J
Jay Patravali
Researcher at Oregon State University
Publications - 8
Citations - 1212
Jay Patravali is an academic researcher from Oregon State University. The author has contributed to research in topics: Deep learning & Computer science. The author has an hindex of 4, co-authored 7 publications receiving 613 citations. Previous affiliations of Jay Patravali include Carnegie Mellon University & Indian Institute of Science.
Papers
More filters
Journal ArticleDOI
Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved?
Olivier Bernard,Alain Lalande,Clement Zotti,Frederick Cervenansky,Xin Yang,Pheng-Ann Heng,Irem Cetin,Karim Lekadir,Oscar Camara,Miguel Ángel González Ballester,Gerard Sanroma,Sandy Napel,Steffen E. Petersen,Georgios Tziritas,Elias Grinias,Mahendra Khened,Varghese Alex Kollerathu,Ganapathy Krishnamurthi,Marc-Michel Rohé,Xavier Pennec,Maxime Sermesant,Fabian Isensee,Paul F. Jäger,Klaus H. Maier-Hein,Peter M. Full,Ivo Wolf,Sandy Engelhardt,Christian F. Baumgartner,Lisa M. Koch,Jelmer M. Wolterink,Ivana Išgum,Yeonggul Jang,Yoonmi Hong,Jay Patravali,Shubham Jain,Olivier Humbert,Pierre-Marc Jodoin +36 more
TL;DR: How far state-of-the-art deep learning methods can go at assessing CMRI, i.e., segmenting the myocardium and the two ventricles as well as classifying pathologies is measured, to open the door to highly accurate and fully automatic analysis of cardiac CMRI.
Book ChapterDOI
2D-3D Fully Convolutional Neural Networks for Cardiac MR Segmentation
TL;DR: In this article, a 2D and 3D segmentation pipeline for fully automated cardiac MR image segmentation using deep convolutional neural networks (CNNs) was developed and trained end-to-end from scratch using the ACD Challenge 2017 dataset comprising of 100 studies.
Posted Content
2D-3D Fully Convolutional Neural Networks for Cardiac MR Segmentation
TL;DR: In this article, a 2D and 3D segmentation pipeline for fully automated cardiac MR image segmentation using deep convolutional neural networks (CNNs) was developed and trained end-to-end from scratch using the ACD Challenge 2017 dataset comprising of 100 studies.
Proceedings ArticleDOI
Unsupervised Cross-Dataset Adaptation via Probabilistic Amodal 3D Human Pose Completion
TL;DR: A novel probabilistic amodal pose completion framework is devised to address dataset bias and helps achieve state-of-the art performance on unsupervised cross-dataset pose estimation, with a significant improvement in partially-visible unconstrained scenarios.
Proceedings ArticleDOI
Learning Intuitive Physics by Explaining Surprise
TL;DR: An instantiation of the Surprise and Explain framework is developed and its potential in the IntPhys Challenge is demonstrated by placing 2nd at the time of this paper’s submission.