Academic Research Profile

Trustworthy AI for Healthcare Research

Research Philosophy

I focus on developing trustworthy AI methods for healthcare in resource-limited settings, with particular emphasis on bias mitigation and fairness in medical AI systems. My research bridges theoretical foundations with practical healthcare applications, aiming to create AI systems that are not only accurate but also fair, transparent, and reliable.

Publications

80+

Citations

h-index

Students Mentored

📄 Download Academic CV 🎓 Google Scholar Profile

Research Interests & Expertise

🏥

Trustworthy AI in Healthcare

Developing AI systems that are fair, transparent, and reliable for medical applications in resource-limited settings.

Fairness Transparency Reliability

⚖️

Bias Mitigation in Medical AI

Creating frameworks to detect and reduce algorithmic bias in medical AI systems across demographic groups.

Demographic Bias Fairness Metrics Mitigation

🧠

Multimodal Medical AI

Processing diverse medical data types including clinical records, medical imaging, and IoT sensor streams.

Computer Vision NLP IoT Integration

Future Research Directions

🚀 Emerging Areas

• Federated learning for privacy-preserving medical AI
• Cross-modal bias amplification in multimodal systems
• Explainable AI for clinical decision support

🎯 Research Goals

• Develop universal bias detection frameworks
• Create trustworthy AI benchmarks for healthcare
• Bridge research-to-practice gaps in medical AI

Publications & Research Output

My research has been published in top-tier venues including Scopus-indexed journals and IEEE conferences, with a focus on practical solutions for healthcare AI challenges.

Published

Peer-reviewed publications

Under Review

Manuscripts in review

In Progress

Ongoing research projects

Key Publications

Most Cited Scopus Indexed Healthcare Analytics

A Medical Cyber-Physical System for Predicting Maternal Health in Developing Countries Using Machine Learning

Healthcare Analytics 2024, 100285 • 38 Citations

Architected ensemble learning framework achieving 99% prediction accuracy while reducing algorithmic bias by 34% through novel fairness constraints. Established new benchmarks in maternal health AI and demonstrated the feasibility of deploying trustworthy AI in resource-limited healthcare settings.

Medical AI Bias Mitigation Ensemble Learning Healthcare CPS

📄 Journal Article 🏆 Most Impact 💻 Code Available

SMOTE Oversampling and Near Miss Undersampling Based Diabetes Diagnosis

Published

IEEE ISCC 2023 • 13 Citations

Novel approach combining SMOTE oversampling with Near Miss undersampling for diabetes diagnosis, featuring explainable AI visualization techniques.

Conference XAI

📄 IEEE Conference

An IoT-Based Real-Time Environmental Monitoring System

Published

JARASET 2025 • 8 Citations

Real-time environmental monitoring system with IoT sensors and machine learning for air quality prediction in developing areas.

Journal IoT

📄 Scopus Journal

Papers Under Review

Multi-Disease Prediction Framework

PLOS ONE (Under Review)

Comprehensive framework integrating 4 disease models with real-time inference

Parkinson's Disease Prediction with SMOTE

PLOS ONE (Under Review)

Advanced ML approach with bias mitigation and explainable AI techniques

📚 View Complete Publication List on Google Scholar

Research Experience & Positions

Research Assistant

Medical Cyber-Physical Systems

Shanto-Mariam University of Creative Technology

Current

Duration: Jul 2022 – Present | Advisor: Prof. Mohammad Mobarak Hossain

Key Research Contributions:

Architected ensemble learning framework for maternal health (99% accuracy)
Reduced algorithmic bias by 34% through novel fairness constraints
Implemented homomorphic encryption on IoT devices
Published 3 co-author papers with 45+ combined citations
Conducted systematic review of 100+ IoT healthcare papers

Medical AI Bias Mitigation IoT Security

Research Assistant

Computer Vision in Medicine

EMPATHY LAB, Independent University Bangladesh

2 Years

Duration: Jan 2023 – Dec 2024

Reviewed 125+ papers on healthcare CV applications
Identified CNNs and Vision Transformers as leading techniques
Highlighted key domains: surgical assistance, disease detection
Conducted systematic analysis of medical imaging applications

Computer Vision Medical Imaging Literature Review

Research Assistant

Healthcare AI & IoT Systems

Rising Research Lab

2+ Years

Duration: Aug 2022 – Feb 2025 | Independent Research

Developed multi-disease prediction system (92% average accuracy)
Engineered IoT air quality monitoring device (99.5% uptime)
Implemented AutoML pipeline (60% development time reduction)
Integrated 4 disease models with real-time inference

Disease Prediction IoT Systems AutoML

Key Research Projects

Featured Research Open Source

M-TRUST: Medical AI Bias Mitigation Framework

Comprehensive open-source bias detection toolkit for medical AI with 4 fairness metrics: demographic parity, equalized odds, calibration, and disparate impact. Achieved 30.8% reduction in demographic bias while maintaining 98% model accuracy.

30.8%

Bias Reduction

98%

Accuracy Maintained

📦 PyPI Package 💻 GitHub

Research Impact

First open-source medical AI bias toolkit
Comprehensive fairness evaluation framework
Industry-ready deployment capability
Advancing trustworthy AI research

Maternal Health Risk Prediction System

ML-based risk stratification achieving 99% accuracy with homomorphic encryption for privacy-preserving computation.

99%

Accuracy

Citations

XGBoost Homomorphic Encryption Privacy-Preserving

📄 Healthcare Analytics 2024

Multimodal Chest X-ray Diagnosis System

CheXNet-based system for 14 thoracic diseases with bias analysis. Discovered 4.3× bias amplification in multimodal data.

Diseases

4.3×

Bias Discovery

CheXNet BioBERT Grad-CAM

💻 GitHub Repository

Teaching & Mentoring Experience

AI Training Leadership

Leading AI training batch as part of EDGE Project, mentoring 50+ students in ML/AI

Research Presentations

Presenting at inter-university competitions and academic conferences

Interview Panel

Evaluating technical skills and mentoring future researchers

AI Trainer

Enhancing Digital Government Economy (EDGE) Project
May 2023 - Nov 2023

Teaching Contributions:

• Trained 50+ students in ML/AI through hands-on projects
• Developed comprehensive curriculum covering supervised learning, clustering, regression
• Created educational materials with clear explanations of complex ML concepts
• Provided detailed feedback on student ML models and improvement strategies

50+

Students Trained

Research Mentor

UITS Summer Research Program
Summer 2023, 2024

Mentoring Achievements:

• Mentored 2 research groups (total 7 students) on ML and IoT projects
• Guided teams from conception to publication: literature review, design, implementation
• Achieved research outcomes: 1 paper accepted at Scopus-indexed journal
• Supervised 1 additional paper currently under review

Students Mentored

Academic Service & Leadership

Conference & Event Organization

UITS Zero One Fest 2022

Poster Presentation Coordinator

Leadership Roles

Senior Executive

Research Hub Wings, UITS Computer Club (2022-2023)

Awards & Recognition

🏆

1st Place

Inter-university PowerPoint Presentation Competition

UITS 2020

🥈

2nd Place

Inter-university Programming Contest

45 teams, UITS 2022

⚡

Fastest Problem Solver

UITS Victory Day Programming Contest

2021

⭐

Employee of the Month

Outstanding AI Engineering Performance

Primacy Infotech LTD, Aug 2023

📊

Research Impact

80+ Citations across 9 Publications

h-index: 4

📝

Journal Reviewer

Elsevier "Informatics and Health"

September 2025

📄 View Certificate

🌟

Open Source Contributor

M-TRUST Framework on PyPI

First Medical AI Bias Toolkit