MRI-Based Habitat Radiomics for Differentiating Early-Stage Endometrial Carcinoma from Submucous Leiomyoma: A...

Area of Science:

• Radiology
• Oncology
• Medical Imaging Analysis

Background:

• Distinguishing early-stage endometrial carcinoma (EC) from submucous leiomyoma (SML) preoperatively is clinically important.
• Accurate differentiation impacts treatment decisions and patient outcomes.

Purpose of the Study:

• To evaluate the effectiveness of magnetic resonance imaging (MRI)-based habitat radiomics for preoperative differentiation between EC and SML.
• To determine if habitat radiomics offers improved diagnostic performance compared to whole-tumor analysis.

Main Methods:

• A retrospective study included 231 patients with uterine lesions (97 EC, 134 SML) who underwent MRI.
• MRI scans were segmented into habitats using k-means clustering on T1WI, T2WI, and ADC maps.
• Radiomic features were extracted from whole tumors and habitats, with selection via Pearson correlation and LASSO regression, followed by logistic regression modeling.

Main Results:

• The habitat_3 model achieved the highest AUC (0.907) in the training cohort.
• In external validation, the habitat_3 model demonstrated superior performance (AUC=0.881) compared to the whole-tumor model (AUC=0.751).
• The habitat_3 model showed incrementally improved predictive performance in the external validation cohort.

Conclusions:

• MRI-based habitat radiomics provides valuable information for preoperative differentiation of early-stage EC and SML.
• Habitat_3 radiomics shows promise for enhancing diagnostic accuracy in distinguishing these conditions.