Optical Coherence Tomography Split-Spectrum Amplitude-Decorrelation Optoretinography Detects Early Central Cone Photoreceptor Dysfunction in Retinal Dystrophies

Abstract

Purpose: To investigate if split-spectrum amplitude-decorrelation optoretinography (SSADOR) can detect and measure macular cone dysfunction in inherited retinal dystrophies (IRDs).

Methods: This study was a case series of participants presenting with various IRD pathologies. Participants were recruited from the Ophthalmic Genetics clinic at the Casey Eye Institute from February to August 2023. Multimodal and SSADOR imaging was obtained in all cases.

Results: We recruited nine participants, including four with macular dystrophy, one with fundus flavimaculatus, one with cone dystrophy, and three with retinitis pigmentosa. SSADOR decorrelation maps identified areas of cone functional impairment consistent with disease phenotypes. A correlation between the SSADOR signal and retinal sensitivity measured by microperimetry within the central 20° diameter area was observed. Additionally, SSADOR was able to demonstrate a decreased signal in mild cases when microperimetry measurements were still normal but subtle changes were also apparent on structural OCT.

Conclusions: SSADOR is sensitive at detecting functional changes in macular cones, even prior to abnormalities in perimetry testing. We highlight the potential benefits of this innovative technology for the early detection of cone dysfunction and their potential contributions to earlier diagnosis and more accurate monitoring of progression.

Translational Relevance: SSADOR is an innovative technology that detects early macular cone function changes, allowing for early diagnosis and precise monitoring of cone dysfunction progression. By serving as a potential clinical trial endpoint, SSADOR facilitates the translation of scientific findings into practical applications, ultimately improving patient care and outcomes.

Introduction

Inherited retinal dystrophies (IRDs) represent a collection of rare monogenic diseases that result in progressive photoreceptor degeneration.^1,2 Pathogenic variants in over 300 genes have been identified to lead to photoreceptor dysfunction by a variety of mechanisms, including loss of protein function impacting the phototransduction cascade or retinoid cycle, protein misfolding leading to endoplasmic stress, impairment of ciliary function, disruption of metabolic pathways, oxidative stress, microglial activation, and autophagy dysregulation.^3–8 

Novel treatment strategies, such as gene therapy and stem cell therapy, show promise of curing IRDs or delaying vision loss.^9–13 The success of therapeutic trials for IRDs depends not only on the ability to identify and quantify clinically meaningful outcome measures but also on being able to detect these changes over a reasonable time period. The ideal endpoint should have a high level of repeatability and reproducibility with minimal measurement error, be sensitive to change, be correlated with patient-reported outcomes, and minimize patient fatigue.^14,15 Best-corrected visual acuity is one of the most common endpoint in clinical trials. However, this primarily assesses only the foveal region and lacks information on extrafoveal and peripheral impairments. This limitation is notable in many IRDs, including retinitis pigmentosa (RP), where visual acuity can remain normal for decades¹⁶ despite a decline in dark adaptation, visual fields, and contrast sensitivity.¹⁷ Full-field stimulus testing (FST) measures the intensity necessary to stimulate the most sensitive part of the retina. It offers high sensitivity but exhibits low spatial correlation; although it correlates to clinically meaningful outcomes it has yet to be accepted by regulators as a primary endpoint.¹⁸ Perimetry, such as visual field, effectively maps extrafoveal sensitivity but relies on subjective performance, can be prone to learning or placebo effects, and imposes notable patient fatigue. Fundus-guided microperimetry provides high-resolution macular testing¹⁹ but has the same disadvantages as visual fields. Full-field electroretinography(ffERG) measures the electrical signals generated by all photoreceptor cells in the retina (120 million rods and 6 million cones); however, although it delivers a direct objective measurement of photoreceptor function, it has high variability and low spatial resolution.²⁰ It may not be sensitive enough to detect changes when smaller areas are treated, such as is commonly done in gene or stem cell therapy. Multifocal ERG measures topographical retinal sensitivity within a 50° diameter of the central visual field.²¹ It provides spatial information of central retinal function, but it also shares the same limitations as ffERG, including high variability and low spatial resolution. 

