AI-Driven Decision Support Systems for Neonatal Care: A Reinforcement Learning Approach
DOI:
https://doi.org/10.52783/jns.v14.3398Keywords:
Neonatal care, Reinforcement learning, Clinical decision support, Surgical decision-making, Artificial intelligenceAbstract
Neonatal surgical care involves high-stakes decision-making under conditions of uncertainty, urgency, and limited physiological feedback. Clinicians often rely on experience and generalised guidelines to decide whether to proceed with surgical intervention—an approach that may not always capture the nuances of individual patient presentations. Artificial Intelligence (AI), particularly Reinforcement Learning (RL), offers a promising avenue to improve consistency and adaptiveness in such critical care scenarios. Unlike traditional machine learning models, RL learns optimal decision strategies by balancing rewards and risks through iterative feedback, making it suitable for sequential and high-impact clinical environments. This study presents a conceptual AI-driven Decision Support System (DSS) that leverages RL principles to assist clinicians in binary surgical decisions for neonates. The system employs manually constructed clinical state–action mappings, expert-informed reward logic, and an explainable Q-table rather than relying on patient data or simulations. It features a three-layered architecture, visual decision flowchart, and event tree to support transparent reasoning. Through hypothetical clinical scenarios and clinician-oriented workflow modelling, the system demonstrates potential for low-resource settings and academic prototyping. While preliminary, the model offers a scalable, modifiable foundation for future integration with real-world clinical platforms, aiming to enhance decision quality in neonatal surgical care.
Downloads
Metrics
References
Catania, L. J. (2021). 4 - AI applications in the business and administration of health care. In L. J. Catania (Ed.), Foundations of Artificial Intelligence in Healthcare and Bioscience (pp. 79–123). Academic Press. https://doi.org/https://doi.org/10.1016/B978-0-12-824477-7.00003-1
Guez-Barber, D., & Pilon, B. (2024). Parental impact during and after neonatal intensive care admission. Seminars in Perinatology, 48(5), 151926. https://doi.org/https://doi.org/10.1016/j.semperi.2024.151926
Jaile, J. C., Brady, J. D., Nelson, P., Sourour, W., Almodovar, M. C., Macicek, S., Pettitt, T. W., & Pigula, F. A. (2024). Cardiac Resynchronization Therapy for Pacing-Related Dysfunction Post Cardiac Surgery in Neonates. Annals of Thoracic Surgery Short Reports, 2(4), 825–828. https://doi.org/https://doi.org/10.1016/j.atssr.2024.05.007
Jeong, H., & Kamaleswaran, R. (2022). Pivotal challenges in artificial intelligence and machine learning applications for neonatal care. Seminars in Fetal and Neonatal Medicine, 27(5), 101393. https://doi.org/https://doi.org/10.1016/j.siny.2022.101393
Jyoti, J., Laing, S., Spence, K., Griffiths, N., & Popat, H. (2023). Parents’ perspectives on their baby’s pain management in a surgical neonatal intensive care unit: The parents’ awareness and involvement in pain management (PAIN-PAM) study - Part 2. Journal of Neonatal Nursing, 29(6), 839–845. https://doi.org/https://doi.org/10.1016/j.jnn.2023.06.007
Khan, S. M. (2025). Chapter 14 - AI-enabled decision support systems in clinical practice. In S. M. Khan (Ed.), Fundamentals of AI for Medical Education, Research and Practice (pp. 305–330). Academic Press. https://doi.org/https://doi.org/10.1016/B978-0-443-33584-6.00014-1
Kovalchuk, S. V, Kopanitsa, G. D., Derevitskii, I. V, Matveev, G. A., & Savitskaya, D. A. (2022). Three-stage intelligent support of clinical decision making for higher trust, validity, and explainability. Journal of Biomedical Informatics, 127, 104013. https://doi.org/https://doi.org/10.1016/j.jbi.2022.104013
Lakhan, A., Nedoma, J., Mohammed, M. A., Deveci, M., Fajkus, M., Marhoon, H. A., Memon, S., & Martinek, R. (2024). Fiber-optics IoT healthcare system based on deep reinforcement learning combinatorial constraint scheduling for hybrid telemedicine applications. Computers in Biology and Medicine, 178, 108694. https://doi.org/https://doi.org/10.1016/j.compbiomed.2024.108694
Levin, C., Kagan, T., Rosen, S., & Saban, M. (2024). An evaluation of the capabilities of language models and nurses in providing neonatal clinical decision support. International Journal of Nursing Studies, 155, 104771. https://doi.org/https://doi.org/10.1016/j.ijnurstu.2024.104771
Muntean, A., Marsland, L., Sikdar, O., Harris, C., Ade-Ajayi, N., Patel, S. B., Cook, J., Sellars, M., Greenough, A., Nicolaides, K., & Davenport, M. (2025). Neonatal Surgery for Congenital Lung Malformations: Indications, Outcomes and Association With Malignancy. Journal of Pediatric Surgery, 60(5), 162253. https://doi.org/https://doi.org/10.1016/j.jpedsurg.2025.162253
Nadhir, A. M., Mounir, B., Abdelkader, L., & Hammoudeh, M. (2025). Enhancing Cybersecurity in Healthcare IoT Systems Using Reinforcement Learning. Transportation Research Procedia, 84, 113–120. https://doi.org/https://doi.org/10.1016/j.trpro.2025.03.053
Onapakala, K., Varma, M. N., Nagesh, M. A., Maturi, S., Kumari, P. L., Saibaba, C. M. H., Bommisetty, J., & Donthi, R. (2024). Multi-Task Deep Learning Approaches For Named Entity Recognition, Sentiment Analysis, And Summarization In Natural Language Processing. African Journal of Biomedical Research, 27(4S), 5713–5720. https://doi.org/10.53555/AJBR.v27i4S.4670
Roayaei, M., & Soltani, Z. (2025). Chapter 2 - Advancing healthcare: Reinforcement learning applications for personalized healthcare. In S. Mahajan & A. K. Pandit (Eds.), Innovations in Biomedical Engineering (pp. 33–86). Academic Press. https://doi.org/https://doi.org/10.1016/B978-0-443-30146-9.00002-2
Thakre, B., Yadav, U., & Bondre, S. V. (2025). Chapter 8 - Deep reinforcement learning in healthcare and biomedical application. In S. Mahajan & A. K. Pandit (Eds.), Innovations in Biomedical Engineering (pp. 241–299). Academic Press. https://doi.org/https://doi.org/10.1016/B978-0-443-30146-9.00008-3
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
