Structural Abnormalities of the Optic Nerve and Retina in Huntington's Disease Pre-Clinical and Clinical Settings

Deep Learning and Machine Learning Algorithms for Retinal Image Analysis in Neurodegenerative Disease: Systematic Review of Datasets and Models

Purpose

Retinal images contain rich biomarker information for neurodegenerative disease. Recently, deep learning models have been used for automated neurodegenerative disease diagnosis and risk prediction using retinal images with good results.

Methods

In this review, we systematically report studies with datasets of retinal images from patients with neurodegenerative diseases, including Alzheimer's disease, Huntington's disease, Parkinson's disease, amyotrophic lateral sclerosis, and others. We also review and characterize the models in the current literature which have been used for classification, regression, or segmentation problems using retinal images in patients with neurodegenerative diseases.

Results

Our review found several existing datasets and models with various imaging modalities primarily in patients with Alzheimer's disease, with most datasets on the order of tens to a few hundred images. We found limited data available for the other neurodegenerative diseases. Although cross-sectional imaging data for Alzheimer's disease is becoming more abundant, datasets with longitudinal imaging of any disease are lacking.

Conclusions

The use of bilateral and multimodal imaging together with metadata seems to improve model performance, thus multimodal bilateral image datasets with patient metadata are needed. We identified several deep learning tools that have been useful in this context including feature extraction algorithms specifically for retinal images, retinal image preprocessing techniques, transfer learning, feature fusion, and attention mapping. Importantly, we also consider the limitations common to these models in real-world clinical applications.

Translational Relevance

This systematic review evaluates the deep learning models and retinal features relevant in the evaluation of retinal images of patients with neurodegenerative disease.

Characterizing differences in retinal and choroidal microvasculature and structure in individuals with Huntington's Disease compared to healthy controls: A cross-sectional prospective study

OBJECTIVE

To characterize retinal and choroidal microvascular and structural changes in patients who are gene positive for mutant huntingtin protein (mHtt) with symptoms of Huntington's Disease (HD).

METHODS

This study is a cross-sectional comparison of patients who are gene positive for mHtt and exhibit symptoms of HD, either motor manifest or prodromal (HD group), and cognitively normal individuals without a family history of HD (control group). HD patients were diagnosed by Duke movement disorder neurologists based on the Unified Huntington's Disease Rating Scale (UHDRS). Fovea and optic nerve centered OCT and OCTA images were captured using Zeiss Cirrus HD-5000 with AngioPlex. Outcome metrics included central subfield thickness (CST), peripapillary retinal nerve fiber layer (pRNFL) thickness, ganglion cell-inner plexiform layer (GCIPL) thickness, and choroidal vascularity index (CVI) on OCT, and foveal avascular zone (FAZ) area, vessel density (VD), perfusion density (PD), capillary perfusion density (CPD), and capillary flux index (CFI) on OCTA. Generalized estimating equation (GEE) models were used to account for inter-eye correlation.

RESULTS

Forty-four eyes of 23 patients in the HD group and 77 eyes of 39 patients in the control group were analyzed. Average GCIPL thickness and FAZ area were decreased in the HD group compared to controls (p = 0.001, p < 0.001). No other imaging metrics were significantly different between groups.

CONCLUSIONS

Patients in the HD group had decreased GCIPL thickness and smaller FAZ area, highlighting the potential use of retinal biomarkers in detecting neurodegenerative changes in HD.

