GUIDE-US: Grade-Informed Unpaired Distillation of Encoder Knowledge from Histopathology to Micro-UltraSound

TL;DR

GUIDE-US introduces a novel unpaired knowledge distillation method that enables micro-ultrasound models to better detect significant prostate cancer by leveraging histopathology data without requiring paired images or registration.

Contribution

This work presents a new unpaired distillation approach that improves micro-US prostate cancer grading by incorporating histopathology knowledge without needing paired data or registration.

Findings

Increases sensitivity to clinically significant prostate cancer by 3.5% at 60% specificity.

Improves overall sensitivity for prostate cancer detection at 60% specificity.

Advances clinical feasibility of non-invasive prostate cancer risk stratification.

Abstract

Purpose: Non-invasive grading of prostate cancer (PCa) from micro-ultrasound (micro-US) could expedite triage and guide biopsies toward the most aggressive regions, yet current models struggle to infer tissue micro-structure at coarse imaging resolutions. Methods: We introduce an unpaired histopathology knowledge-distillation strategy that trains a micro-US encoder to emulate the embedding distribution of a pretrained histopathology foundation model, conditioned on International Society of Urological Pathology (ISUP) grades. Training requires no patient-level pairing or image registration, and histopathology inputs are not used at inference. Results: Compared to the current state of the art, our approach increases sensitivity to clinically significant PCa (csPCa) at 60% specificity by 3.5% and improves overall sensitivity at 60% specificity by 1.2%. Conclusion: By enabling earlier and more dependable cancer risk stratification solely from imaging, our method advances clinical feasibility. Source code will be publicly released upon publication.