Papers & Posters

Agentic AI to Augment Unstructured Data for Information Exchange and LLM

M. Habibi and M. Nourani — IEEE Journal of Biomedical and Health Informatics (JBHI), 2026

This work presents a multi-agent, agentic AI platform for converting unstructured clinical narratives into standardized structured formats, achieving improved extraction performance and reduced hallucinations while supporting cross-institutional interoperability.

Reconstruction of T1-Weighted Contrast-Enhanced MRI for Glioblastoma Radiogenomic Classification

F. Parsaee, M. C. Stefan, and M. Habibi — IEEE Engineering in Medicine and Biology Society (EMBC), Copenhagen, July 2026

This paper proposes a contrast-free deep learning framework that synthesizes T1wCE information from conventional T1w MRI for glioblastoma radiogenomic classification, achieving comparable performance while reducing reliance on gadolinium-based contrast agents.

A Scalable Dual-Camera Platform for Behavioral Assessment in Healthcare

M. Habibi, M. Nourani, and D. H. Sullivan — IEEE International Conference on Healthcare Informatics (ICHI), Minneapolis, MN, June 2026

This research describes a scalable dual-camera video-based monitoring platform that integrates computer vision and AI to assess patient behavior, posture, ambulation, and object interactions for clinical monitoring and risk assessment.

An Integrated Deep Learning Architecture for Illumination-Aware Object Detection

M. Habibi and M. Nourani — IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI), Santa Fe, NM, March 2026

This study proposes an illumination-aware object detection framework that combines image enhancement and detection in a jointly trained architecture, improving robustness and efficiency for low-light visual monitoring.

AI-Based Performance Analysis for Track and Field Athletes

M. Habibi, M. Nourani, and M. M. Nourani — 47th Annual IEEE Engineering in Medicine and Biology Society (EMBC), Copenhagen, Denmark, July 2025

This paper outlines an AI-driven methodology for analyzing and ranking track and field athletes using historical race data, enabling coaches to assess performance trends, consistency, ART2 neural clustering to quantify progress and future competitiveness.

Video-Based Human-Object Interaction Analysis for Patient Behavioral Monitoring

M. Habibi, Z. Delaram, M. Nourani, and D. H. Sullivan — IEEE 13th International Conference on Healthcare Informatics (ICHI), Rende, Italy, June 2025

This work develops a computer vision and machine learning framework for systematic patient behavior analysis through posture recognition, ambulation assessment, and interaction detection with clinically relevant objects.

Utilizing Generative AI for Patient Behavioral Assessment

M. Habibi and M. Nourani — IEEE 18th Dallas Circuits and Systems Conference (DCAS), 2025

This publication presents a vision-language framework for clinical environments that integrates computer vision and large language models to analyze patient behavior and generate context-aware clinical reports in a non-intrusive and scalable manner.

An AI-Driven Camera-Based Platform for Patient Ambulation Assessment

M. Habibi, M. Nourani, and D. H. Sullivan — 46th Annual IEEE Engineering in Medicine and Biology Society (EMBC), 2024

This article details a video-based ambulation assessment platform that uses deep neural networks to detect body and head position, estimate movement, and classify posture for healthcare monitoring.

AI-Based Kinematic Analysis for Track Athletes

M. Habibi, M. Nourani, and M. M. Nourani — IEEE International Conference on Smart Computing (SMARTCOMP), Osaka, Japan, 2024

This research introduces an AI-driven framework for running posture analysis that combines video pose estimation, kinematic analysis, and posture classification to support technique refinement, performance enhancement, and injury prevention.

Introducing a New Classification Method using a Combined Approach of Machine Learning and Multi-Criteria Decision Making

M. Habibi, Z. Delaram, and M. Kouchaki — International Conference on Science and Technology, Iran, 2022 (Farsi)

Combining Machine Learning (ML) and Multi-Criteria Decision Making (MCDM) creates powerful hybrid frameworks that enhance predictive accuracy while ensuring explainability and handling complex, high-dimensional data.

Improved HOPNET Routing Protocol Using the Bee Colony Algorithm (Bee-HOPNET)

M. Habibi, Z. Delaram, and M. Kouchaki — 18th International Conference on Recent Research in Science and Technology, Dec. 2019 (Farsi)

This paper presents an artificial intelligence-based routing method for MANETs that uses bee colony optimization to improve the HOPNET protocol in terms of delay, routing overhead, packet delivery, and energy consumption.

A Scalable Dual-Camera Platform for Behavioral Assessment in Healthcare

This poster presents a scalable dual-camera AI platform for behavioral assessment in healthcare that combines object detection, posture recognition, multi-view tracking, and 3D triangulation to monitor patient activity and support clinical decision-making.

Video-Based Human-Object Interaction Analysis for Healthcare

This poster presents an AI-based video monitoring framework for healthcare that detects patient posture and object interactions from camera data to generate clinically useful behavior analysis and support patient monitoring.

A Camera-Based Platform for Patient Ambulation Assessment

This poster presents a camera-based AI platform for patient ambulation assessment that detects and tracks body and head movement, estimates 3D position, and classifies posture from video to support healthcare monitoring.