An imaging-based approach that provides objective, spatially high-resolution measurements with short imaging time and no ocular contact would be preferable over psychophysical and electrophysical tests. However, identifying biomarkers correlating with photoreceptor function has been challenging. Optical coherence tomography (OCT) has provided some surrogate measurements of retinal structure that correlate with functional measurements.^22,23 Multiple layers in the outer retina can be identified²⁴: outer nuclear layer, inner segment ellipsoid zone (EZ), interdigitation zone, and retinal pigment epithelium (RPE)–Bruch’s membrane complex. Changes in the EZ area or fundus autofluorescence (FAF) have been deemed acceptable outcome measures by the U.S. Food and Drug Administration.²⁵ However, such measurements are usually indicative of irreversible photoreceptor death, and biomarkers that can be used at earlier stages of disease would be useful. 

Early studies noted that OCT could measure light-induced outer retinal thickness changes,^26,27 but the magnitudes (∼100 nm) were small and below the imaging depth resolution (∼5 µm), thus leading to inconsistent results. More sophisticated techniques have used adaptive optics (AO) and phase-sensitive detection to characterize photoreceptor outer segment (OS) alterations (e.g., optical path length changes). Resolving and tracking single photoreceptors with state-of-the-art AO OCT–optoretinography (ORG) shows a swift, millisecond-scale shrinkage of OSs by tens of nanometers, followed by a gradual elongation lasting hundreds of milliseconds and spanning hundreds of nanometers^28–32 (Supplementary Fig. S1). ORG is measurement of the intrinsic cone optical properties in response to light stimuli.³⁰ Although the exact origin of the ORG contrast change is still under investigation, the prevailing hypothesis attributes the later swelling phase—which our split-spectrum amplitude-decorrelation optoretinography (SSADOR) approach measures—to the increased osmotic pressure associated with activation of the phototransduction cascade.³² Therefore, ORG represents a novel imaging-based approach to evaluate photoreceptor function, which is non-contact, does not rely on subjective feedback, and promises high spatial resolution.^33–38 However, AO imaging is impractical in clinical settings due to a very small scanning area and high dependence on patient cooperation. 

We previously demonstrated that OCT–SSADOR can detect and measure photoreceptor light response in healthy subjects.³⁹ Unlike AO ORG, SSADOR does not require complex and expensive optical instrument to resolve single photoreceptors. This advancement allows OCT–SSADOR to cover a much larger retinal area and facilitates clinical translation. It is worth noting that OCT–SSADOR does not directly measure OS length changes; instead, it compares OCT B-scans obtained before and after the test flash. Amplitude decorrelation is sensitive to both nanometer-scale morphological and subtle reflectivity changes.^39,40 We have shown that SSADOR decorrelation correlates with flash intensity and therefore photoreceptor response magnitude. Furthermore, we have aimed to establish the clinical application of our work by examining a spectrum of IRD patients, who manifest with macular dystrophy, cone–rod dystrophy, and retinitis pigmentosa. 

Materials and Methods

Informed consent was obtained following an approved protocol by the Institutional Review Board at Oregon Health & Science University, and the study adhered to the tenets of the Declaration of Helsinki. Nine patients who visited the ophthalmic genetics clinic at the Casey Eye Institute between February 2023 and August 2023 underwent SSADOR imaging using our prototype ultrahigh-resolution spectral-domain OCT (SD-OCT). 

The case series include patients presenting with various pathologies, such as macular dystrophy, cone dystrophy, and retinitis pigmentosa, diagnosed clinically and confirmed with genotyping. Exclusion criteria included the existence or history of any ocular pathology that might impair interpretation, significant cataract, advanced macular pathology with complete loss of photoreceptors, or inability to fixate. Cases with suboptimal SSADOR measurements due to low OCT signal (i.e., from ocular opacity) or severe eye movement were also excluded from this study. 

Measurement of best-corrected visual acuity and slit-lamp biomicroscopy were performed. Ultra-widefield fundus photography and autofluorescence were obtained with an Optos 200Tx confocal scanning laser ophthalmoscope (Optos, Dunfermline, UK). Functional assessments of cones and rods were obtained by ffERG (Diagnosys LLC, Lowell, MA). Central retinal sensitivity assessment (68-point grids covering central 20° diameter) was performed by microperimetry (macular integrity assessment [MAIA]; iCare USA, Raleigh, NC). Structural macular assessments and macular cone function assessments were obtained by the in-house ultrahigh-resolution SD-OCT instrument and SSADOR.³⁹ 

Ultrahigh-Resolution SD-OCT

The study employed a prototype high-speed, ultrahigh-resolution SD-OCT instrument, which has been described previously.^41,42 The axial resolution is 2.4 µm in air (∼1.8 µm in the retina), or about two times better than current commercial SD-OCT. The lateral resolution is approximately 10 µm. The incident power on the cornea is 2.1 mW nominal. Advancements in OCT spectrometer and line-scan camera technology enabled a high imaging A-scan rate of 250 kHz while maintaining >98-dB peak sensitivity. High imaging speed expands the SSADOR field of view while maintaining the necessary short inter-volume interval (∼0.5 second) (Supplementary Fig. S2). 

