H4 - Optimizing Operating Points for High Performance Lesion Detection and Segmentation Using Lesion Size Reweighting

Brennan Nichyporuk, Justin Szeto, Douglas Arnold, Tal Arbel

Show abstract - Show schedule - PDF - Reviews

There are many clinical contexts which require accurate detection and segmentation of all focal pathologies (e.g. lesions, tumours) in patient images. In cases where there are a mix of small and large lesions, standard binary cross entropy loss will result in better segmentation of large lesions at the expense of missing small ones. Adjusting the operating point to accurately detect all lesions generally leads to oversegmentation of large lesions. In this work, we propose a novel reweighing strategy to eliminate this performance gap, increasing small pathology detection performance while maintaining segmentation accuracy. We show that our reweighing strategy vastly outperforms competing strategies based on experiments on a large scale, multi-scanner, multi-center dataset of Multiple Sclerosis patient images.
Hide abstract

Thursday 8th July
H4-12 (short): Application: Radiology - 16:45 - 17:30 (UTC+2)
Hide schedule

Can't display slides, your browser doesn't support embedding PDFs.

Download slides