Ontology highlight
ABSTRACT:
SUBMITTER: Kanakasabapathy MK
PROVIDER: S-EPMC8943917 | biostudies-literature | 2021 Jun
REPOSITORIES: biostudies-literature
Kanakasabapathy Manoj Kumar MK Thirumalaraju Prudhvi P Kandula Hemanth H Doshi Fenil F Sivakumar Anjali Devi AD Kartik Deeksha D Gupta Raghav R Pooniwala Rohan R Branda John A JA Tsibris Athe M AM Kuritzkes Daniel R DR Petrozza John C JC Bormann Charles L CL Shafiee Hadi H
Nature biomedical engineering 20210610 6
In machine learning for image-based medical diagnostics, supervised convolutional neural networks are typically trained with large and expertly annotated datasets obtained using high-resolution imaging systems. Moreover, the network's performance can degrade substantially when applied to a dataset with a different distribution. Here, we show that adversarial learning can be used to develop high-performing networks trained on unannotated medical images of varying image quality. Specifically, we u ...[more]