In traditional medical education, learners are mostly trained to diagnose and treat patients through supervised practice. Artificial Intelligence and simulation techniques can complement such an educational practice. In this paper, we present GLARE-Edu, an innovative system in which AI knowledge-based methodologies and simulation are exploited to train learners “how to act” on patients based on the evidence-based best practices provided by clinical practice guidelines. GLARE-Edu is being developed by a multi-disciplinary team involving physicians and AI experts, within the AI-LEAP (LEArning Personalization of AI and with AI) Italian project. GLARE-Edu is domain-independent: it supports the acquisition of clinical guidelines and case studies in a computer format. Based on acquired guidelines (and case studies), it provides a series of educational facilities: (i) navigation, to navigate the structured representation of the guidelines provided by GLARE-Edu, (ii) automated simulation, to show learners how a guideline would suggest to act, step-by-step, on a specific case, and (iii) (self)verification, asking learners how they would treat a case, and comparing step-by-step the learner’s proposal with the suggestions of the proper guideline. In this paper, we describe GLARE-Edu architecture and general features, and we demonstrate our approach through a concrete application to the melanoma guideline and we propose a preliminary evaluation.

On the basis of our 25-year experience with the GLARE (Guideline Acquisition, Representation and Execution) clinical decision support system, we have started to analyze the adoption of computer-interpretable clinical guidelines (CIGs) and AI techniques to train and test medical students about how to act on patients. Moving from decision support to the educational task involves significant research challenges. In this paper, we propose a new facility that supports teachers in the definition of tests, by selecting and hiding to students specific parts of the CIG, and asking students how they would act on the given case study (patient) in the selected parts. Students are provided with a medical ontology to identify proper actions/decisions, and students' proposals are then automatically compared with what the CIG (considered as a "golden standard") would suggest to do to the patient through knowledge representation and reasoning techniques. Our basic explanation mechanism exploits the medical ontology to show to students the differences (if any) between their proposals and the ones of the CIG.

Education is a challenging task in the healthcare context. Clinical practice guidelines (CPGs) encode the “best” evidence-based medical procedures to manage a specific disease, and are fundamental tools to specify how to act on patients following evidence-based recommendations. In this paper, we propose to complement traditional medical education by introducing advanced AI techniques to facilitate the learning of CPGs. The starting point of our approach is GLARE (Guideline Acquisition, Representation and Execution), a system to acquire and represent Computer-Interpretable Clinical Guidelines (CIGs), and to execute them on patients. In particular, we propose a new computer-based tool to support students’ self-verification. Our tool proposes students a “shadow” of node in the CIG representation, and let students fill it with concrete medical actions and decisions. At each step, our tool compares and evaluates the student's proposals with the “golden standard” provided by the CIG.

Mantle cell lymphoma (MCL) has a heterogeneous clinical course, making robust, usable prognostic tools essential for risk-adapted care. In this paper, we worked at the definition of an effective and deployable prognostic model, tailored to a specific MCL population. Our work has taken advantage of AutoML-Med, a framework supporting end-users in the data analysis pipeline. Our experiments have provided promising results, leading to an explainable model that outperformed traditional prognostic indexes, paving the way for future investigations.