Biochemical Prostate Cancer Recurrence Prediction: Thinking Fast & Slow

TL;DR

This paper introduces a novel two-stage multiple instance learning approach, inspired by the 'thinking fast and slow' concept, to predict biochemical recurrence time in prostate cancer from whole slide images, achieving promising validation results.

Contribution

It proposes a new two-stage MIL method leveraging different resolution patches for improved prostate cancer recurrence prediction.

Findings

Achieved a mean C-index of 0.733 on internal validation.

Achieved a C-index of 0.603 on LEOPARD challenge validation.

Attention visualization highlights relevant tissue areas for prediction.

Abstract

Time to biochemical recurrence in prostate cancer is essential for prognostic monitoring of the progression of patients after prostatectomy, which assesses the efficacy of the surgery. In this work, we proposed to leverage multiple instance learning through a two-stage ``thinking fast \& slow'' strategy for the time to recurrence (TTR) prediction. The first (``thinking fast'') stage finds the most relevant WSI area for biochemical recurrence and the second (``thinking slow'') stage leverages higher resolution patches to predict TTR. Our approach reveals a mean C-index ($Ci$) of 0.733 ($\theta=0.059$) on our internal validation and $Ci=0.603$ on the LEOPARD challenge validation set. Post hoc attention visualization shows that the most attentive area contributes to the TTR prediction.