Recently, the research collaboration between the Bioinformatics & Intelligent Medicine team and Tianjin Medical University Cancer Hospital has resulted in the publication titled "PhiHER2: Phenotype-informed weakly supervised model for HER2 status prediction from pathological images" in the journal Bioinformatics/ISMB2024.

HER2 status identification enables physicians to assess the prognosis risk and determine the treatment schedule for patients. In clinical practice, pathological slides serve as the gold standard, offering morphological information on cellular structure and tumoral regions. Computational analysis of pathological images has the potential to discover morphological patterns associated with HER2 molecular targets and achieve precise status prediction. However, pathological images are typically equipped with high-resolution attributes, and HER2 expression in breast cancer images often manifests the intra-tumoral heterogeneity. In this work, we present a phenotype-informed weakly-supervised multiple instance learning architecture (PhiHER2) for the prediction of the HER2 status from pathological images of breast cancer. Specifically, a hierarchical prototype clustering module is designed to identify representative phenotypes across whole slide images. These phenotype embeddings are then integrated into a cross-attention module, enhancing feature interaction and aggregation on instances. This yields a phenotype-based feature space that leverages the intra-tumoral morphological heterogeneity for HER2 status prediction.

Extensive results demonstrate that PhiHER2 captures a better WSI-level representation by the typical phenotype guidance and significantly outperforms existing methods on real-world datasets. Additionally, interpretability analyses of both phenotypes and WSIs provide explicit insights into the heterogeneity of morphological patterns associated with molecular HER2 status.

doi：https://doi.org/10.1093/bioinformatics/btae236