Privacy preserving disease risk assessment model using machine learning classifiers

Abstract

In the pre-digital era, medical diagnosis used to consume a lot of time and human resources. But in this digital age, the entire process can be efficiently done with the help of machines. So a new term called prognosis is introduced in this modern era where scientific prediction of the likely development of a disease and its outcome can be done. Machine learning techniques are utilized for prognosis. Though machine learning algorithms solve many problems in the healthcare field, they cannot flourish further without privacy and security assurances as the healthcare field consists of more sensitive data. Our application addresses this issue and provides a fully secured disease risk assessment using an encryption algorithm called RSA. The main objective here is to build a system with the following aspects, Functional Aspect: Disease prediction using machine learning technique decision trees. Non-Functional Aspect: Providing privacy and confidentiality of data using encryption techniques like RSA. Data privacy and confidentiality is provided with the help of the RSA algorithm. RSA is the first successful public key cryptographic algorithm.