What question did this study set out to answer?

The aim is to develop and evaluate CNN models for detecting basement membrane breaches in OSCC using a unique dataset.

April 23, 2026Journal of Oral Biology and Craniofacial Research0 citationsOpen Access

Automated detection of basement-membrane breaches in oral squamous cell carcinoma : A comparative study of convolutional neural network architectures

SAS Samyuktha Aarthi MYMonal Yuwanati PYPradeep Kumar Yadalam

Key Points

The aim is to develop and evaluate CNN models for detecting basement membrane breaches in OSCC using a unique dataset.
Created a dataset of 66 H&E stained images with expert annotations for 'break' and 'normal' categories.
Evaluated five CNN architectures: two custom models and three pre-trained models under the same conditions.
Measured performance using accuracy and AUC metrics on test sets.
ResNet-18 achieved the highest test accuracy of 86.67% and AUC of 0.87.
MobileNetV2 followed with 80% accuracy and AUC of 0.83.
Custom models performed lower, around 73%, indicating transfer learning models are more effective.

Abstract

Oral squamous cell carcinoma (OSCC) remains a leading cause of cancer-related morbidity and mortality worldwide. The ability to detect early histopathological evidence of invasion—particularly basement membrane (BM) discontinuity at the epithelium–connective tissue interface—is crucial for timely intervention; yet, this subtle feature is often overlooked in automated image analysis research. This study aimed to develop and benchmark convolutional neural network (CNN) models for the automated detection of BM breaches in OSCC, introducing the first publicly available, expert-annotated dataset dedicated to this diagnostic hallmark. We assembled 66 high-resolution hematoxylin–eosin (H&E) stained histology images, evenly split into “break” (BM discontinuity with tumour invasion) and “normal” categories. Expert oral pathologists annotated the epithelial–connective tissue interface, producing region-of-interest metadata for each image. The dataset was divided into training, validation, and test sets. Five CNN architectures were evaluated under same training conditions: two custom models (SimpleCNN and DeepCNN) and three transfer learning models (ResNet-18, MobileNetV2, EfficientNet-B0). Performance was measured by test accuracy and AUC. ResNet-18 was best performing CNN with 86.67% accuracy and 0.87 AUC, effectively discriminate the lesion with break from normal mucosa. MobileNetV2 followed with 80% accuracy (AUC 0.83). SimpleCNN, DeepCNN, and EfficientNet-B0 achieved about 73%. Overall, transfer learning models outperformed custom architectures, highlighting the benefits of pre-trained features in this histopathological task. This study highlights AI's role in detecting epithelial-basement membrane breaks and a key microscopic feature for diagnosing invasion, demonstrating how deep learning enables precise, consistent, objective evaluations.

AIに質問

Bookmark

View Full Paper