Enhancing Predictive Accuracy for Recurrence-Free Survival in Head and Neck Tumor: A Comparative Study of Weighted...

Area of Science:

• Oncology
• Medical Imaging
• Radiomics

Background:

• Predicting recurrence-free survival (RFS) in head and neck (H&N) cancer is complex due to tumor heterogeneity.
• Traditional clinical predictors have limitations in accurately forecasting RFS.

Purpose of the Study:

• To compare the prognostic efficacy of clinical predictors against advanced radiomics features for H&N cancer RFS.
• To explore weighted fusion techniques for enhancing RFS prediction accuracy.

Main Methods:

• Utilized clinical data and radiomic features from CT and PET scans.
• Employed weighted fusion algorithms for patient risk stratification.
• Evaluated predictive performance using Kaplan-Meier analysis and confidence interval tests.

Main Results:

• The weighted fusion model, with 90% emphasis on PET features, significantly outperformed individual modalities.
• Combined radiomics and clinical data through weighted fusion improved RFS prediction accuracy.
• Individual clinical and radiomics models did not achieve statistical significance for survival differentiation.

Conclusions:

• Integrating radiomic features with clinical data via weighted fusion enhances RFS prediction in H&N cancer.
• Multi-modal data and PET imaging show potential for more reliable prognostic models.
• This approach represents a step towards precision medicine in H&N cancer care.