Single-Stage vs. Multi-Stage Machine Learning Algorithms for Prostate Segmentation in Magnetic Resonance Images

Anurag Garikipati, Rajesh Venkataraman

Keywords: Machine Learning, Prostate Segmentation, Magnetic Resonance Imaging

TL;DR: Comparing multi-stage vs. single-stage machine learning algorithms for prostate segmentation in MRI

Track: short paper

Paper Type: methodological development

Abstract: Fusion of magnetic resonance images (MRI) with ultrasound has led to major improvements in precision diagnostics for prostate cancer. A key step in the fusion process is segmentation of the prostate in MRI and machine learning (ML) has proven to be a valuable tool for segmentation. In this paper, we compare two ML workflows for prostate segmentation; a single-stage and multi-stage ML algorithm to address the challenges of prostate segmentation.