Actively Recruiting
Developing an AI-Based Risk Prediction Model Integrating Clinical and Imaging Data for Head and Neck Cancerous Lesions
Led by The Affiliated Nanjing Drum Tower Hospital of Nanjing University Medical School · Updated on 2026-09-02
3000
Participants Needed
1
Research Sites
104 weeks
Total Duration
AI-Summary
What This Trial Is About
Researchers are developing and validating a clinical prediction model to assess the risk of head and neck cancerous lesions using deep learning combined with artificial intelligence algorithms. This study involves a national multicenter cohort of approximately 11,000 patients with voice disorders and addresses the challenge of distinguishing early-stage laryngeal cancers from benign voice disorders, which often present with similar symptoms like hoarseness. The research aims to improve early detection and optimize healthcare resource allocation through advanced AI technology. The study uses a two-stage, multimodal AI framework. The first stage integrates demographic data, behavioral exposures, and clinical symptoms to create a non-invasive Clinical Screening Model suitable for primary care settings. The second stage uses deep learning to analyze endoscopic images for precise diagnosis, differentiating between malignant lesions, benign vocal fold lesions, and normal anatomy. A cloud-based web application supports real-time risk estimation. Participants provide demographic data, clinical symptoms, and undergo laryngoscopic imaging during their first outpatient visit. Researchers analyze these data to develop and validate the AI models. The primary outcome is the laryngoscopic report diagnosis on the first day, with secondary outcomes including demographic data and voice handicap index VHI-10 scores. The study spans from initial screening to comprehensive data collection, supporting improved diagnostic accuracy and clinical decision-making.
CONDITIONS
Brief Title
A Machine Learning-Based Risk Prediction Model for Head and Neck Cancerous Lesions
Research Team
J
junyi Wang, master
Not the Right Trial for You?
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here