Saved in:
Bibliographic Details
Main Authors: Tomy, Chris, Vali, Mo, Pertzborn, David, Alamatouri, Tammam, Mühlig, Anna, Guntinas-Lichius, Orlando, Xylander, Anna, Fantuzzi, Eric Michele, Negro, Matteo, Crisafi, Francesco, Lio, Pietro, Azevedo, Tiago
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.18119
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908843516100608
author Tomy, Chris
Vali, Mo
Pertzborn, David
Alamatouri, Tammam
Mühlig, Anna
Guntinas-Lichius, Orlando
Xylander, Anna
Fantuzzi, Eric Michele
Negro, Matteo
Crisafi, Francesco
Lio, Pietro
Azevedo, Tiago
author_facet Tomy, Chris
Vali, Mo
Pertzborn, David
Alamatouri, Tammam
Mühlig, Anna
Guntinas-Lichius, Orlando
Xylander, Anna
Fantuzzi, Eric Michele
Negro, Matteo
Crisafi, Francesco
Lio, Pietro
Azevedo, Tiago
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2602_18119
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RamanSeg: Interpretability-driven Deep Learning on Raman Spectra for Cancer Diagnosis
Tomy, Chris
Vali, Mo
Pertzborn, David
Alamatouri, Tammam
Mühlig, Anna
Guntinas-Lichius, Orlando
Xylander, Anna
Fantuzzi, Eric Michele
Negro, Matteo
Crisafi, Francesco
Lio, Pietro
Azevedo, Tiago
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
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.
title RamanSeg: Interpretability-driven Deep Learning on Raman Spectra for Cancer Diagnosis
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2602.18119