STAMP: Multi-pattern Attention-aware Multiple Instance Learning for STAS Diagnosis in Multi-center Histopathology Images

TL;DR

This paper introduces STAMP, a novel deep learning framework utilizing multi-pattern attention and multiple instance learning to improve the accuracy and efficiency of STAS diagnosis in multi-center lung adenocarcinoma histopathology images.

Contribution

The study presents a multi-pattern attention-aware multiple instance learning model with a dual-branch architecture and transformer-based encoding for better STAS detection across diverse datasets.

Findings

STAMP achieved AUCs of 0.8058, 0.8017, and 0.7928 on three datasets.

The model outperformed existing methods in STAS diagnosis accuracy.

Transformer-based instance encoding enhanced feature discriminability.

Abstract

Spread through air spaces (STAS) constitutes a novel invasive pattern in lung adenocarcinoma (LUAD), associated with tumor recurrence and diminished survival rates. However, large-scale STAS diagnosis in LUAD remains a labor-intensive endeavor, compounded by the propensity for oversight and misdiagnosis due to its distinctive pathological characteristics and morphological features. Consequently, there is a pressing clinical imperative to leverage deep learning models for STAS diagnosis. This study initially assembled histopathological images from STAS patients at the Second Xiangya Hospital and the Third Xiangya Hospital of Central South University, alongside the TCGA-LUAD cohort. Three senior pathologists conducted cross-verification annotations to construct the STAS-SXY, STAS-TXY, and STAS-TCGA datasets. We then propose a multi-pattern attention-aware multiple instance learning framework, named STAMP, to analyze and diagnose the presence of STAS across multi-center histopathology images. Specifically, the dual-branch architecture guides the model to learn STAS-associated pathological features from distinct semantic spaces. Transformer-based instance encoding and a multi-pattern attention aggregation modules dynamically selects regions closely associated with STAS pathology, suppressing irrelevant noise and enhancing the discriminative power of global representations. Moreover, a similarity regularization constraint prevents feature redundancy across branches, thereby improving overall diagnostic accuracy. Extensive experiments demonstrated that STAMP achieved competitive diagnostic results on STAS-SXY, STAS-TXY and STAS-TCGA, with AUCs of 0.8058, 0.8017, and 0.7928, respectively, surpassing the clinical level.