Peer-reviewed research, distributed systems architectures, and digital health innovations authored by team members and academic research collaborators.
M. Hasan, A. Fukuda, R. I. Maruf, F. Yokota, A. Ahmed
Examines the nature and genesis of data input and system errors in remote healthcare platforms. Analyzes where, how, and by whom errors are introduced during field telemetry and vital sign acquisition, proposing point-of-care verification mechanisms to safeguard clinical decision reliability.
M. B. Sampa, N. H. Abdul Aziz, M. S. Rahman, M. Hasan, N. A. Ab. Aziz
Systematically reviews empirical literature evaluating key determinants of eHealth adoption across patients and healthcare practitioners. Synthesizes adoption dynamics spanning teleconsultation, electronic medical records (EMRs), and mHealth, identifying critical technical, organizational, and behavioral drivers to inform national digital health rollouts.
A. Ahmed, M. Hasan, M. B. Sampa, K. M. Hossein, Y. Nohara, N. Nakashima
Presents the concept, system architecture, and field implementation of the Portable Health Clinic (PHC), a remote telemedicine and diagnostic platform co-developed with Kyushu University. Details hardware packaging, edge sensor integration, encrypted cloud data pipelines, and deployment findings across rural and unreached communities.
A. Chowdhury, S. Dey, S. Hossain, M. Hasan, S. Chowdhury
Heart failure is a critical global health challenge demanding precise prognostic instruments. This study evaluates machine learning techniques to predict survival and time-to-event outcomes in heart failure patients, combining clinical parameters and explainable AI (XAI) to optimize clinical cardiovascular decision support.
M. S. Rahman, M. A. R. Khan, M. Nijim, M. Al Aqqad, H. Tomioka, B. Shin, M. Hasan
Proposes a privacy-preserving federated deep learning framework for nomophobia risk classification from smartphone telemetry while maintaining client data residency. Integrates Differential Privacy (DP-SGD) with encrypted transmission of masked local parameter updates, achieving 99.12% accuracy and 0.9997 AUC under FedAvg across distributed multi-client topologies.