What question did this study set out to answer?

Develop a deep learning framework to accurately predict WHO Type I and Type II subtypes of epithelial ovarian cancer (EOC).

February 26, 2026BMC Medical Imaging0 citationsOpen Access

A dual-center study: multimodal fusion-based deep learning approach for pathological subtype prediction of type I and type II ovarian cancer

THTianle Hong WHWeicheng Huang WLWenqing Lu

Key Points

Develop a deep learning framework to accurately predict WHO Type I and Type II subtypes of epithelial ovarian cancer (EOC).
Developed a multimodal deep learning model combining CT, ultrasound, and clinical data.
Performed retrospective analysis on 240 EOC patients with postoperative histopathological subtyping.
Conducted external validation with 17 additional EOC patients.
Used ResNet-50 with Cross-attention mechanisms for data fusion.
Analyzed contributions of imaging features through SHAP analysis.
The USECNC + CL model achieved an AUC of 0.87 during cross-validation.
Baseline models showed varying AUCs: EC (0.81), NC (0.69), US (0.77), Clinical data (0.79).
External validation for the USECNC + CL model yielded AUC of 0.850 and accuracy of 0.824.
SHAP analysis ranked imaging features, highlighting EC as most important, followed by US and NC.
CA-125 levels indicated higher predictive value compared to imaging features.

Abstract

Accurate prediction of WHO Type I and Type II pathological classifications in epithelial ovarian cancer (EOC) patients is critical for developing effective personalized treatment strategies. This study aims to develop a multimodal deep learning framework integrating enhanced CT (EC), non-contrast CT (NC), transvaginal ultrasound (US), and clinical data to predict EOC subtypes. Furthermore, we investigate the differential contributions of distinct modalities to model predictions through interpretable analysis. A retrospective analysis was conducted on 240 EOC patients who underwent postoperative histopathological subtyping alongside EC, NC and US. A total of 17 EOC patients from other centers were enrolled for external validation. The multimodal imaging data fusion deep learning model USECNC, integrating ResNet-50 with Cross-attention mechanisms, was designed to synergistically fuse multimodal image features. The integrated model USECNC + CL (USECNC+Clinical Data) combines clinical data and imaging features at the decision layer for final prediction. The USECNC + CL model demonstrated the best performance in predicting EOC pathological subtypes, achieving an AUC of 0.87 (5-fold cross-validation range: 0.82–0.92). Baseline models based on EC, NC, US, and clinical data achieved AUC of 0.81, 0.69, 0.77 and 0.79. The optimal radiomics model achieved an AUC of 0.72. The USECNC + CL model also achieved good predictive performance in the external validation set (AUC = 0.850, ACC = 0.824). In the SHAP analysis, the contribution of imaging features is ranked from highest to lowest as follows: EC, US, NC. Meanwhile, compared to imaging fusion features, CA125 demonstrates a superior contribution. The integration of preoperative multimodal data into a deep learning model enables accurate prediction of pathological subtypes in EOC patients. Interpretability analyses confirmed the critical diagnostic contributions of EC vascular features and CA-125 levels. Accurate preoperative type I/II prediction can effectively assist in the formulation of chemotherapy regimens and surgical plans for patients.

Demander à l'IA

Bookmark

View Full Paper