Mohamed Abul Hassan, Ph.D.

Biography

Mohamed Abul Hassan is a Project Scientist with the Biomedical Engineering Department at the University of California, Davis.  His research aims to improve medical decision-making, enhance patient outcomes, and streamline healthcare processes using machine learning, computer vision algorithms. This extends to working with novel medical imaging modalities and unstructured clinical data, analyzing patterns, and providing recommendations to healthcare providers. In the past eight years, he has worked on unique healthcare projects addressing clinical needs for surgical guidance, disease diagnosis, and remote health monitoring.

Mohamed believes that healthcare should be a universal enabler, transcending barriers and making health services accessible to all, regardless of their geographic, economic, or social status.

Interests 

Education 

🎓 Ph.D. in Electronic Engineering, 2018

🎓 M.S. in Electronic Engineering, 2015

🎓 B.E. in Electrical and Electronic                 Engineering, 2012


Updates


Publications

Augmenting Early Stroke Diagnosis with an Eye-Tracker

This study introduces an innovative diagnostic tool that utilizes a machine learning algorithm-driven eye tracker to enhance early diagnosis of PCS. 

 
Preprint: https://www.researchsquare.com/article/rs-4656842/v1

Automatic Recognition of Food Ingestion Environment Using Egocentric Camera

This work looks at the ingestion environment which is very insightful for dietary assessment and recognizing patterns. We used a neural network with a two-stage training framework based on transfer learning from the Place365 Database.
Oral Presentation: CVPR MetaFood Workshop 2024

Paper | Cite

FLIm Augmented Semi-Supervised Model to Detect Residual Tumor during TORS

We proposed to overcome the limitation of IFSA that fails 50% in identifying positive surgical margins during TORS by using FLIm with an anomaly detection classification model to detect residual tumor intraoperatively InVivo
Oral Presentation: MICCAI 2023

Paper | Supplement | Cite

Webcamera-Based Neurological Eye Examination 

This study takes a step toward AI-assisted stroke diagnosis for telehealth. We developed to detect the pupil center to estimate clinically relevant conjugate eye movement without the need for calibration or sophisticated eye trackers. 
Published: IEEE Journal of Biomedical and Health Informatics 

Paper | Code | Dataset | Cite

Using Biological Variables to Improve Tumor Classification 

We enhanced the classification efficacy of the FLIm augmented classification model by accounting for the diversity of the tissues within the oral cavity and the oropharynx. 
Published: IEEE Transactions on Biomedical Engineering

Paper | Cite

Non-calibrated Eye Tracker for Detecting Abnormal Eye Movement 

Here we overcame the limitation of eye trackers for clinical use by adopting a non-calibrated approach to measure clinically relevant eye movement information between patients and controls.  
Published: IEEE Transactions on Biomedical Engineering (Featured Article) 

Paper | Code | Dataset | Cite

Estimating Driver's Vital Signs using Webcamera

Non-contact estimation of heart rate is a challenging task under low illumination conditions. Here we developed an algorithm to estimate heart rate based on ballistocardiography head motion.  




Paper | Cite

Estimating Heart Rate and Breathing Rate using Webcamera

Here we developed an algorithm based on maximizing quasi-periodic information to estimate vital signals from a facial video under natural conditions. 
Published: Biomedical Optics Express (Featured Article) 

Paper | Code | Cite