Poisson Ordinal Network for Gleason Group Estimation Using Bi-Parametric MRI

TL;DR

This paper introduces a novel Poisson ordinal network (PON) that estimates prostate cancer Gleason groups directly from MRI scans, reducing the need for biopsies by modeling class dependencies and variances.

Contribution

The study proposes a new Poisson ordinal network that captures class dependencies and variances, improving MRI-based Gleason group estimation accuracy.

Findings

PON outperforms existing methods on biopsy-labelled MRI data.

The model effectively captures ordinal relationships between Gleason groups.

Contrastive learning with a memory bank enhances intra-class variance regularization.

Abstract

The Gleason groups serve as the primary histological grading system for prostate cancer, providing crucial insights into the cancer's potential for growth and metastasis. In clinical practice, pathologists determine the Gleason groups based on specimens obtained from ultrasound-guided biopsies. In this study, we investigate the feasibility of directly estimating the Gleason groups from MRI scans to reduce otherwise required biopsies. We identify two characteristics of this task, ordinality and the resulting dependent yet unknown variances between Gleason groups. In addition to the inter- / intra- observer variability in a multi-step Gleason scoring process based on the interpretation of Gleason patterns, our MR-based prediction is also subject to specimen sampling variance and, to a lesser degree, varying MR imaging protocols. To address this challenge, we propose a novel Poisson ordinal network (PON). PONs model the prediction using a Poisson distribution and leverages Poisson encoding and Poisson focal loss to capture a learnable dependency between ordinal classes (here, Gleason groups), rather than relying solely on the numerical ground-truth (e.g. Gleason Groups 1-5 or Gleason Scores 6-10). To improve this modelling efficacy, PONs also employ contrastive learning with a memory bank to regularise intra-class variance, decoupling the memory requirement of contrast learning from the batch size. Experimental results based on the images labelled by saturation biopsies from 265 prior-biopsy-blind patients, across two tasks demonstrate the superiority and effectiveness of our proposed method.