Title: Rough information theory based approach to manage uncertainty in remote healthcare

Authors: Sayan Das; Jaya Sil

Addresses: Department of Computer Science and Technology, Indian Institute of Engineering Science and Technology, Shibpur, Howrah – 711103, West Bengal, India ' Department of Computer Science and Technology, Indian Institute of Engineering Science and Technology, Shibpur, Howrah – 711103, West Bengal, India

Abstract: Scarcity of doctors and lack of skilled manpower in rural India is a real challenge in diagnosing the people even for the common health-related problems such as flu, diarrhea etc. Due to a lack of domain knowledge and expertise, health assistants are unable to consistently categories patients as diseased or not. The paper aims to develop a rough information theoretic approach for diagnosing patients with minimum false cases by reducing uncertainty in health dataset. Knowledge granulation of rough set theory is used to partition the patients into positive region (PR, certainly diagnosed), and boundary region (BR, possibly diagnosed). Conditional entropy of patients in BR is measured, considering patients of PR and accordingly feature values of patients in BR are revised. The model is suitable for providing primary healthcare to the patients based on diagnosis; however, not substitutions of doctors and in emergency cases, patients are referred to the experts.

Keywords: uncertainty management; rough set; information theory; remote healthcare; conditional entropy; fuzzy c-mean; rule base; decision system.

DOI: 10.1504/IJFCM.2022.124361

International Journal of Fuzzy Computation and Modelling, 2022 Vol.4 No.1, pp.73 - 103

Received: 24 Mar 2021
Accepted: 27 Jul 2021
Published online: 25 Jul 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article