Hyperspectral Imaging and the Retina: Worth the Wave?

Hyperspectral retinal imaging biomarkers of ocular and systemic diseases

Hyperspectral imaging is a frontier in the field of medical imaging technology. It enables the simultaneous collection of spectroscopic and spatial data. Structural and physiological information encoded in these data can be used to identify and localise typically elusive biomarkers. Studies of retinal hyperspectral imaging have provided novel insights into disease pathophysiology and new ways of non-invasive diagnosis and monitoring of retinal and systemic diseases. This review provides a concise overview of recent advances in retinal hyperspectral imaging.

Retinal optical coherence tomography intensity spatial correlation features as new biomarkers for confirmed Alzheimer's disease

BACKGROUND

The nature and severity of Alzheimer's disease (AD) pathologies in the retina and brain correspond. However, retinal biomarkers need to be validated in clinical cohorts with confirmed AD biomarkers and optical coherence tomography (OCT). The main objective of this study was to investigate whether retinal metrics measured by OCT aid in the early screening and brain pathology monitoring for confirmed AD.

METHODS

This was a case-control study. All participants underwent retinal OCT imaging, and neurological examinations, including amyloid-β (Aβ) positron emission tomography. Participants were subdivided into cognitively normal (CN), mild cognitive impairment (MCI), and AD-derived dementia (ADD). Except retinal thickness, we developed the grey level co-occurrence matrix algorithm to extract retinal OCT intensity spatial correlation features (OCT-ISCF), including angular second matrix (ASM), correlation (COR), and homogeneity (HOM), one-way analysis of variance was used to compare the differences in retinal parameters among the groups, and to analyze the correlation with brain Aβ plaques and cognitive scores. The repeatability and robustness of OCT-ISCF were evaluated using experimental and simulation methods.

RESULTS

This study enrolled 82 participants, subdivided into 20 CN, 22 MCI, and 40 ADD. Compared with the CN, the thickness of retinal nerve fiber layer and myoid and ellipsoid zone were significantly thinner (P < 0.05), and ASM, COR, and HOM in several retinal sublayers changed significantly in the ADD (P < 0.05). Notably, the MCI showed significant differences in ASM and COR in the outer segment of photoreceptor compared with the CN (P < 0.05). The changing pattern of OCT-ISCF with interclass correlation coefficients above 0.8 differed from that caused by speckle noise, and was affected by OCT image quality index. Moreover, the retinal OCT-ISCF were more strongly correlated with brain Aβ plaque burden and MoCA scores than retinal thickness. The accuracy using retinal OCT-ISCF (AUC = 0.935, 0.830) was better than that using retinal thickness (AUC = 0.795, 0.705) in detecting ADD and MCI.

CONCLUSIONS

The study demonstrates that retinal OCT-ISCF enhance the association and detection efficacy of AD pathology compared to retinal thickness, suggesting retinal OCT-ISCF have the potential to be new biomarkers for AD.

Rationale and design of the BeyeOMARKER study: prospective evaluation of blood- and eye-based biomarkers for early detection of Alzheimer's disease pathology in the eye clinic

BACKGROUND

Alzheimer's disease (AD) is a common, complex and multifactorial disease that may require screening across multiple routes of referral to enable early detection and subsequent future implementation of tailored interventions. Blood- and eye-based biomarkers show promise as low-cost, scalable and patient-friendly tools for early AD detection given their ability to provide information on AD pathophysiological changes and manifestations in the retina, respectively. Eye clinics provide an intriguing real-world proof-of-concept setting to evaluate the performance of these potential AD screening tools given the intricate connections between the eye and brain, presumed enrichment for AD pathology in the aging population with eye disorders, and the potential for an accelerated diagnostic pathway for under-recognized patient groups.

