2025/09/27 by Jia Li, Xinghao Wang, Fanxin Zeng +8 · 1 voice
Computer Science · Medicine · Biochemistry, Genetics and Molecular Biology · #Topic Modeling #Radiomics and Machine Learning in Medical Imaging #Biomedical Text Mining and Ontologies
paper · doi:10.1016/j.imed.2025.03.004
openalex publication_date 2025/09/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/06/28
Background Colorectal cancer(CRC) presents a significant global health challenge due to its rising incidence. Early detection and precise prediction of molecular target expression, guided by key biomarkers such as CK-20, Ki-67, P-53, and microsatellite instability, are essential for personalizing treatment strategies and improving patient outcomes. However, current approaches often struggle to capture intricate semantic and relational structures within vast medical records and corpus. Methods We developed CRC-BERT-GCN, a pre-trained deep learning framework that integrates graph neural network, to fully utilize the semantic information in predicting cancer phenotypes. We retrospectively collected 6,468 CRC patients from a tertiary hospital in Beijing, from 2012 to 2022, electronic health record (EHR)and radiology reports were collected and divided into training sets (70%), validation sets (10%) and tests (20%) stratified by molecular phenotypes. The model performance is evaluated using accuracy, precision, recall, and F1-score. Ablation studies and attention analysis were performed to interpret the model’s performance and internal mechanisms. Results CRC-BERT-GCN outperformed all basic models and achieved the best performance across all evaluation metrics, with an F1-Score of 73.67%. Compared to baseline models, it improved accuracy by 4.25% and F1-Score by 4.39%, demonstrating the effectiveness of integrating GCN for biomarker prediction. Moreover, the results were interpreted using attention heatmap. Conclusion CRC-BERT-GCN may demonstrate superior performance in predicting colorectal cancer biomarkers, and outperforming baseline models, highlighting its potential for advancing personalized treatment strategies through improved molecular target prediction.