Advancing healthcare cloud security through machine learning for proactive threat detection and prevention
Jagendra Singh, Neha Garg, Nikita Manne, S. Madhumala Vinnakota, Vijaya Kumar Koppula, Rinki Tyagi · 2025
In the modern healthcare environment, cloud computing technology has drastically transformed the storage, organization, and availability of patient records. However, with these benefits come substantial security concerns, as the field is under-going an increase in cyber-attacks and related security breaches. Thus, it is essential to investigate and implement a solution to improve security. The present re-search considers using machine learning algorithms within a healthcare cloud environment to increase security. A dataset implemented in the research included a variety of user information, such as authentication logs, transmission rates, and communication frequencies, forming a dataset of 1220 users within the cloud system. The artificial neural network, decision tree, support vector machine and the k-nearest neighbor machine learning algorithms were trained and tested on the data to predict abnormal communication patterns, unusual activities, protocol mismatches, or network outliers. The artificial neural network algorithm demonstrated the best performance with an accuracy of 96.78%. However, the differences were relatively small with all of the methods demonstrating good performance, with accuracy rates of 96.78%, 93.45%, 92.34%, and 87.6% for ANN, decision tree, SVM, and KNN, respectively. Precision and recall information indicates a similarity in the artifacts’ abilities to identify false positives and negatives. Overall, the present research is relevant in the context of the increasing im-portance of cybersecurity and machine learning applications.