Objective: To develop an algorithm that can predict Hoehn and Yahr (H–Y) and Unified Parkinson’s Disease Rating Scale part III (UPDRS-III) scores using fundus photography among PD patients.

Background: Until now, other than complex neurological tests, there have been no readily accessible and reliable indicators of neurologic dysfunction among Parkinson’s disease (PD) patients. Thus, we conducted a study to determine the role of fundus photography as a noninvasive and readily available tool for assessing neurologic dysfunction among PD patients using deep learning methods.

Method: Fundus photographs of a total of 615 participants (266 participants with PD and 349 participants with atypical motor abnormality) who visited the neurology department of Kangbuk Samsung Hospital from August 2020 to April 2021 were analyzed in this study. A convolutional neural network was developed to predict both the H–Y and UPDRS-III scores based on fundus photography findings and participants’ demographic characteristics. The area under the receiver operating characteristic curve (AUC) was calculated for sensitivity and specificity analyses for both the internal and external validation datasets.

Results: For the internal validation dataset, the sensitivity was 83.23% (95% confidence interval (CI), 82.07%–84.38%) and 82.61% (95% CI 81.38%–83.83%) for the H–Y and UPDRS-III scores, respectively. The specificity was 66.81% (95% CI 64.97%–68.65%) and 65.75% (95% CI 62.56%–68.94%) for the H–Y and UPDRS-III scores, respectively. For the external validation dataset, the sensitivity and specificity were 70.73% and 66.66%, respectively. Lastly, the calculated AUC and ACC were 0.67 and 70.45%, respectively.

Conclusion: Our study offers amalgamative insights into the neurological dysfunction among PD patients by providing information on how to apply a deep learning method to evaluate the association between the retina and brain. Our study data might help clarify recent research findings regarding dopamine pathologic cascades between the retina and brain among PD patients. However, further research is needed to expand the clinical implication of this algorithm.

« Back to 2023 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/neurologic-dysfunction-assessment-in-parkinsons-disease-based-on-fundus-photographs-using-deep-learning/