AI-based Predictive and Interventional System for Early Detection of Non-compliance Risks With Oral Therapies in Lymphoma Patients, Integrating the Complete Care Pathway and an Interoperable Clinical Interface With Algorithms Paired With Explainability Tools

What this Trial Is About

This research focuses on patients with Non-Hodgkin Lymphomas who require strict treatment protocols including intensive intravenous chemotherapy or oral targeted therapies. The study aims to assess patient adherence to oral therapies using an artificial intelligence (AI) tool to improve early detection of non-compliance risks. This approach seeks to enhance clinical intervention accuracy and responsiveness by overcoming limitations of current adherence monitoring methods. The study involves two groups: a retrospective cohort of 350 lymphoma patients monitored from 2019 to 2024 on an empirical basis, and a prospective cohort of up to 210 patients starting in November 2025. The prospective group will be followed to develop and validate an AI-based decision-support system that analyzes clinical data and uses machine learning to identify patients at risk of non-compliance. The system will generate real-time alerts and tailored recommendations for healthcare providers via an intuitive interface. Participants will be treated at the Hematology Department and provide informed consent. Researchers will monitor compliance trends using clinical data such as dispensing history, laboratory results, and drug interactions. The primary outcome is the accuracy of the AI tool measured by ROC-AUC in 2027, with secondary outcomes including the F1-score by the end of 2027. The study will continue through February 2029, involving ongoing data collection, analysis, and clinical validation of the AI system.

AI-based Predictive and Interventional System for Early Detection of Non-compliance Risks With Oral Therapies in Lymphoma Patients.