SSADOR Imaging

Functional photoreceptor imaging was performed by sequentially acquiring five repeated OCT volumes. To optimize parameters governing the A-scan sampling density and inter-volume interval, the raster scan protocol consists of 600 × 200 A-scans covering a 3 × 1-mm² retinal area. Each volume takes 0.5 second to acquire, and the entire SSADOR scan lasts 2.5 seconds (Supplementary Fig. 3A). A white light flash, lasting 0.2 second and bleaching ∼15% of exposed cone pigments, is delivered at the beginning of the third OCT volume. A fovea-centered 3 × 3-mm² area can be covered using three montage scans, which follow a 2-minute recovery period between scans but do not require change of gaze. 

Visualizing and Measuring Cone Light Response

SSADOR calculations proceeded in three phases. First, the reconstructed OCT volumes were spatially registered to a reference volume (i.e., volume 2 in the SSADOR dataset). Second, we used 4 × spectral split to calculate split-spectrum amplitude decorrelation between volume pairs V2–V4 and V2–V5 as the response signal, and volume pair V2–V1 served as the baseline. Our previous study found that the 4 × split achieved the highest decorrelation signal-to-noise ratio in normal eyes. SSADOR decorrelation also did not show significant difference between the V2–V4 and V2–V5 pairs; therefore, the ones with least motion artifact were used. Finally, the calculated voxel-wise decorrelation was integrated over a 10-µm depth range that encompassed the photoreceptor OS tip and posterior portion of the OS (Supplementary Fig. 3B). 

SSADOR can be visualized in two dimensions as a response map, analogous to the microperimetry grid plot but with much higher sampling density and topographical resolution. Alternatively, averaging SSADOR over the flashed area yields the quantitative metric mean decorrelation, which reflects an aggregated cone response. Averaging can be performed over the macula (i.e., global mean decorrelation) within each SSADOR scan field of view or by using a predefined pattern (e.g., a 150-µm diameter circle, or local mean). Averaging reduces statistical variations but trades topographical resolution. Additionally, one can correlate SSADOR local mean decorrelation with microperimetry spot sensitivity when the former is calculated at the same retinal location as the latter and using the corresponding averaging kernel size (e.g., 150-µm diameter circle for Goldmann III), thus allowing intermodality verification. 

Correlating SSADOR Local Mean Decorrelation and Retinal Sensitivity

The MAIA microperimeter provided the fundus tracking data and coordinates of the test stimuli presented. We used semiautomatic image registration (MATLAB Image Processing Toolbox; MathWorks, Natick, MA) to register the en face OCT projection to the MAIA fundus photograph. If automated registration failed, we manually selected control points (e.g., using vessel bifurcations) to estimate the image transform. The registered images were overlaid and visually inspected to verify spatial correspondence. The SSADOR local mean decorrelation was subsequently computed on the locations corresponding to the MAIA test grid and using a 150-µm-diameter circle. Two exceptions were made: (1) the sampling location was moved away from the retinal blood vessels if there was an overlap, and (2) the SSADOR measurements were excluded if OCT motion correction failed within the region (i.e., no valid SSADOR data). However, only a small number of measurements were affected (on average, one or two out of ∼16 locations per eye), and we do not expect the exceptions significantly impact the correlation analysis. 

Statistical Analysis

Descriptive statistics (correlation and Bland–Altamn plots) were used to assess the test–retest repeatability. We calculated Pearson correlation coefficients to compare SSADOR local mean decorrelation versus mesopic retinal sensitivity. 

Results

ORG Response Versus Flash Intensity

We adjusted the test flash intensity to modulate the magnitude of cone response in normal controls (five eyes of five subjects), with the bleach ranging from ∼1.9% to 15.0% of total cone opsin. SSADOR mean decorrelation, calculated over the two center flashed areas, showed strong positive correlation with respect to cone bleach fraction (i.e., response magnitude). The response curve can be approximated using an exponential function with high accuracy, achieving a coefficient of determination of R² = 0.99 (Supplementary Fig. S4). 

