Semantic Segmentation for Pixel-Wise Visualization of Breast Cancer in Deep Ultraviolet-Excited Fluorescence Images
Open Access
- 25 April 2026
- journal article
- Published by Japan Society of Histochemistry & Cytochemistry in ACTA HISTOCHEMICA ET CYTOCHEMICA
- Vol. 59 (2) , 101-109
Abstract
Microscopy with ultraviolet surface excitation (MUSE) enables rapid fluorescence imaging of tissue surfaces, but MUSE images differ markedly from conventional hematoxylin and eosin images, making cancer delineation challenging for routine pathological practice. We investigated the feasibility of deep learning-based semantic segmentation, which assigns a class label to each pixel, for pixel-wise breast cancer detection in MUSE images. Fresh breast tissues from 30 mastectomy patients with breast cancer were stained with terbium and Hoechst and imaged by MUSE. A total of 150 cancerous images (five per case) were manually annotated into cancerous and non-cancerous classes, and 300 non-cancerous images (ten per case) were collected. Models were trained and evaluated using five-fold nested cross-validation, comparing a cancer-only (CO) model trained solely on cancerous images with a cancer plus non-cancer (CN) model trained on both cancerous and non-cancerous images. The CO model achieved a higher Dice score than the CN model (CO, 0.7478; CN, 0.7343). Sliding window-based majority voting post-processing reduced scattered false-positive areas and improved Dice scores (CO, 0.7984; CN, 0.7849). These results support the feasibility of deep learning-based semantic segmentation for visualizing breast cancer regions and provide a basis for future quantitative applications using MUSE images.Keywords
This publication has 30 references indexed in Scilit:
- Frequency and survival of delayed breast cancer diagnosis in women participating at screening mammography in the Netherlands: a population-based studyThe Lancet Regional Health - Europe, 2025
- All Sizes Matter: Improving Volumetric Brain Segmentation on Small Lesions2024
- Next-Generation Morphometry for pathomics-data mining in histopathologyNature Communications, 2023
- Slide OverThe American Journal of Pathology, 2022
- Epithelium segmentation using deep learning in H&E-stained prostate specimens with immunohistochemistry as reference standardScientific Reports, 2019
- Rapid histopathological imaging of skin and breast cancer surgical specimens using immersion microscopy with ultraviolet surface excitationScientific Reports, 2018
- Machine Learning Methods for Histopathological Image AnalysisComputational and Structural Biotechnology Journal, 2018
- Microscopy with ultraviolet surface excitation for rapid slide-free histologyNature Biomedical Engineering, 2017
- A survey on deep learning in medical image analysisMedical Image Analysis, 2017
- Focally positive margins in breast conserving surgery: Predictors, residual disease, and local recurrenceEuropean Journal of Surgical Oncology, 2017