METHODS

The BeyeOMARKER study is a prospective, observational, longitudinal cohort study aiming to include individuals visiting an eye-clinic. Inclusion criteria entail being ≥ 50 years old and having no prior dementia diagnosis. Excluded eye-conditions include traumatic insults, superficial inflammation, and conditions in surrounding structures of the eye that are not engaged in vision. The BeyeOMARKER cohort (n = 700) will undergo blood collection to assess plasma p-tau217 levels and a brief cognitive screening at the eye clinic. All participants will subsequently be invited for annual longitudinal follow-up including remotely administered cognitive screening and questionnaires. The BeyeOMARKER + cohort (n = 150), consisting of 100 plasma p-tau217 positive participants and 50 matched negative controls selected from the BeyeOMARKER cohort, will additionally undergo Aβ-PET and tau-PET, MRI, retinal imaging including hyperspectral imaging (primary), widefield imaging, optical coherence tomography (OCT) and OCT-Angiography (secondary), and cognitive and cortical vision assessments.

RESULTS

We aim to implement the current protocol between April 2024 until March 2027. Primary outcomes include the performance of plasma p-tau217 and hyperspectral retinal imaging to detect AD pathology (using Aβ- and tau-PET visual read as reference standard) and to detect cognitive decline. Initial follow-up is ~ 2 years but may be extended with additional funding.

CONCLUSIONS

We envision that the BeyeOMARKER study will demonstrate the feasibility of early AD detection based on blood- and eye-based biomarkers in alternative screening settings, and will improve our understanding of the eye-brain connection.

TRIAL REGISTRATION

The BeyeOMARKER study (Eudamed CIV ID: CIV-NL-23-09-044086; registration date: 19th of March 2024) is approved by the ethical review board of the Amsterdam UMC.

Fluorescence excitation-scanning hyperspectral imaging with scalable 2D-3D deep learning framework for colorectal cancer detection

Colorectal cancer is one of the top contributors to cancer-related deaths in the United States, with over 100,000 estimated cases in 2020 and over 50,000 deaths. The most common screening technique is minimally invasive colonoscopy using either reflected white light endoscopy or narrow-band imaging. However, current imaging modalities have only moderate sensitivity and specificity for lesion detection. We have developed a novel fluorescence excitation-scanning hyperspectral imaging (HSI) approach to sample image and spectroscopic data simultaneously on microscope and endoscope platforms for enhanced diagnostic potential. Unfortunately, fluorescence excitation-scanning HSI datasets pose major challenges for data processing, interpretability, and classification due to their high dimensionality. Here, we present an end-to-end scalable Artificial Intelligence (AI) framework built for classification of excitation-scanning HSI microscopy data that provides accurate image classification and interpretability of the AI decision-making process. The developed AI framework is able to perform real-time HSI classification with different speed/classification performance trade-offs by tailoring the dimensionality of the dataset, supporting different dimensions of deep learning models, and varying the architecture of deep learning models. We have also incorporated tools to visualize the exact location of the lesion detected by the AI decision-making process and to provide heatmap-based pixel-by-pixel interpretability. In addition, our deep learning framework provides wavelength-dependent impact as a heatmap, which allows visualization of the contributions of HSI wavelength bands during the AI decision-making process. This framework is well-suited for HSI microscope and endoscope platforms, where real-time analysis and visualization of classification results are required by clinicians.

Distinct retinal reflectance spectra from retinal hyperspectral imaging in Parkinson's disease

BACKGROUND

Recent developments in the retinal hyperspectral imaging method have indicated its potential in addressing challenges posed by neurodegenerative disorders, such as Alzheimer's disease. This human clinical study is the first to assess reflectance spectra obtained from this imaging as a tool for diagnosing patients with Parkinson's disease (PD).

METHODS

Retinal hyperspectral imaging was conducted on a total of 40 participants, including 20 patients with PD and 20 controls. Following preprocessing, retinal reflectance spectra were computed for the macular retina defined by four rectangular regions. Linear discriminant analysis classifiers underwent training to discern patients with PD from control participants. To assess the performance of the selected features, nested leave-one-out cross-validation was employed using machine learning. The indicated values include the area under the curve (AUC) and the corresponding 95% confidence interval (CI).

RESULTS

Retinal reflectance spectra of PD patients exhibited variations in the spectral regions, particularly at shorter wavelengths (superonasal retina, wavelength < 490 nm; inferonasal retina, wavelength < 510 nm) when compared to those of controls. Retinal reflectance spectra yielded an AUC of 0.60 (95% CI: 0.43-0.78) and 0.60 (95% CI: 0.43-0.78) for the superonasal and inferonasal retina, respectively, distinguishing individuals with and without PD.

