Original research Vol. 1, No. 2 pp. 501–510

An Artificial Intelligence-Based Predictive Framework for Early Detection and Mitigation of Ransomware Attacks in Healthcare Information Systems

Jesse et. al.*
Jesse Alegieuno Security engineer, AI secure. Student, department of information technology and management, school Illinois institute of technology
Published: July 1, 2026 97 views 35 downloads

Abstract

Ransomware attacks have emerged as one of the most significant cybersecurity threats to Healthcare Information Systems (HIS), compromising patient data, disrupting clinical operations, and threatening the continuity of healthcare services. Conventional signature-based detection techniques are increasingly ineffective against sophisticated ransomware variants because they typically identify attacks only after file encryption has commenced. This study developed and evaluated an Artificial Intelligence (AI)-based predictive framework for the early detection and mitigation of ransomware attacks in healthcare environments. An experimental research design was adopted using publicly available ransomware datasets and healthcare-related cybersecurity data. The datasets were preprocessed through feature extraction and selection to identify behavioural indicators associated with the pre-encryption stages of ransomware attacks. Five AI algorithms, namely Random Forest, Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Long Short-Term Memory (LSTM), were trained and evaluated using Accuracy, Precision, Recall, F1-score, Receiver Operating Characteristic Area Under the Curve (ROC-AUC), False Positive Rate (FPR), and Mean Time to Detection (MTTD). The findings revealed that behavioural indicators, including abnormal file modification, API call sequence anomalies, privilege escalation, registry modifications, and network beaconing, are reliable predictors of ransomware activity before encryption occurs. Among the evaluated models, the LSTM algorithm achieved the highest predictive performance with an accuracy of 99.1%, an F1-score of 0.990, and the shortest detection time. The proposed AI-driven framework further improved ransomware resilience through continuous behavioural monitoring, explainable artificial intelligence, and automated incident response mechanisms. The study concludes that AI-based predictive detection provides a more effective and proactive cybersecurity strategy than conventional reactive approaches and offers a practical framework for enhancing the security and resilience of Healthcare Information Systems against evolving ransomware threats.

Keywords: Artificial Intelligence, Ransomware, Healthcare Information Systems, Machine Learning, Deep Learning, Cybersecurity, Behavioural Analysis, Early Detection, Incident Response, Explainable Artificial Intelligence.

Cite This Article

et. al., J. (2026). An Artificial Intelligence-Based Predictive Framework for Early Detection and Mitigation of Ransomware Attacks in Healthcare Information Systems. UniOffaJIS, 1(2), 501–510.
Format: APA (6th/7th Ed.)

Authors

Jesse et. al. Corresponding
Jesse Alegieuno Security engineer, AI secure. Student, department of information technology and management, school Illinois institute of technology

Article Info

Type: Original research

Published: Jul 1, 2026

Volume: 1

Issue: 2

Pages: 501–510