Test–Retest Repeatability

Five normally sighted subjects and five IRD patients had repeated SSADOR images from the same visit. The SSADOR local mean was sampled using a 150-µm-diameter circle on a rectangular grid of 500-µm spacing (i.e., comparable to the customized microperimetry grid). The correlation and Bland–Altman plots indicate that pointwise SSADOR mean decorrelation showed high test–retest repeatability. Representative SSADOR decorrelation maps from two IRD patients, acquired in the same imaging session with instrument realignment also showed good qualitative agreements between repeated scans (Supplementary Fig. S5). 

Clinical Characteristics of IRD Patients

Among the nine patients enrolled, four patients had been diagnosed with a macular dystrophy, one with fundus flavimaculatus, one with cone dystrophy, and three with retinitis pigmentosa. The lens status of all participants is shown in Supplementary Table S1. Ultrahigh-resolution SD-OCT and SSADOR were obtained in all cases. 

Case 1

A 34-year-old female was diagnosed with ABCA4-related cone dystrophy. She harbored two heterozygous variants in ABCA4. She presented complaining of mild central distortion without an impact on activities of daily living. Her visual acuity was 20/20 and retinal exam revealed only blunted foveal reflexes in both eyes (OU). FAF showed parafoveal hypoautofluorescence with a small hyperautofluorescent ring at the border. Microperimetry sensitivity remained almost normal (represented by yellow to green dots). SSADOR showed a bright round decorrelation signal around the central fixation point surrounded by a low SSADOR decorrelation signal in the parafoveal area. The degree of SSADOR signal loss was greater than expected compared to the decreased retinal sensitivity in microperimetry. Full-field ERG revealed mild to moderately decreased cone responses but with normal rod responses OU, consistent with a generalized cone dystrophy (Fig. 1). 

Figure 1.

 

Right eye of 34-year-old patient diagnosed with ABCA4-related cone dystrophy. Color fundus photography revealed blunt foveal reflex, and FAF showed a central hypoautofluorescence spot with a small ffighyperautofluorescent ring at the border. Subtle thinning in the parafoveal EZ line was observed in macular OCT. Full-field ERG revealed decreased cone response with intact rod response. The normal amplitudes and implicit times of the scotopic DA0.01 and photopic LA3.0 responses are shown as blue and red boxes, respectively, below the a-wave and above the b-wave of each trace. Macular sensitivity tested by microperimetry remained almost normal (represented by yellow and green dots). A round-shaped, high SSADOR signal at the center fixation point surrounded by a low SSADOR signal in the parafoveal area were observed. Similar to the SSADOR findings, bright projection signals from the cone OS were observed in the center area.

Case 2

A 56-year-old male was diagnosed with occult macular dystrophy that was thus far negative to genetic testing. He complained of nyctalopia and decreased distance and near vision in the past 7 years. His visual acuity was now 20/20 in the right eye and 20/25 in the left eye. The fundus examination was unremarkable. FAF in both eyes demonstrated a small hyper-autofluorescence spot in the foveal center. OCT images showed bilateral mild EZ disruption and thickening of the external limiting membrane. Microperimetry showed normal retinal sensitivity in both eyes, and the SSADOR map showed reduced decorrelation in the parafoveal area, suggesting reduced cone responses to light (Fig. 2). 

Figure 2.

 

Right eye of 56-year-old patient diagnosed with occult macular dystrophy. Fundus photography revealed unremarkable findings, whereas FAF demonstrated a small hyperautofluorescence band in the foveal center. Macular OCT showed mild EZ disruption and thickening of the external limiting membrane in the central area. Microperimetry indicated intact retinal sensitivity in both eyes, but an earlier decrease in SSADOR signal in the parafoveal area was observed. The OS projection image shows a dark signal area corresponding to the area of hyperautofluorescence band in FAF and disruption of the EZ in OCT.

Case 3

A 66-year-old female was diagnosed with macular dystrophy and harbored a heterozygous pathogenic variant in PRPH2. She complained of decreased acuity and distortion in the left eye. Her visual acuity measured 20/20 in the right eye and 20/50 in the left eye. Fundus examination revealed pigmentary mottling in the macula that corresponded to confluent regions of hypoautofluorescence and intermixed hyperautofluorescence on FAF. OCT images demonstrated diffuse outer retinal atrophy, EZ and RPE disruption with some preserved scattered islands of outer retinal structures. Both microperimetry and SSADOR effectively detected decreased macular cone function in the areas of atrophy. However, in the central fovea, where microperimetry showed retained sensitivity near fixation, the SSADOR map was able to detect severe loss of cone responses to light (Fig. 3). 

Figure 3.