CONCLUSION

Reflectance spectra obtained from retinal hyperspectral imaging tended to decrease at shorter wavelengths across a broad spectral range in PD patients. Further investigations building upon these preliminary findings are imperative to focus on the retinal spectral signatures associated with PD pathological hallmarks, including α-synuclein.

Hyperspectral Imaging Detects Clitoral Vascular Issues in Gender-Affirming Surgery

The aim of this study was to assess the efficacy of hyperspectral imaging (HSI) as an intraoperative perfusion imaging modality during gender affirmation surgery (GAS). The hypothesis posited that HSI could quantify perfusion to the clitoral complex, thereby enabling the prediction of either uneventful wound healing or the occurrence of necrosis. In this non-randomised prospective clinical study, we enrolled 30 patients who underwent GAS in the form of vaginoplasty with the preparation of a clitoral complex from 2020 to 2024 and compared patients' characteristics as well as HSI data regarding clitoris necrosis. Individuals demonstrating uneventful wound healing pertaining to the clitoral complex were designated as Group A. Patients with complete necrosis of the neo-clitoris were assigned to Group B. Patient characteristics were collected and subsequently a comparative analysis carried out. No significant difference in patient characteristics was observed between the two groups. Necrosis occurred when both StO2 and NIR PI parameters fell below 40%. For the simultaneous occurrence of StO2 and NIR PI of 40% or less, a sensitivity of 92% and specificity of 72% was calculated. Intraoperatively, the onset of necrosis in the clitoral complex can be reliably predicted with the assistance of HSI.

Naevi Characterization Using Hyperspectral Imaging: A Pilot Study

PURPOSE

The prevalence of choroidal naevi is common and has been found to be up to 10%. Little is known regarding the optical properties of choroidal naevi. A novel hyperspectral eye fundus camera was used to investigate choroidal naevi's optical density spectra in the retina.

METHODS

In an ophthalmology clinic setting, patients with choroidal naevi were included in the study. Visual acuity and pressure were tested. Following mydriatics, optical coherence tomography and fundus photography were taken as a reference, after which a hyperspectral image with 12 nm spectral resolution at 450-700 nm was taken. The optical density spectra was measured across the area of the naevus.

RESULTS

Nine patients with 11 naevi were examined. The visual acuity was not affected by any of the naevi. All the naevi were flat as measured either with the optical coherence tomography and/or on inspection, and only one naevi had a risk factor (orange pigmentation). The Wasserstein distance between the background and the naevi was higher at 695 nm compared to 555 nm (p = .002). The naevi could be grouped into three clusters based on the extracted optical density spectra.

CONCLUSION

Choroidal naevi are better visible in longer wavelengths compared to shorter wavelengths. This finding can be used to contour and follow choroidal naevi. Choroidal naevi expose different optical density spectra that can be grouped into three different clusters. One of these clusters has an optical density spectra resembling the absorption spectra of lipofuscin, which may indicate the content of this pigment.

Hyperspectral Retinal Imaging as a Non-Invasive Marker to Determine Brain Amyloid Status

Background

As an extension of the central nervous system (CNS), the retina shares many similarities with the brain and can manifest signs of various neurological diseases, including Alzheimer's disease (AD).

Objective

To investigate the retinal spectral features and develop a classification model to differentiate individuals with different brain amyloid levels.

Methods

Sixty-six participants with varying brain amyloid-β protein levels were non-invasively imaged using a hyperspectral retinal camera in the wavelength range of 450-900 nm in 5 nm steps. Multiple retina features from the central and superior views were selected and analyzed to identify their variability among individuals with different brain amyloid loads.

Results

The retinal reflectance spectra in the 450-585 nm wavelengths exhibited a significant difference in individuals with increasing brain amyloid. The retinal features in the superior view showed higher inter-subject variability. A classification model was trained to differentiate individuals with varying amyloid levels using the spectra of extracted retinal features. The performance of the spectral classification model was dependent upon retinal features and showed 0.758-0.879 accuracy, 0.718-0.909 sensitivity, 0.764-0.912 specificity, and 0.745-0.891 area under curve for the right eye.

