes. For clinical diagnosis and ventricle volume prediction, the best algorithms strongly outperform simple baselines in predictive ability. However, for ADAS-Cog13 no single submitted prediction method was significantly better than random guesswork. Two ensemble methods based on taking the mean and median over all predictions, obtained top scores on almost all tasks. Better than average performance at diagnosis prediction was generally associated with the additional inclusion of features from cerebrospinal fluid (CSF) samples and diffusion tensor imaging (DTI). On the other hand, better performance at ventricle volume prediction was associated with inclusion of summary statistics, such as the slope or maxima/minima of patient-specific biomarkers. On a limited, cross-sectional subset of the data emulating clinical trials, performance of the best algorithms at predicting clinical diagnosis decreased only slightly (2 percentage points) compared to the full longitudinal dataset. The submission system remains open via the website https://tadpole.grand-challenge.org, while TADPOLE SHARE (https://tadpole-share.github.io/) collates code for submissions. TADPOLE's unique results suggest that current prediction algorithms provide sufficient accuracy to exploit biomarkers related to clinical diagnosis and ventricle volume, for cohort refinement in clinical trials for Alzheimer's disease. However, results call into question the usage of cognitive test scores for patient selection and as a primary endpoint in clinical trials.
{"title":"The Alzheimer's Disease Prediction Of Longitudinal Evolution (TADPOLE) Challenge: Results after 1 Year Follow-up","authors":"Razvan V. Marinescu, N. Oxtoby, A. Young, E. Bron, A. Toga, M. Weiner, F. Barkhof, Nick C Fox, A. Eshaghi, Tina Toni, Marcin Salaterski, V. Lunina, M. Ansart, S. Durrleman, Pascal Lu, S. Iddi, Dan Li, W. Thompson, M. Donohue, A. Nahon, Yarden Levy, Dan Halbersberg, M. Cohen, Huiling Liao, Tengfei Li, Kaixian Yu, Hongtu Zhu, Jose Gerardo Tamez-Peña, A. Ismail, Timothy Wood, H. C. Bravo, Minh Nguyen, Nanbo Sun, Jiashi Feng, B. Yeo, Gan Chen, Kexin Qi, Shi-Yu Chen, D. Qiu, I. Buciuman, A. Kelner, R. Pop, Denisa Rimocea, M. Ghazi, M. Nielsen, S. Ourselin, Lauge Sørensen, Vikram Venkatraghavan, Keli Liu, C. Rabe, P. Manser, S. Hill, J. Howlett, Zhiyue Huang, S. Kiddle, S. Mukherjee, Anaïs Rouanet, B. Taschler, B. Tom, S. White, N. Faux, S. Sedai, Javier de Velasco Oriol, Edgar E. V. Clemente, K. Estrada, Leon M. Aksman, A. Altmann, C. Stonnington, Yalin Wang, Jianfeng Wu, Vivek Devadas, Clémentine Fourrier, L. L. Rakêt, Aristeidis Sotiras, G. Erus, J. Doshi, C. Davatzikos, J. Vogel, Andrew Doyle, Angela Tam, A","doi":"10.59275/j.melba.2021-2dcc","DOIUrl":"https://doi.org/10.59275/j.melba.2021-2dcc","url":null,"abstract":"Accurate prediction of progression in subjects at risk of Alzheimer's disease is crucial for enrolling the right subjects in clinical trials. However, a prospective comparison of state-of-the-art algorithms for predicting disease onset and progression is currently lacking. We present the findings of \"The Alzheimer's Disease Prediction Of Longitudinal Evolution\" (TADPOLE) Challenge, which compared the performance of 92 algorithms from 33 international teams at predicting the future trajectory of 219 individuals at risk of Alzheimer's disease. Challenge participants were required to make a prediction, for each month of a 5-year future time period, of three key outcomes: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog13), and total volume of the ventricles. The methods used by challenge participants included multivariate linear regression, machine learning methods such as support vector machines and deep neural networks, as well as disease progression models. No single submission was best at predicting all three outcomes. For clinical diagnosis and ventricle volume prediction, the best algorithms strongly outperform simple baselines in predictive ability. However, for ADAS-Cog13 no single submitted prediction method was significantly better than random guesswork. Two ensemble methods based on taking the mean and median over all predictions, obtained top scores on almost all tasks. Better than average performance at diagnosis prediction was generally associated with the additional inclusion of features from cerebrospinal fluid (CSF) samples and diffusion tensor imaging (DTI). On the other hand, better performance at ventricle volume prediction was associated with inclusion of summary statistics, such as the slope or maxima/minima of patient-specific biomarkers. On a limited, cross-sectional subset of the data emulating clinical trials, performance of the best algorithms at predicting clinical diagnosis decreased only slightly (2 percentage points) compared to the full longitudinal dataset. The submission system remains open