EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explainable CNN with spatial attention for oral cancer patch selection using a custom scoring metric

Subhija Eranjoli Nalupurakkal, Reju Vaninirappuputhenpurayil Gopalan

Biomedical Engineering / Biomedizinische Technik · 05.06.2026

DOI: 10.1515/bmt-2025-0308 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomedical Engineering / Biomedizinische Technik, online erschienen am 05.06.2026. Verfasst von Subhija Eranjoli Nalupurakkal und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Instrumentation, Cochin University of Science and Technology, Cochin, Kerala, India. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Rashtriya Uchchatar Shiksha Abhiyan (RUSA) 2.0,.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

This study aims to propose an algorithm to select feature-rich patches from histopathology images, eliminating irrelevant regions, to support the computer-aided diagnosis of oral cancer.

Methods

To identify interpretable and diverse feature-rich patches, Haralick and local binary pattern (LBP) features are extracted and cosine similarity is calculated between patches. A trained convolutional neural network (CNN) model with spatial attention, along with a SHapley Additive exPlanation (SHAP) explainer, is used to estimate each patch’s contribution using SHAP values. These values, combined with cosine similarity, guide the selection of four patches that significantly influence the diagnosis while ensuring diversity. Features from the selected patches are extracted using a ResNet-50 model, concatenated, and normalized using Z-score normalization. The resulting feature vector is used to classify images as normal or cancerous.

Results

The approach was tested on a public dataset of 1,224 oral histopathology images from 230 patients, covering 100× and 400× magnifications. It achieved 97.53 % accuracy, 97.68 % precision, and 97.36 % sensitivity using a voting classifier, and 97.93 %, 99.43 %, and 97.76 % respectively with a stacking classifier, outperforming existing cancer detection methods.

Conclusions

The proposed algorithm effectively classifies oral biopsy images with minimal training, enhancing diagnostic reliability and offering valuable support for pathologists.

Schlagwörter

Stichwörter der Autor:innen

SHAP valuecancer detectioncomputer-aided diagnosishistopathologyoral squamous cell carcinomapatch selection

Autor:innen und Institutionen

  • Department of Instrumentation, Cochin University of Science and Technology, Cochin, Kerala, India.

Förderung

  • Rashtriya Uchchatar Shiksha Abhiyan (RUSA) 2.0,

Zitieren

APA

Eranjoli Nalupurakkal, S., & Vaninirappuputhenpurayil Gopalan, R. (2026). Explainable CNN with spatial attention for oral cancer patch selection using a custom scoring metric. Biomedical Engineering / Biomedizinische Technik. https://doi.org/10.1515/bmt-2025-0308

Vancouver

Eranjoli Nalupurakkal S, Vaninirappuputhenpurayil Gopalan R. Explainable CNN with spatial attention for oral cancer patch selection using a custom scoring metric. Biomedical Engineering / Biomedizinische Technik. 2026. doi:10.1515/bmt-2025-0308

BibTeX
@article{EranjoliNalupurakkal2026,
  author = {Eranjoli Nalupurakkal, Subhija and Vaninirappuputhenpurayil Gopalan, Reju},
  title = {Explainable CNN with spatial attention for oral cancer patch selection using a custom scoring metric},
  journal = {Biomedical Engineering / Biomedizinische Technik},
  year = {2026},
  doi = {10.1515/bmt-2025-0308}
}
Kostenlos starten →