This magnetic resonance imaging multimodal fusion study aims to automate the staging of endometrial cancer using deep learning and to compare the diagnostic performance of deep learning with that of radiologists in the staging of endometrial cancer. This study retrospectively investigated 122 patients with pathologically confirmed early EC from January 1, 2025 to December 31, 2021. Of these patients, 68 were in the International Federation of Gynecology and Obstetrics (FIGO) stage IA, and 54 were in FIGO stage IB. Based on the Swin transformer model and its proprietary SW-MSA (shift window multiple self-coherence) module, magnetic resonance imaging (MRI) images in each of the three planes (sagittal, coronal, and transverse) are cropped, enhanced, and classified, and fusion experiments in the three planes are performed simultaneously. Selecting one plane for the experiment, the accuracy of IA and IB classification was 0.988 in the sagittal, 0.96 in the coronal, and 0.94 in the transverse position, and classification accuracy after the fusion of three planes reached 1. Finally, the automatic classification method based on the Swin transformer has an accuracy of 1, a recall of 1, and a specificity of 1 for early EC classification. In this study, the multimodal fusion approach accurately classified early EC. It was comparable to what a radiologist would perform and simpler and more precise than previous methods that required segmenting followed by staging.
Endometrial cancer (EC) is the most common gynecologic malignancy, with a steadily increasing incidence worldwide. Abnormal vaginal bleeding, a hallmark symptom, enables early diagnosis, which is critical for improving clinical outcomes. Pelvic magnetic resonance imaging (MRI) serves as the primary imaging modality for EC evaluation, offering detailed visualization of endometrial and myometrial invasion.This study proposes DCS-Net, a multi-task deep learning framework for the automated detection and staging of early-stage EC in MRI. The framework incorporates an advanced object detection module to accurately localize and crop the uterine region, followed by a convolutional neural network for staging classification.Experimental results show that the detection module achieves high localization performance, and the classification network reaches an accuracy of 90.8%. The region-focused approach improves staging accuracy by 5% compared to direct classification using unprocessed images.These results underscore the potential of DCS-Net to improve diagnostic efficiency and accuracy in alignment with clinical workflows. Future work will explore the integration of multi-parametric MRI data to further enhance diagnostic performance and address broader clinical needs.