Funded Studies
Study Rationale: Early detection of Parksinson’s disease (PD) leads to improved care and better quality of life. However, the current methods for diagnosing PD, physical examination and brain imaging, are not particularly accurate in the early stages of the disease. Recent research has shown that subtle changes in the retina and the optic nerve occur early in PD. Measuring these changes may provide a simpler and more accurate means of early diagnosis. Although these changes can be seen in images acquired using techniques called fundus photography and optical coherence tomography (OCT), they are subtle and can be challenging to interpret.
Hypothesis: We hypothesize that we can leverage our large database of fundus photographs and OCT images to teach an artificial intelligence algorithm (AI) to detect the early signs of PD. This method could pave the way for more accurate, cost-effective and practical means for early diagnosis and intervention.
Study Design: Fundus photography and OCT are non-invasive imaging techniques that can capture structural and vascular changes in the back of the eye. We have access to four large databases of ocular images from individuals with and without PD; these databases include other health data, as well as the date of PD diagnosis. In this project, we will use advanced AI techniques to search for patterns and features in these images that can help us identify individuals with PD and predict whether someone might develop the disease in the next 5 years.
Impact on Diagnosis/Treatment of Parkinson’s disease: If successful, this project has the potential to revolutionize PD diagnosis and treatment by identifying non-invasive ocular biomarkers and developing a predictive model that could enable earlier detection, timely intervention and more personalized treatment strategies, ultimately improving quality of life for individuals with PD.
Next Steps for Development: The next steps toward clinical application would involve validating the identified ocular biomarkers and predictive models in larger, multi-center studies and, ultimately, by prospectively identifying individuals seen in eye clinics who are at risk of developing PD and providing them with additional neurological screening.