Conclusions

This study highlights the spectral variation of retinal features associated with brain amyloid loads. It also demonstrates the feasibility of the retinal hyperspectral imaging technique as a potential method to identify individuals in the preclinical phase of AD as an inexpensive alternative to brain imaging.

A High-Resolution Hyperspectral Imaging System for the Retina

In this study, we developed an imaging system that can acquire and produce high-resolution hyperspectral images of the retina. Our system combines the view from a high-resolution RGB camera and a snapshot hyperspectral camera together. The method is fast and can be constructed into a compact imaging device. We tested our system by imaging a calibrated color chart, biological tissues ex vivo, and a phantom of the human retina. By using image pansharpening methods, we were able to produce a high-resolution hyperspectral image. The images from the hyperspectral camera alone have a spatial resolution of 0.2 mm/pixel, whereas the pansharpened images have a spatial resolution of 0.1 mm/pixel, a 2x increase in spatial resolution. Our method has the potential to capture images of the retina rapidly. Our method preserves both the spatial and spectral fidelity, as shown by comparing the original hyperspectral images with the pansharpened images. The high-resolution hyperspectral imaging device can have a variety of applications in retina examinations.

Potential Retinal Biomarkers in Alzheimer's Disease

Alzheimer's disease (AD) represents a major diagnostic challenge, as early detection is crucial for effective intervention. This review examines the diagnostic challenges facing current AD evaluations and explores the emerging field of retinal alterations as early indicators. Recognizing the potential of the retina as a noninvasive window to the brain, we emphasize the importance of identifying retinal biomarkers in the early stages of AD. However, the examination of AD is not without its challenges, as the similarities shared with other retinal diseases introduce complexity in the search for AD-specific markers. In this review, we address the relevance of using the retina for the early diagnosis of AD and the complex challenges associated with the search for AD-specific retinal biomarkers. We provide a comprehensive overview of the current landscape and highlight avenues for progress in AD diagnosis by retinal examination.

Retinal electrophysiology in central nervous system disorders. A review of human and mouse studies

The retina and brain share similar neurochemistry and neurodevelopmental origins, with the retina, often viewed as a "window to the brain." With retinal measures of structure and function becoming easier to obtain in clinical populations there is a growing interest in using retinal findings as potential biomarkers for disorders affecting the central nervous system. Functional retinal biomarkers, such as the electroretinogram, show promise in neurological disorders, despite having limitations imposed by the existence of overlapping genetic markers, clinical traits or the effects of medications that may reduce their specificity in some conditions. This narrative review summarizes the principal functional retinal findings in central nervous system disorders and related mouse models and provides a background to the main excitatory and inhibitory retinal neurotransmitters that have been implicated to explain the visual electrophysiological findings. These changes in retinal neurochemistry may contribute to our understanding of these conditions based on the findings of retinal electrophysiological tests such as the flash, pattern, multifocal electroretinograms, and electro-oculogram. It is likely that future applications of signal analysis and machine learning algorithms will offer new insights into the pathophysiology, classification, and progression of these clinical disorders including autism, attention deficit/hyperactivity disorder, bipolar disorder, schizophrenia, depression, Parkinson's, and Alzheimer's disease. New clinical applications of visual electrophysiology to this field may lead to earlier, more accurate diagnoses and better targeted therapeutic interventions benefiting individual patients and clinicians managing these individuals and their families.

Using Ensemble OCT-Derived Features beyond Intensity Features for Enhanced Stargardt Atrophy Prediction with Deep Learning

Stargardt disease is the most common form of juvenile-onset macular dystrophy. Spectral-domain optical coherence tomography (SD-OCT) imaging provides an opportunity to directly measure changes to retinal layers due to Stargardt atrophy. Generally, atrophy segmentation and prediction can be conducted using mean intensity feature maps generated from the relevant retinal layers. In this paper, we report an approach using advanced OCT-derived features to augment and enhance data beyond the commonly used mean intensity features for enhanced prediction of Stargardt atrophy with an ensemble deep learning neural network. With all the relevant retinal layers, this neural network architecture achieves a median Dice coefficient of 0.830 for six-month predictions and 0.828 for twelve-month predictions, showing a significant improvement over a neural network using only mean intensity, which achieved Dice coefficients of 0.744 and 0.762 for six-month and twelve-month predictions, respectively. When using feature maps generated from different layers of the retina, significant differences in performance were observed. This study shows promising results for using multiple OCT-derived features beyond intensity for assessing the prognosis of Stargardt disease and quantifying the rate of progression.