 

Right eye of 66-year-old patient who harbored a heterozygous pathogenic variant in PRPH2. Pigmentary mottling was observed in the posterior pole corresponding to the confluent regions of hypoautofluorescence and intermixed hyperautofluorescence in FAF. OCT demonstrated outer retinal atrophy and EZ and RPE disruption, with some preserved islands of outer retinal structures. Microperimetry showed good retinal sensitivity around fixation areas, but the SSADOR map revealed severe loss of cone function. Full-field ERG revealed a mild decrease in rod and cone responses. The normal amplitudes and implicit times of scotopic DA0.01 and photopic LA3.0 responses are shown as blue and red boxes, respectively, below the a-wave and above the b-wave of each trace.

Cases 4 to 6

Three patients were diagnosed with CRX-related macular dystrophy, ABCA4-related fundus flavimaculatus, and biallelic MFSD8-related macular dystrophy. The first two had 20/20 visual acuity level OU, and OCT demonstrated parafoveal disruption of the EZ and RPE and outer nuclear layer thinning that preserved the central fovea. The third had visual acuity of 20/70 OU and central retinal atrophic region with EZ loss at leading edges on OCT. The ffERG in all cases revealed normal cone responses. Microperimetry showed decreased retinal sensitivity in the foveal or parafoveal region corresponding to the topographic reduction in cone responses on SSADOR (Fig. 4). 

Figure 4.

 

SSADOR map of three IRD patients with different degrees of macular atrophy. (A) CRX-related macular dystrophy patient who still showed a preserved cone EZ in the center. (B) ABCA4-related fundus flavimaculatus patient who showed scattered areas of outer retinal atrophy and EZ and RPE disruption. (C) MFSD8-related macular dystrophy patient who had severe macular atrophy in the center.

Cases 7 to 9

Three patients diagnosed with retinitis pigmentosa (two with biallelic USH2A pathogenic variants and one with mild sectoral RP and inconclusive genotyping) were imaged with SSADOR. All three patients had normal central visual acuity. The ORG signal in the central area was decreased. Although microperimetry was not available for the patient in Fig. 5A, the severity of decreased signal on SSADOR correlated to the loss of cone EZ observed in the OCT images (Fig. 5). 

Figure 5.

 

Multimodality imaging and SSADOR maps for three RP patients. (A) Inferior sectoral RP demonstrated a focal area of bone spicule and pigmentary change in the inferior retina; intact structural OCT and nearly normal SSADOR signal were observed in the center. (B, C) USH2A-related RP patients who showed scattered bone spicule and pigmentary changes in fundus photographs; macular OCT revealed outer retinal atrophy and EZ disruption preserving the center fovea. Microperimetry and SSADOR showed a high correlation of retinal sensitivity and SSADOR signal. Kinetic perimetry demonstrated superior defects in sectoral RP; pericentral ring scotoma and nasal defect sparing central vision were observed in advance RP.

ORG Compared to Microperimetry

We compared the pointwise microperimetry retinal sensitivity with corresponding SSADOR local mean decorrelation. Mean decorrelation was calculated within a 150-µm-diameter circle and at corresponding locations of the microperimetry grid. When including all points, the SSADOR versus retinal sensitivity had a Pearson correlation coefficient of 0.48 (P < 0.001; 95% confidence interval [CI], 0.33–0.61). Our results suggest that SSADOR had a high sensitivity for detecting early cone impairments, even when mesopic microperimetry sensitivity appeared normal or almost normal. When we excluded these points (i.e., measurements below the 95% CI, indicating normal sensitivity but reduced SSADOR response; open red circle in Supplementary Fig. S6), the Pearson correlation coefficient was 0.62 (P < 0.001; 95% CI, 0.48–0.72), suggesting a good correspondence between SSADOR local mean and microperimetry sensitivity. 

Discussion

In a previous study, Lassoued et al.⁴⁴ used AO-assisted OCT optoretinography to study cone dysfunction along the transition zone in three retinitis pigmentosa eyes. They detected residual, albeit reduced, cone functionality along the transition zone and distinguished three cone types (S, M, and L) based on their differential response to three different wavelength stimuli. Using AO to correct for optical aberrations of the human eye, their approach enabled the measurement of individual cone photoreceptors with high spatial resolution. However, the use of AO proved to be expensive and complex for clinical settings, and it demands specialized technical skills for image acquisition. Moreover, such instruments have a very limited field of view (e.g., 1 × 1 mm²), requiring extensive montaging to locate and capture most pathologies. 

