Privacy Preserving Analytics in Outsourced Healthcare System
D. Sudha Devi1, S.Sudendar2
1Dr. D. Sudha Devi*, Department of Computing, Data Science, Coimbatore Institute of Technology, Coimbatore, India.
2S. Sudendar, Department of Computing, Data Science, Coimbatore Institute of Technology, Coimbatore, India.
Manuscript received on June 16, 2020. | Revised Manuscript received on June 27, 2020. | Manuscript published on July 10, 2020. | PP: 329-333 | Volume-9 Issue-9, July 2020 | Retrieval Number: 100.1/ijitee.I7209079920 | DOI: 10.35940/ijitee.I7209.079920
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The most data intensive industry today is the healthcare system. The advancement in technology has revolutionized the traditional healthcare practices and led to enhanced E-Healthcare System. Modern healthcare systems generate voluminous amount of digital health data. These E-Health data are shared between patients and among groups of physicians and medical technicians for processing. Due to the demand for continuous availability and handling of these massive E-Health data, mostly these data are outsourced to cloud storage. Being cloud-based computing, the sensitive patient data is stored in a third-party server where data analytics are performed, hence more concern about security raises. This paper proposes a secure analytics system which preserves the privacy of patients’ data. In this system, before outsourcing, the data are encrypted using Paillier homomorphic encryption which allows computations to be performed over encrypted dataset. Then Decision Tree Machine Learning algorithm is used over this encrypted dataset to build the classifier model. This encrypted model is outsourced to cloud server and the predictions about patient’s health status is displayed to the user on request. In this system nowhere the data is decrypted throughout the process which ensures the privacy of patients’ sensitive data.
Keywords: E-healthcare, Homomorphic encryption, Decision tree classifier, Cloud server, Privacy preserving, Machine learning.
Scope of the Article: Machine learning