A Prospective Validation Study of Radiomics in the Differential Diagnosis of Uterine Leiomyoma and Uterine Sarcoma

Uterine fibroids are the most common benign gynecological tumors among women of reproductive age in China, with a prevalence of 20-30% among women over 30 years old and a trend toward younger onset. Despite advances in minimally invasive techniques and pharmacological therapies during the "12th Five-Year Plan," the incidence of uterine fibroids continues to rise due to rapid socioeconomic development, environmental changes, lifestyle shifts, and delayed childbearing. As a result, uterine fibroids have become a major public health concern. Understanding the mechanisms underlying the onset, recurrence, and malignant transformation of fibroids, developing fertility-preserving individualized treatment strategies, and identifying high-risk populations remain key challenges in reproductive and women's health research.

To address these challenges, our multicenter collaborative group, led by Tongji Hospital and supported by the National Clinical Research Center for Obstetrics and Gynecology, has established a large-scale systematic database integrating clinical, imaging, pathological, laboratory, and molecular data from multiple tertiary hospitals. Based on multicenter clinical big data collected from January 2012 to January 2025, we have developed a multimodal early-warning model for the malignant transformation of uterine fibroids. This model, primarily incorporating T2WI and DWI features and optimized using a support vector machine (SVM) algorithm, demonstrated high sensitivity and specificity in retrospective analysis and internal validation, suggesting promising potential for identifying high-risk individuals.

However, retrospective designs inherently limit the assessment of the model's real-world clinical applicability, generalizability, and stability. Therefore, beginning in January 2025, we plan to conduct a prospective validation study in consecutively enrolled patients to evaluate the model's diagnostic performance in routine clinical practice, its feasibility in real-world diagnostic workflows, and its potential value for early screening, follow-up management, and individualized treatment of high-risk populations. This study is expected to provide robust evidence to improve early detection, reduce malignant transformation risk, and ultimately enhance clinical outcomes and public health impact.