Imaging Amyloid and Tau in the Retina: Current Research and Future Directions

BACKGROUND

The retina is a key focus in the search for biomarkers of Alzheimer's disease (AD) because of its accessibility and shared development with the brain. The pathological hallmarks of AD, amyloid beta (Aβ), and hyperphosphorylated tau (pTau) have been identified in the retina, although histopathologic findings have been mixed. Several imaging-based approaches have been developed to detect retinal AD pathology in vivo. Here, we review the research related to imaging AD-related pathology in the retina and implications for future biomarker research.

EVIDENCE ACQUISITION

Electronic searches of published literature were conducted using PubMed and Google Scholar.

RESULTS

Curcumin fluorescence and hyperspectral imaging are both promising methods for detecting retinal Aβ, although both require validation in larger cohorts. Challenges remain in distinguishing curcumin-labeled Aβ from background fluorescence and standardization of dosing and quantification methods. Hyperspectral imaging is limited by confounding signals from other retinal features and variability in reflectance spectra between individuals. To date, evidence of tau aggregation in the retina is limited to histopathologic studies. New avenues of research are on the horizon, including near-infrared fluorescence imaging, novel Aβ labeling techniques, and small molecule retinal tau tracers. Artificial intelligence (AI) approaches, including machine learning models and deep learning-based image analysis, are active areas of investigation.

CONCLUSIONS

Although the histopathological evidence seems promising, methods for imaging retinal Aβ require further validation, and in vivo imaging of retinal tau remains elusive. AI approaches may hold the greatest promise for the discovery of a characteristic retinal imaging profile of AD. Elucidating the role of Aβ and pTau in the retina will provide key insights into the complex processes involved in aging and in neurodegenerative disease.

Multispectral retinal video-ophthalmoscope with fiber optic illumination

Multispectral imaging is used in various applications including astronomy, industry and agriculture. In retinal imaging, the single-shot multispectral image stack is typically acquired and analyzed. This multispectral analysis can provide information on various structural or metabolic properties. This paper describes the multispectral improvement of a video-ophthalmoscope, which can acquire retinal video sequences of the optic nerve head and peripapillary area using various spectral light illumination. The description of the multispectral video imaging is provided and several applications are described. These applications include multispectral retinal photoplethysmography, visualization of spontaneous vein pulsation and multispectral RGB image generation.

Retinal imaging biomarkers of neurodegenerative diseases

The timely detection of neurodegenerative diseases is central to improving clinical care as well as enabling the development and deployment of disease-modifying therapies. Retinal imaging is emerging as a method to detect features of a number of neurodegenerative diseases, given the anatomical and functional similarities between the retina and the brain. This review provides an overview of the current status of retinal imaging biomarkers of neurodegenerative diseases including Alzheimer's disease, Parkinson's disease, Lewy body dementia, frontotemporal dementia, Huntington's disease and multiple sclerosis. Whilst research findings are promising, efforts to harmonise study designs and imaging methods will be important in translating these findings into clinical care. Doing so may mean that eye care providers will play important roles in the detection of a variety of neurodegenerative diseases in future.

Current and novel multi-imaging modalities to assess retinal oxygenation and blood flow

Retinal ischemia characterizes the underlying pathology in a multitude of retinal diseases that can ultimately lead to vision loss. A variety of novel imaging modalities have been developed to characterize retinal ischemia by measuring retinal oxygenation and blood flow in-vivo. These technologies offer valuable insight into the earliest pathophysiologic changes within the retina and provide physicians and researchers with new diagnostic and monitoring capabilities. Future retinal imaging technologies with the capability to provide affordable, noninvasive, and comprehensive data on oxygen saturation, vasculature, and blood flow mechanics are needed. This review will highlight current and future trends in multimodal imaging to assess retinal blood flow and oxygenation.