Hershel Mehta

TL;DR: An algorithm that can detect pneumonia from chest X-rays at a level exceeding practicing radiologists is developed, and it is found that CheXNet exceeds average radiologist performance on the F1 metric.

TL;DR: CheXNeXt, a convolutional neural network to concurrently detect the presence of 14 different pathologies, including pneumonia, pleural effusion, pulmonary masses, and nodules in frontal-view chest radiographs, achieved radiologist-level performance on 11 pathologies and did not achieve radiologists' level performance on 3 pathologies.

TL;DR: Optogenetic and brain imaging approaches reveal a causal brainwide dynamical mechanism for the hedonic-anhedonic transition and test the hypothesis that elevated medial prefrontal cortex (mPFC) excitability exerts suppressive control over the interactions between two distant subcortical regions: the dopaminergic midbrain and the striatum.

TL;DR: MURA, a large dataset of musculoskeletal radiographs containing 40,561 images from 14,863 studies, where each study is manually labeled by radiologists as either normal or abnormal is introduced, and a 169-layer DenseNet baseline model is trained to detect and localize abnormalities.

TL;DR: A 169-layer densely connected convolutional network is trained to detect and localize abnormalities in MURA, a large dataset of musculoskeletal radiographs containing 40,895 images from 14,982 studies, and it is found that the model achieves performance comparable to that of radiologists.