Publication Type
Journal Article
Publication Date (Issue Year)
2024
Journal Name
Journal of Cybersecurity and Privacy
Abstract
Globally, 77% of the elderly aged 65 and above suffer from multiple chronic ailments, according to recent research. However, several barriers within the healthcare system in the developing world hinder the adoption of home-based patient management, hence the need for the IoMT, whose application raises security concerns, particularly in authentication. Several authentication techniques have been proposed; however, they lack a balance of security and usability. This paper proposes a Naive Bayes based adaptive user authentication app that calculates the risk associated with a login attempt on an Android device for elderly users, using their health conditions, risk score, and available authenticators. This authentication technique guided by the MAPE-KHMT framework makes use of embedded smartphone sensors. Results indicate a 100% and 98.6% accuracy in usable-security metrics, while cross-validation and normalization results also support the accuracy, efficiency, effectiveness, and usability of our model with room for scaling it up without computational costs and generalizing it beyond SSA. The post-deployment evaluation also confirms that users found the app usable and secure. A few areas need further refinement to improve the accuracy, usability, security, and acceptance but the model shows potential to improve users’ compliance with IoMT security, thereby promoting the attainment of SDG3.
Keywords
elderly patients, SSA, chronic ailments, risk calculation, adaptive authentication, smartphone, usable security
Rsif Scholar Name
Prudence Munyaradzi Mavhemwa
Thematic Area
ICTs Including Big Data and Artificial Intelligence
Africa Host University (AHU)
University of Rwanda (UR), Rwanda
Funding Statement
The APC was funded by PASET-RSIF
Recommended Citation
Mavhemwa, P. M., Zennaro, M., Nsengiyumva, P., & Nzanywayingoma, F. (2024). An android-based Internet of Medical Things adaptive user authentication and authorization model for the elderly. Journal of Cybersecurity and Privacy, 4, 993-1017. https://doi.org/10.3390/JCP4040046