via the website https://tadpole.grand-challenge.org, while TADPOLE SHARE (https://tadpole-share.github.io/) collates code for submissions. TADPOLE's unique results suggest that current prediction algorithms provide sufficient accuracy to exploit biomarkers related to clinical diagnosis and ventricle volume, for cohort refinement in clinical trials for Alzheimer's disease. However, results call into question the usage of cognitive test scores for patient selection and as a primary endpoint in clinical trials.","PeriodicalId":75083,"journal":{"name":"The journal of machine learning for biomedical imaging","volume":"45 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89767080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-01-08DOI: 10.59275/j.melba.2022-8b6a
Lucas Fidon, S. Ourselin, T. Vercauteren
Limiting failures of machine learning systems is of paramount importance for safety-critical applications. In order to improve the robustness of machine learning systems, Distributionally Robust Optimization (DRO) has been proposed as a generalization of Empirical Risk Minimization (ERM). However, its use in deep learning has been severely restricted due to the relative inefficiency of the optimizers available for DRO in comparison to the wide-spread variants of Stochastic Gradient Descent (SGD) optimizers for ERM.We propose SGD with hardness weighted sampling, a principled and efficient optimization method for DRO in machine learning that is particularly suited in the context of deep learning. Similar to a hard example mining strategy in practice, the proposed algorithm is straightforward to implement and computationally as efficient as SGD-based optimizers used for deep learning, requiring minimal overhead computation. In contrast to typical ad hoc hard mining approaches, we prove the convergence of our DRO algorithm for over-parameterized deep learning networks with ReLU activation and finite number of layers and parameters.Our experiments on fetal brain 3D MRI segmentation and brain tumor segmentation in MRI demonstrate the feasibility and the usefulness of our approach. Using our hardness weighted sampling for training a state-of-the-art deep learning pipeline leads to improved robustness to anatomical variabilities in automatic fetal brain 3D MRI segmentation using deep learning and to improved robustness to the image protocol variations in brain tumor segmentation.a decrease of 2% of the interquartile range of the Dice scores for the enhanced tumor and the tumor core regions.Our code is available at https://github.com/LucasFidon/HardnessWeightedSampler
{"title":"Distributionally Robust Deep Learning using Hardness Weighted Sampling","authors":"Lucas Fidon, S. Ourselin, T. Vercauteren","doi":"10.59275/j.melba.2022-8b6a","DOIUrl":"https://doi.org/10.59275/j.melba.2022-8b6a","url":null,"abstract":"Limiting failures of machine learning systems is of paramount importance for safety-critical applications. In order to improve the robustness of machine learning systems, Distributionally Robust Optimization (DRO) has been proposed as a generalization of Empirical Risk Minimization (ERM). However, its use in deep learning has been severely restricted due to the relative inefficiency of the optimizers available for DRO in comparison to the wide-spread variants of Stochastic Gradient Descent (SGD) optimizers for ERM.We propose SGD with hardness weighted sampling, a principled and efficient optimization method for DRO in machine learning that is particularly suited in the context of deep learning. Similar to a hard example mining strategy in practice, the proposed algorithm is straightforward to implement and computationally as efficient as SGD-based optimizers used for deep learning, requiring minimal overhead computation. In contrast to typical ad hoc hard mining approaches, we prove the convergence of our DRO algorithm for over-parameterized deep learning networks with ReLU activation and finite number of layers and parameters.Our experiments on fetal brain 3D MRI segmentation and brain tumor segmentation in MRI demonstrate the feasibility and the usefulness of our approach. Using our hardness weighted sampling for training a state-of-the-art deep learning pipeline leads to improved robustness to anatomical variabilities in automatic fetal brain 3D MRI segmentation using deep learning and to improved robustness to the image protocol variations in brain tumor segmentation.a decrease of 2% of the interquartile range of the Dice scores for the enhanced tumor and the tumor core regions.Our code is available at https://github.com/LucasFidon/HardnessWeightedSampler","PeriodicalId":75083,"journal":{"name":"The journal of machine learning for biomedical imaging","volume":"148 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77869875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：