Usability Evaluation and Classification of mHealth Applications for Type 2 Diabetes Mellitus Using MARS and ID3 Algorithm

Kamaldeep Gupta; Sharmistha Roy; Ayman Altameem; Raghvendra Kumar; Abdul Khader Jilani Saudagar; Ramesh Chandra Poonia

doi:10.3390/ijerph19126999

Usability Evaluation and Classification of mHealth Applications for Type 2 Diabetes Mellitus Using MARS and ID3 Algorithm

Int J Environ Res Public Health. 2022 Jun 8;19(12):6999. doi: 10.3390/ijerph19126999.

Authors

Kamaldeep Gupta¹, Sharmistha Roy¹, Ayman Altameem², Raghvendra Kumar³, Abdul Khader Jilani Saudagar⁴, Ramesh Chandra Poonia⁵

Affiliations

¹ Faculty of Computing and Information Technology, Usha Martin University, Ranchi 835103, India.
² Department of Computer Science and Engineering, College of Applied Studies and Community Services, King Saud University, Riyadh 11533, Saudi Arabia.
³ Department of Computer Science and Engineering, GIET University, Rayagada 765022, India.
⁴ Information Systems Department, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11432, Saudi Arabia.
⁵ Department of Computer Science, CHRIST (Deemed to be University), Bangalore 560029, India.

Abstract

The rapid growth of mHealth applications for Type 2 Diabetes Mellitus (T2DM) patients' self-management has motivated the evaluation of these applications from both the usability and user point of view. The objective of this study was to identify mHealth applications that focus on T2DM from the Android store and rate them from the usability perspective using the MARS tool. Additionally, a classification of these mHealth applications was conducted using the ID3 algorithm to identify the most preferred application. The usability of the applications was assessed by two experts using MARS. A total of 11 mHealth applications were identified from the initial search, which fulfilled our inclusion criteria. The usability of the applications was rated using the MARS scale, from 1 (inadequate) to 5 (excellent). The Functionality (3.23) and Aesthetics (3.22) attributes had the highest score, whereas Information (3.1) had the lowest score. Among the 11 applications, "mySugr" had the highest average MARS score for both Application Quality (4.1/5) as well as Application Subjective Quality (4.5/5). Moreover, from the classification conducted using the ID3 algorithm, it was observed that 6 out of 11 mHealth applications were preferred for the self-management of T2DM.

Keywords: ID3; MARS; T2DM; decision making; mHealth applications; usability.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Algorithms
Diabetes Mellitus, Type 2* / therapy
Humans
Mobile Applications*
Neoplasm Proteins
Self-Management*
Telemedicine*

Substances

Neoplasm Proteins

Grants and funding

Project number (RSP2022R498), King Saud University, Riyadh, Saudi Arabia.