Agentic AI-Driven Data Product for Automated Healthcare Insights Generation
DOI:
https://doi.org/10.63282/3050-9262.IJAIDSML-V6I1P125Keywords:
Agentic Artificial Intelligence, Healthcare Data Analytics, Predictive Modeling, Random Forest, Automated Insight Generation, Appointment No-Show Prediction, Machine Learning, Explainable AI, Data Product, Feature Engineering, Clinical Decision Support, Autonomous Systems, Healthcare Operations, Artificial Intelligence in Medicine, Data-Driven HealthcareAbstract
This study implements the Agentic AI data product development and delivery process for healthcare appointment no-show prediction with agentic AI for autonomous insights generation post-implementation. The Healthcare Appointment Dataset was obtained from Kaggle, used to predict attendance for scheduled appointments versus no-shows. Notable variables include (but are not limited to) patient age, gender, scholarship attendance, and the number of days a patient was forced to wait (after requesting an appointment) for their appointment. The prediction was thus based on attendance (or not); the model produced an applicable output from a relevant research question with 77.33% accuracy and consistently significant results predictive of practical application within healthcare settings. Moreover, agentic AI was executed to facilitate insights generation independently based on the model results without researcher-induced intervention. The fields produced by the nature of insights discovered reveal that the higher average days of waiting time for the appointment they had, coupled with the lower average age of patients, the better predictive capacity for no-show status. Thus, performance prediction modeling accompanied by agentic insights generation for results assessment increases interpretability, efficacy, and decision support capabilities within the healthcare analytics world. Furthermore, future research will be conducted with a multimodal source reinforcement learning approach to improve prediction accuracy and real-time available metrics solidifying agentic AI within predictive healthcare endeavors for real-world applicability
References
[1] J. Bajwa, U. Munir, A. Nori and B. Williams, “Artificial intelligence in healthcare: transforming the practice of medicine,” Future Healthcare Journal, vol. 8, no. 2, pp. e188–e194, 2021.
[2] A. Bandi, B. Kongari, R. Naguru, S. Pasnoor and S. V. Vilipala, “The rise of agentic AI: A review of definitions, frameworks, architectures, applications, evaluation metrics, and challenges,” Future Internet, vol. 17, no. 9, p. 404, 2025.
[3] M. Chustecki, “Benefits and risks of AI in health care: Narrative review,” Interactive Journal of Medical Research, vol. 13, no. 1, p. e53616, 2024.
[4] D. Ferber et al., “Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology,” Nature Cancer, pp. 1–13, 2025.
[5] A. F. A. Hamdan and A. A. Bakar, “Machine learning predictions on outpatient no-show appointments in a Malaysia major tertiary hospital,” The Malaysian Journal of Medical Sciences (MJMS), vol. 30, no. 5, p. 169, 2023.
[6] D. Liu, W. Y. Shin, E. Sprecher, K. Conroy, O. Santiago, G. Wachtel and M. Santillana, “Machine learning approaches to predicting no-shows in pediatric medical appointments,” NPJ Digital Medicine, vol. 5, no. 1, p. 50, 2022.
[7] A. Mumuni and F. Mumuni, “Automated data processing and feature engineering for deep learning and big data applications: A survey,” Journal of Information and Intelligence, vol. 3, no. 2, pp. 113–153, 2025.
[8] L. H. A. Salazar, W. D. Parreira, A. M. D. R. Fernandes and V. R. Q. Leithardt, “No-show in medical appointments with machine learning techniques: A systematic literature review,” Information, vol. 13, no. 11, p. 507, 2022.
[9] R. Sapkota, K. I. Roumeliotis and M. Karkee, “AI agents vs. agentic AI: A conceptual taxonomy, applications and challenges,” arXiv preprint arXiv:2505.10468, 2025.
[10] A. Tunmise, A. Esther, G. Franca and H. Paranjabey, “Automated feature engineering and transformation in real-time ETL pipelines for predictive modeling,” 2025.
