What question did this study set out to answer?

This work aims to improve the retrieval of breast histopathology images using advanced graph neural network models.

April 29, 2026

Leveraging Medical Foundation Model Features in Graph Neural Network‐Based Retrieval of Breast Histopathology Images

NSNematollah SaeidiUniversity of Isfahan HKHossein KarshenasUniversity of Isfahan BSBijan ShoushtarianUniversity of Isfahan

Key Points

This work aims to improve the retrieval of breast histopathology images using advanced graph neural network models.
Leveraged features from foundation models in a novel graph autoencoder model.
Evaluated on two publicly available histology image datasets (BreakHis and BACH).
Compared performance with pre-trained convolutional neural networks.
Achieved average mAP/mMV scores of 96.7%/91.5% for BreakHis and 97.6%/94.2% for BACH.
Models using foundation model features surpassed those with pre-trained convolutional networks by up to 15.5%.

Abstract

ABSTRACT Breast cancer is the most common cancer type in women worldwide. Early detection and appropriate treatment can significantly reduce its impact. While histopathology examinations play a vital role in rapid and accurate diagnosis, they often require experienced medical experts for proper recognition and cancer grading. Automated image retrieval systems have the potential to assist pathologists in identifying cancerous tissues, thereby accelerating the diagnostic process. Nevertheless, proposing an accurate image retrieval model is challenging due to considerable variability among the tissue and cell patterns in histological images. In this work, we leverage the features from foundation models in a novel attention‐based adversarially regularized variational graph autoencoder model for breast histological image retrieval. Our results confirm the superior performance of models trained with foundation model features compared to those using pre‐trained convolutional neural networks (up to 7.7% and 15.5% for mAP and mMV, respectively), with the pre‐trained general‐purpose self‐supervised model for computational pathology (UNI) delivering the best overall performance. Evaluated on two publicly available histology image datasets of breast cancer, our top‐performing model, trained with UNI features, achieved average mAP/mMV scores of 96.7%/91.5% and 97.6%/94.2% for the BreakHis and BACH datasets, respectively. Our proposed retrieval model has the potential to be used in clinical settings to enhance diagnostic performance and ultimately benefit patients.

Ask AI

Helpful

Bookmark

View Full Paper