RamanSeg: Interpretability-driven Deep Learning on Raman Spectra for Cancer Diagnosis

TL;DR

RamanSeg introduces an interpretable deep learning model for cancer diagnosis using Raman spectra, achieving high segmentation accuracy and providing a transparent alternative to traditional black-box methods.

Contribution

The paper presents RamanSeg, a novel prototype-based, interpretable deep learning architecture for Raman spectra analysis in cancer diagnosis, with variants balancing interpretability and performance.

Findings

RamanSeg achieved a mean foreground Dice score of 67.3%.

The model outperformed a U-Net baseline in interpretability and accuracy.

Proposed methods enable transparent and effective cancer tissue segmentation.

Abstract

Histopathology, the current gold standard for cancer diagnosis, involves the manual examination of tissue samples after chemical staining, a time-consuming process requiring expert analysis. Raman spectroscopy is an alternative, stain-free method of extracting information from samples. Using nnU-Net, we trained a segmentation model on a novel dataset of spatial Raman spectra aligned with tumour annotations, achieving a mean foreground Dice score of 80.9%, surpassing previous work. Furthermore, we propose a novel, interpretable, prototype-based architecture called RamanSeg. RamanSeg classifies pixels based on discovered regions of the training set, generating a segmentation mask. Two variants of RamanSeg allow a trade-off between interpretability and performance: one with prototype projection and another projection-free version. The projection-free RamanSeg outperformed a U-Net baseline with a mean foreground Dice score of 67.3%, offering a meaningful improvement over a black-box training approach.