The article "Medical-informed machine learning: integrating prior knowledge into medical decision systems" by Christel Sirocchi, Alessandro Bogliolo, and Sara Montagna, presented at the 18th Conference on Computational Intelligence Methods for Bioinformatics and Biostatistics, highlights the importance of integrating medical domain knowledge into machine learning (ML) models for clinical applications. The authors argue that while ML has shown promise in various medical domains, its impact on clinical care remains limited due to the complexity and variability of clinical data. They advocate for the integration of medical knowledge at each stage of the ML pipeline, including data preprocessing, feature engineering, model training, and output evaluation. The study reviews existing integration strategies and maps them onto these stages, providing a structured overview. A case study on diabetes prediction is used to demonstrate the effectiveness of these strategies. Domain knowledge, formalized as rules, causal networks, intervals, and formulas, is integrated at each stage, resulting in a range of hybrid models. The results show that integrated models outperform purely data-driven approaches in terms of accuracy, interpretability, data efficiency, and adherence to clinical guidelines. Notably, these models maintain performance under limited data scenarios. The authors identify two main challenges for future research: refining the representation of medical domain knowledge and fine-tuning its contribution to the ML model. They conclude by providing guidance for future integration efforts, emphasizing the need for more comprehensive integration strategies tailored to the healthcare sector.The article "Medical-informed machine learning: integrating prior knowledge into medical decision systems" by Christel Sirocchi, Alessandro Bogliolo, and Sara Montagna, presented at the 18th Conference on Computational Intelligence Methods for Bioinformatics and Biostatistics, highlights the importance of integrating medical domain knowledge into machine learning (ML) models for clinical applications. The authors argue that while ML has shown promise in various medical domains, its impact on clinical care remains limited due to the complexity and variability of clinical data. They advocate for the integration of medical knowledge at each stage of the ML pipeline, including data preprocessing, feature engineering, model training, and output evaluation. The study reviews existing integration strategies and maps them onto these stages, providing a structured overview. A case study on diabetes prediction is used to demonstrate the effectiveness of these strategies. Domain knowledge, formalized as rules, causal networks, intervals, and formulas, is integrated at each stage, resulting in a range of hybrid models. The results show that integrated models outperform purely data-driven approaches in terms of accuracy, interpretability, data efficiency, and adherence to clinical guidelines. Notably, these models maintain performance under limited data scenarios. The authors identify two main challenges for future research: refining the representation of medical domain knowledge and fine-tuning its contribution to the ML model. They conclude by providing guidance for future integration efforts, emphasizing the need for more comprehensive integration strategies tailored to the healthcare sector.