Eye Tracking & Computer Vision

Eye tracking technology monitors eye movements to gather data on how patients perceive and interact with their environment. In neurodegenerative disease research, eye tracking, particularly the study of saccades (rapid eye movements), can help us detect signs of disease often before more general symptoms emerge. Subtle alterations in saccade patterns can indicate early signs of diseases like Parkinson's and PSP. Our AI models analyse time series data from eye tracking devices to detect these changes and monitor disease progression. For example, our hierarchical machine learning approach has shown that analysing entire saccade waveforms can accurately distinguish between Parkinson's Disease, PSP, and healthy controls.

Plot demonstrating the trajectory of pro-saccades and anti-saccades. The features that are analysed from these trajectories include angular displacement, fixation time, latency, and amplitude.

Computer vision, a field of AI focused on enabling machines to interpret and process visual information, is changing medical imaging analysis. We use computer vision techniques to analyse MRI brain scans, identifying changes in the brain associated with neurodegenerative diseases. These methods allow for precise analysis, potentially improving the accuracy of diagnoses by detecting early markers of disease progression. Additionally, computer vision can be used to track eye movements, enhancing our understanding of how patients perceive and interact with their environment. Our research uses this technique to help diagnose and classify between diseases.

Publications

Patel, S., Bredemeyer, O., FitzGerald, J. J., & Antoniades, C. A. (2024). Hierarchical Machine Learning Classification of Parkinsonian Disorders using Saccadic Eye Movements: A Development and Validation Study. https://arxiv.org/abs/2407.16063

Bredemeyer, O., Patel, S., FitzGerald, J. J., & Antoniades, C. A. (2022). Oculomotor deficits in Parkinson’s disease: Increasing sensitivity using multivariate approaches. Frontiers in Digital Health, 4. https://doi.org/10.3389/fdgth.2022.939677

Su, Z. H., Patel, S., Bredemeyer, O., FitzGerald, J. J., & Antoniades, C. A. (2022). Parkinson’s disease deficits in time perception to auditory as well as visual stimuli – A large online study. Frontiers in Neuroscience, 16. https://doi.org/10.3389/fnins.2022.995438

Lu, Z., Buchanan, T., Kennard, C., FitzGerald, J. J., & Antoniades, C. A. (2019). The effect of levodopa on saccades – Oxford Quantification in Parkinsonism study. Parkinsonism & Related Disorders, 68, 49–56. https://doi.org/10.1016/j.parkreldis.2019.09.029

FitzGerald, J. J., & Antoniades, C. A. (2016). Eye movements and deep brain stimulation. Current Opinion in Neurology, 29(1), 69. https://doi.org/10.1097/WCO.0000000000000276