Deep Reinforcement Learning for personalized diagnostic decision pathways using Electronic Health Records: A comparative study on anemia and Systemic Lupus Erythematosus

• Published in: Artificial intelligence in medicine 2024-11, Vol.157, p.102994, Article 102994
• Main Authors: Muyama, Lillian, Neuraz, Antoine, Coulet, Adrien
• Format: Article
• Language: English
• Subjects: Algorithms; Anemia; Anemia - diagnosis; Artificial Intelligence; Clinical Decision-Making; Computer Science; Critical Pathways; Decision pathways; Decision support; Deep Learning; Deep reinforcement learning; Diagnosis; Electronic Health Records; Human health and pathology; Humans; Life Sciences; Lupus; Lupus Erythematosus, Systemic - diagnosis; Precision Medicine - methods
• ISSN: 0933-3657; 1873-2860; 1873-2860
• DOI: 10.1016/j.artmed.2024.102994

Clinical diagnoses are typically made by following a series of steps recommended by guidelines that are authored by colleges of experts. Accordingly, guidelines play a crucial role in rationalizing clinical decisions. However, they suffer from limitations, as they are designed to cover the majority of the population and often fail to account for patients with uncommon conditions. Moreover, their updates are long and expensive, making them unsuitable for emerging diseases and new medical practices. Inspired by guidelines, we formulate the task of diagnosis as a sequential decision-making problem and study the use of Deep Reinforcement Learning (DRL) algorithms to learn the optimal sequence of actions to perform in order to obtain a correct diagnosis from Electronic Health Records (EHRs), which we name a diagnostic decision pathway. We apply DRL to synthetic yet realistic EHRs and develop two clinical use cases: Anemia diagnosis, where the decision pathways follow a decision tree schema, and Systemic Lupus Erythematosus (SLE) diagnosis, which follows a weighted criteria score. We particularly evaluate the robustness of our approaches to noise and missing data, as these frequently occur in EHRs. In both use cases, even with imperfect data, our best DRL algorithms exhibit competitive performance compared to traditional classifiers, with the added advantage of progressively generating a pathway to the suggested diagnosis, which can both guide and explain the decision-making process. DRL offers the opportunity to learn personalized decision pathways for diagnosis. Our two use cases illustrate the advantages of this approach: they generate step-by-step pathways that are explainable, and their performance is competitive when compared to state-of-the-art methods. •We adapt the reinforcement learning framework to diagnosis decision support.•Our approach progressively constructs optimal sequences of actions to reach a diagnosis, which we refer to as diagnostic decision pathways.•We perform an empirical analysis on synthetic Electronic Health Records that shows that our performs competitively with state-of-the-art classifiers.•Our approach is robust to noise and missing data.•The resulting pathways are explainable and follow clinical reasoning.