🤖 AI Summary
Clinical trials for knee osteoarthritis (KOA) suffer from high costs, inefficient data collection, and limitations of existing predictive models—including static architectures and exclusive focus on the knee joint. Method: We propose the first reinforcement learning–based active sensing framework explicitly optimized for clinical trial objectives. Operating at the patient level, it dynamically monitors multi-site disease progression across the body, integrating multimodal data (imaging, clinical, and functional). Using Proximal Policy Optimization (PPO) with a custom sparse reward function, it autonomously determines optimal follow-up timing and data modalities. Contribution/Results: Our end-to-end, clinical-goal-driven framework enables adaptive longitudinal data acquisition—breaking from static modeling paradigms. Experiments demonstrate significant improvement over state-of-the-art methods in KOA progression prediction: average follow-up cost reduced by 32%, and data information gain increased by 19.7%. This work provides a key technical foundation for deployable next-generation adaptive clinical trials.
📝 Abstract
Osteoarthritis (OA) is the most common musculoskeletal disease, with knee OA (KOA) being one of the leading causes of disability and a significant economic burden. Predicting KOA progression is crucial for improving patient outcomes, optimizing healthcare resources, studying the disease, and developing new treatments. The latter application particularly requires one to understand the disease progression in order to collect the most informative data at the right time. Existing methods, however, are limited by their static nature and their focus on individual joints, leading to suboptimal predictive performance and downstream utility. Our study proposes a new method that allows to dynamically monitor patients rather than individual joints with KOA using a novel Active Sensing (AS) approach powered by Reinforcement Learning (RL). Our key idea is to directly optimize for the downstream task by training an agent that maximizes informative data collection while minimizing overall costs. Our RL-based method leverages a specially designed reward function to monitor disease progression across multiple body parts, employs multimodal deep learning, and requires no human input during testing. Extensive numerical experiments demonstrate that our approach outperforms current state-of-the-art models, paving the way for the next generation of KOA trials.