Retina Oculomics in Neurodegenerative Disease

Ophthalmic biomarkers have long played a critical role in diagnosing and managing ocular diseases. Oculomics has emerged as a field that utilizes ocular imaging biomarkers to provide insights into systemic diseases. Advances in diagnostic and imaging technologies including electroretinography, optical coherence tomography (OCT), confocal scanning laser ophthalmoscopy, fluorescence lifetime imaging ophthalmoscopy, and OCT angiography have revolutionized the ability to understand systemic diseases and even detect them earlier than clinical manifestations for earlier intervention. With the advent of increasingly large ophthalmic imaging datasets, machine learning models can be integrated into these ocular imaging biomarkers to provide further insights and prognostic predictions of neurodegenerative disease. In this manuscript, we review the use of ophthalmic imaging to provide insights into neurodegenerative diseases including Alzheimer Disease, Parkinson Disease, Amyotrophic Lateral Sclerosis, and Huntington Disease. We discuss recent advances in ophthalmic technology including eye-tracking technology and integration of artificial intelligence techniques to further provide insights into these neurodegenerative diseases. Ultimately, oculomics opens the opportunity to detect and monitor systemic diseases at a higher acuity. Thus, earlier detection of systemic diseases may allow for timely intervention for improving the quality of life in patients with neurodegenerative disease.

Retinal thickness as a biomarker of cognitive impairment in manifest Huntington's disease

BACKGROUND

Cognitive decline has been reported in premanifest and manifest Huntington's disease but reliable biomarkers are lacking. Inner retinal layer thickness seems to be a good biomarker of cognition in other neurodegenerative diseases.

OBJECTIVE

To explore the relationship between optical coherence tomography-derived metrics and global cognition in Huntington's Disease.

METHODS

Thirty-six patients with Huntington's disease (16 premanifest and 20 manifest) and 36 controls matched by age, sex, smoking status, and hypertension status underwent macular volumetric and peripapillary optical coherence tomography scans. Disease duration, motor status, global cognition and CAG repeats were recorded in patients. Group differences in imaging parameters and their association with clinical outcomes were analyzed using linear mixed-effect models.

RESULTS

Premanifest and manifest Huntington's disease patients presented thinner retinal external limiting membrane-Bruch's membrane complex, and manifest patients had thinner temporal peripapillary retinal nerve fiber layer compared to controls. In manifest Huntington's disease, macular thickness was significantly associated with MoCA scores, inner nuclear layer showing the largest regression coefficients. This relationship was consistent after adjusting for age, sex, and education and p-value correction with False Discovery Rate. None of the retinal variables were related to Unified Huntington's Disease Rating Scale score, disease duration, or disease burden. Premanifest patients did not show a significant association between OCT-derived parameters and clinical outcomes in corrected models.

CONCLUSIONS

In line with other neurodegenerative diseases, OCT is a potential biomarker of cognitive status in manifest HD. Future prospective studies are needed to evaluate OCT as a potential surrogate marker of cognitive decline in HD.

Myelinosome organelles in pathological retinas: ubiquitous presence and dual role in ocular proteostasis maintenance

The timely and efficient elimination of aberrant proteins and damaged organelles, formed in response to various genetic and environmental stressors, is a vital need for all cells of the body. Recent lines of evidence point out several non-classical strategies employed by ocular tissues to cope with aberrant constituents generated in the retina and in the retinal pigmented epithelium cells exposed to various stressors. Along with conventional strategies relying upon the intracellular degradation of aberrant constituents through ubiquitin-proteasome and/or lysosome-dependent autophagy proteolysis, two non-conventional mechanisms also contribute to proteostasis maintenance in ocular tissues. An exosome-mediated clearing and a myelinosome-driven secretion mechanism do not require intracellular degradation but provide the export of aberrant constituents and "waste proteins" outside of the cells. The current review is centered on the non-degradative myelinosome-driven secretion mechanism, which operates in the retina of transgenic Huntington's disease R6/1 model mice. Myelinosome-driven secretion is supported by rare organelles myelinosomes that are detected not only in degenerative Huntington's disease R6/1 retina but also in various pathological states of the retina and of the retinal pigmented epithelium. The intra-retinal traffic and inter-cellular exchange of myelinosomes was discussed in the context of a dual role of the myelinosome-driven secretion mechanism for proteostasis maintenance in different ocular compartments. Special focus was made on the interplay between degradative and non-degradative strategies in ocular pathophysiology, to delineate potential therapeutic approaches to counteract several vision diseases.