Towards Automated Eye Movement Characterization for Stroke Patients Using Synthetic Video Data and Machine Learning
Paper in proceeding, 2025
Stroke is a critical medical emergency that can cause permanent disability or death. Rapid identification of stroke, especially in prehospital settings, is crucial for timely treatment. Video analysis and machine learning (ML) could facilitate the prehospital assessment of stroke, but a lack of video data from stroke patients remain a barrier to developing effective models. This study explores the use of synthetic data to develop ML models, generating 73 videos mimicking characteristic eye movements of stroke patients through 3D modeling and animation. Four ML models were developed. Long short-term memory (LSTM) and gated recurrent units (GRU) achieved the best performance (over 84% in accuracy, precision, sensitivity, specificity and F1-Score). These findings highlight the promise of synthetic data for developing ML models for healthcare applications and the potential of ML-driven video analysis in the automated assessment of stroke-related eye movements, supporting advancements in prehospital stroke care.
synthetic data
video analysis
eye movement
machine learning
Stroke
Author
Hoor Jalo
Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering
Eddie Ström
Student at Chalmers
Samuel Ollila
Student at Chalmers
Robin Khatiri
Student at Chalmers
Jacob Westerberg
Student at Chalmers
Teodor Svensson
Student at Chalmers
Petra Redfors
Sahlgrenska University Hospital
University of Gothenburg
Katarina Jood
University of Gothenburg
Sahlgrenska University Hospital
Bengt-Arne Sjöqvist
Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering
Stefan Candefjord
Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering
Studies in health technology and informatics
09269630 (ISSN) 18798365 (eISSN)
Vol. 329 951-955
20th World Congress on Medical and Health Informatics, MEDINFO 2025
Taipei, Taiwan,
Taipei, Taiwan,
Subject Categories (SSIF 2025)
Neurosciences
Neurology
DOI
10.3233/SHTI250980
PubMed
40775998