Jay B. Patel
Massachusetts Institute of Technology
4 Papers
24 Citations
Jay B. Patel is an academic researcher from Massachusetts Institute of Technology. The author has contributed to research in topics: Deep learning & Consistency (statistics). The author has an hindex of 4, co-authored 4 publications.
Chat about Author
Papers
Federated Learning for Breast Density Classification: A Real-World Implementation.
Holger R. Roth,Ken Chang,Praveer Singh,Nir Neumark,Wenqi Li,Vikash Gupta,Sharut Gupta,Liangqiong Qu,Alvin Ihsani,Bernardo Bizzo,Yuhong Wen,Varun Buch,Meesam Shah,Felipe Kitamura,Matheus Ribeiro Furtado de Mendonça,Vitor Lavor,Ahmed Harouni,Colin B. Compas,Jesse Tetreault,Prerna Dogra,Yan Cheng,Selnur Erdal,Richard D. White,Behrooz Hashemian,Thomas J. Schultz,Miao Zhang,Adam McCarthy,B. Min Yun,Elshaimaa Sharaf,Katharina Hoebel,Jay B. Patel,Bryan Chen,Sean Ko,Evan Leibovitz,Etta D. Pisano,Laura Coombs,Daguang Xu,Keith J. Dreyer,Ittai Dayan,Ram C. Naidu,Mona Flores,Daniel L. Rubin,Jayashree Kalpathy-Cramer +42 more
- 08 Oct 2020
TL;DR: This study investigates the use of federated learning (FL) to build medical imaging classification models in a real-world collaborative setting and shows that despite substantial differences among the datasets from all sites and without centralizing data, it can successfully train AI models in federation.
Federated Learning for Breast Density Classification: A Real-World Implementation
Holger R. Roth,Ken Chang,Praveer Singh,Nir Neumark,Wenqi Li,Vikash Gupta,Sharut Gupta,Liangqiong Qu,Alvin Ihsani,Bernardo Bizzo,Yuhong Wen,Varun Buch,Meesam Shah,Felipe Kitamura,Matheus Ribeiro Furtado de Mendonça,Vitor Lavor,Ahmed Harouni,Colin B. Compas,Jesse Tetreault,Prerna Dogra,Yan Cheng,Selnur Erdal,Richard D. White,Behrooz Hashemian,Thomas J. Schultz,Miao Zhang,Adam McCarthy,B. Min Yun,Elshaimaa Sharaf,Katharina Hoebel,Jay B. Patel,Bryan Chen,Sean Ko,Evan Leibovitz,Etta D. Pisano,Laura Coombs,Daguang Xu,Keith J. Dreyer,Ittai Dayan,Ram C. Naidu,Mona Flores,Daniel L. Rubin,Jayashree Kalpathy-Cramer +42 more
TL;DR: This paper investigated the use of federated learning (FL) to build medical imaging classification models in a real-world collaborative setting, and showed that models trained using FL perform 6.3% on average better than their counterparts trained on an institute's local data alone.
24
An exploration of uncertainty information for segmentation quality assessment
Katharina Hoebel,Vincent Andrearczyk,Andrew Beers,Jay B. Patel,Ken Chang,Adrien Depeursinge,Henning Müller,Henning Müller,Jayashree Kalpathy-Cramer +8 more
- 10 Mar 2020
TL;DR: Using uncertainty information offers ways to assess segmentation quality fully automatically without access to ground truth and models trained using weighted categorical cross-entropy offer more meaningful uncertainty information on a voxel-level.
Radiomics Repeatability Pitfalls in a Scan-Rescan MRI Study of Glioblastoma
Katharina Hoebel,Jay B. Patel,Andrew Beers,Ken Chang,Praveer Singh,James M. Brown,Marco C. Pinho,Tracy T. Batchelor,Elizabeth R. Gerstner,Bruce R. Rosen,Jayashree Kalpathy-Cramer +10 more
- 01 Jan 2021
TL;DR: Both normalization and intensity quantization have an effect on the level of repeatability and redundancy of features, emphasizing the importance of both accurate reporting of methodology in radiomics articles and understanding the limitations of choices made in pipeline design.