Loading…
PEARC20 has ended
Welcome to PEARC20!
PEARC20’s theme is “Catch the Wave.” This year’s theme embodies the spirit of the community’s drive to stay on pace and in front of all the new waves in technology, analytics, and a globally connected and diverse workforce. We look forward to this year’s PEARC20 virtual meeting, where we can share scientific discovery and craft the future infrastructure.

The conference will be held in Pacific Time (PT) and the times listed below are in Pacific Time.

The connection information for all PEARC20 workshops, tutorials, plenaries, track presentations, BOFs, Posters, Visualization Showcase, and other affiliated events, are in the PEARC20 virtual conference platform, Brella. If you have issues joining Brella, please email pearcinfo@googlegroups.com.
Back To Schedule
Tuesday, July 28 • 12:00pm - 1:20pm
A Pilot Benchmarking Study of Deep Neural Network Performance on Low Magnification Pathology ROIs

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Deep Neural Networks (DNNs) have successfully demonstrated superior overall performance in many image classification and recognition tasks on H&E histology images. Reported studies typically utilize high quality (20x or 40x) Whole Slide Images (WSIs) in order to deliver optimal performance. However, it remains uncertain how well DNNs can perform on lower quality region-of-interest (ROI) histology images in real life scenarios. The NCI Patient Derived Models Repository (PDMR) database hosts a catalog of low magnification (4x) ROIs of tissue histology images across a total of 60 cancer models, providing an ideal test case for evaluating DNNs performance in real life scenarios. In this study, using 5 pre-trained models, we have benchmarked the NCI PDMR database ROIs on a selected set of popular DNN classifiers. Overall, on the binary carcinoma vs. sarcoma classification test, we have reached 89.57% accuracy on 4x ROIs using our downsizing models and 84.18% accuracy on 4x ROIs using our patch-based models. On the multi-class carcinoma classification test, we have reached 72.06% top-2 accuracy on 4x ROIs using our downsizing models and 78.07% top-2 accuracy on 4x ROIs using our patch-based models. With such accuracies, we can utilize our DNNs to perform crucial tele-pathological tasks in underdeveloped countries and rural areas, enabling scientists to utilize histology images acquired from mobile devices for rapid screening in remote areas.


Tuesday July 28, 2020 12:00pm - 1:20pm PDT
Brella