A Dual-Sequence MRI-Based Radiomics Model for Predicting High-Intensity Focused Ultrasound Ablation Efficacy in...

Area of Science:

• Medical Imaging
• Oncology
• Radiology

Background:

• Adenomyosis treatment often involves high-intensity focused ultrasound (HIFU) ablation.
• Predicting the therapeutic efficacy of HIFU ablation is crucial for patient management.
• Dual-sequence MRI radiomics offers potential for non-invasive outcome prediction.

Purpose of the Study:

• To develop and validate a predictive model for HIFU ablation efficacy in adenomyosis.
• To utilize dual-sequence MRI radiomics combined with clinical data for enhanced prediction.
• To assess the model's performance in predicting varying levels of ablation success.

Main Methods:

• Retrospective analysis of 114 adenomyosis patients undergoing HIFU ablation.
• Development of predictive models using logistic regression and radiomic features from dual-sequence MRI.
• Model evaluation on a test set using ROC curves, Delong test, NRI, and IDI.

Main Results:

• The combined dual-sequence MRI radiomics model achieved an AUC of 0.84 for predicting ablation outcomes.
• The model demonstrated superior net benefit compared to single-sequence models.
• Accurate prediction of ablation success rates exceeding 50% was observed.

Conclusions:

• Dual-sequence MRI radiomics provides a promising tool for predicting HIFU ablation efficacy in adenomyosis.
• The developed model can assist clinicians in anticipating treatment outcomes.
• This approach may optimize patient selection and treatment planning for HIFU procedures.