A Comprehensive Research on Privacy-Preserving in Health Data Analytics
Praveen Kumar Latchupatula, Abdolwadwd Alzubaidy, A V Shreyas., Krishna Prakash R, Yogesh Ramaswamy · 2025
Digital healthcare systems have generated massive volumes of sensitive health data, allowing data-driven medical diagnostics, personalized treatment, and public health research. However, the use of such data increases serious privacy issues, owing to risk of data breaches, unauthorized access, and inappropriate use of personal health information. This has necessitated the improvement of privacy-preserving techniques to ensure secure and ethical use of health data. However, several privacy-preserving techniques, such as differential privacy, homomorphic encryption, secure multiparty computation, and secure sensitive health data, frequently reduce data utility, degrade model accuracy, and result in computational costs. However, this is a serious limitation in healthcare, in which both data accuracy and secure privacy guarantees are crucial for safe clinical decisions and meaningful insights. Additionally, there is a rising need to strike a balance between effective data analysis and strict privacy protection, as health data remain to fuel AI-driven innovations, and confirming their confidentiality without delaying analytic value is vital. Furthermore, existing privacy-preserving techniques often struggle to ensure high utility while confirming strong privacy guarantees, thereby highlighting the necessity for stronger, scalable, and adaptive solutions.