Developing Efficient Personalized Treatment Platforms for Cervix Cancer Using Advanced Radiation, Imaging, Molecular Biology, and Machine Learning

What this Trial Is About

Researchers are studying cervix cancer by using patient data from completed and ongoing clinical trials to develop and validate machine learning tools. The study aims to create knowledge-based systems that improve treatment delivery and help identify patients who may benefit from more intensive treatments. This observational study integrates radiation treatment information, quantitative imaging, and patient outcomes to advance personalized treatment approaches. The research focuses on three main goals. First, automatic delineation of complex tumor targets on MRI and CT scans is performed for various volumes and organs related to cervix cancer. Second, machine learning models are developed to predict normal tissue toxicity using texture analysis and radiation dose maps. Third, the study integrates imaging, staging, pathology, and molecular biology data to predict treatment response and identify high-risk patient groups. Patient data sets include imaging and biomarker information collected during chemoradiation and brachytherapy. Participants existing clinical trial data and imaging are used retrospectively, so no new treatment is given. The research team will generate software tools for automated tumor delineation and develop normal tissue complication plots over three years. They will also identify high-risk patients who might need intensified therapy based on integrated imaging and molecular features. The study includes patients aged 18 to 90 who have completed or are undergoing radiation therapy for cervix cancer. Outcomes measured include software development, toxicity prediction model validation, and risk group identification.

Development of Clinically High Efficient Platforms for Individualised Treatment of Cervix Cancer