Retinal texture biomarkers may help to discriminate between Alzheimer’s, Parkinson’s, and healthy controls

AI Generated Summary*

Texture analysis of optical coherence tomography scans was tested for separating eyes of Alzheimer's disease (AD) and Parkinson's disease (PD) patients from healthy controls. Retrospective macular scans from 20 AD, 28 PD and 27 age-matched controls at one centre yielded fundus images for six inner retinal layers. Grey level co-occurrence matrix and complex wavelet features fed support vector machine classifiers combined by one-versus-one voting with k-fold cross-validation. Full retinal thickness showed no significant group differences across nine macular regions. Median per-eye sensitivity reached 88.7% for controls, 79.5% for AD and 77.8% for PD, best with the nerve fibre layer excluded. Among participants labelled identically in both eyes, correct classification medians ran from 91.4% (2-fold) to 94.4% (10-fold).

*This summary was generated by AI and is published unedited. Oku does not alter these summaries. It may contain errors or omissions and is provided for general informational purposes only. Oku does not guarantee its accuracy, completeness, or reliability. For authoritative information, please refer to the original, peer-reviewed article.