In our study, we employed SSADOR algorithm with a prototype ultrahigh resolution SD-OCT to detect and measured cone light response. The custom-built instrument achieved approximately 2 times finer axial resolution than commercial SD-OCT, enabling the visualization of minute morphological changes of the outer retina. The prototype instrument also offered faster imaging speeds compared to standard commercial instruments. By combining ultrahigh resolution SD-OCT with SSADOR, we achieved simultaneous assessment of both structural and functional aspects. This enhancement in functional imaging occurred without requiring cellular resolution or optical phase detection. Therefore, our technique facilitated a larger field of view (e.g., 3 × 1 mm²) in comparison to prior photoreceptor functional imaging studies. A clinically relevant 3 × 3-mm² field can be achieved using only three montage scans within minutes. The SSADOR imaging protocol is also comparable to standard OCT imaging procedures, facilitating integration into the clinical workflow. 

This case series included SSADOR measurements in nine IRD patients. Overall, the eyes of these patients exhibited areas of reduced SSADOR response, where the decorrelation map revealed cone impairments consistent with clinically observed pathology. In cases of moderate or advanced cone or macular dystrophy, SSADOR decorrelation correlated well with retina sensitivity measured by microperimetry. In cases of mild disease with near normal mesopic retinal sensitivity, it has been shown that SSADOR local mean decorrelation can exhibit a more pronounced reduction than otherwise expected. Our data suggest that SSADOR has a higher sensitivity in detecting functional impairment of cones at an earlier time point; therefore, OCT–SSADOR shows promise in detecting and monitoring subtle cone impairment at early stages of the disease, which could be beneficial for future therapeutic trials. 

ORG measurements are fast and objective, have high spatial resolution, and do not require electrode contact. Combined with ultrahigh-resolution structural imaging, a joint assessment of both photoreceptor function and morphology may improve sensitivity and reliability for IRD staging and monitoring compared with either used alone. SSADOR provides essential information about impaired visual function, potentially offering a means to rescue existing photoreceptors before cone structural damage ensues. Recent advancements in gene augmentation hold promise for reversing photoreceptor dysfunction by introducing new copies of gene into the target retinal cells. This technique relies on the viability of photoreceptor cells for the replacement of gene product to regain functionality. The advantage of ORG lies in its capacity to precisely assess the remaining function of the photoreceptor cell, making it a potential biomarker for optimizing responses to gene therapy clinical trials. 

The main limitation of OCT–SSADOR lies in its susceptibility to eye movements. Effective results hinge on maintaining good central fixation to mitigate motion-related distortions and noises. Although current software can satisfactorily compensate for slower eye motion (e.g., drifting), more rapid saccades result in significant artifacts, leading to artifactually high decorrelation measurements. Mitigation strategies include implementation of hardware eye tracking and real-time scan repositioning to combat eye movement, as well as improving motion correction accuracy by using advanced algorithms such as those demonstrated by Ploner et al.⁴³ 

Our study also indicates that current SSADOR implementation has difficulty measuring severe photoreceptor function loss, representing a floor effect that may limit its utility in monitoring progression in late-stage disease. However, it is expected that advancements in OCT imaging performance and SSADOR computing algorithms can improve SSADOR detection sensitivity, thus enabling measurement of severely reduced photoreceptor response. Our current implementation uses an in-house prototype instrument, which may limit its accessibility to the wider clinical and research community. However, it is worth noting that the SSADOR algorithm is compatible with the raster scanning protocol used by essentially all commercial OCT devices. It is reasonable to expect that next-generation high-performance commercial OCT could offer sufficient imaging performance for SSADOR functional photoreceptor imaging, especially when component costs fall as the technology matures. 

The direction for future progress points toward extending this technology to evaluate functionality across a range of excitable retinal cell types, including rod photoreceptors, retinal ganglion cells, or neuronal cells in the inner plexiform layer.⁴⁵ Utilizing wavelength-dependent responses with ORG could differentiate each cone type (S, M, or L).⁴⁶ As this method continues to evolve, it holds the exciting prospect of enabling the tracking of cell dysfunction across the entire retinal thickness in the context of various inherited retinal diseases. 

Conclusions

The combination of ultrahigh-resolution SD-OCT and SSADOR holds promise as a method to measure macular cone function, especially in patients who have subclinical or very mild disease presentation with preserved cone structure on OCT. SSADOR has the potential to provide higher sensitivity for detecting earlier, subtler cone dysfunction. 

Acknowledgments

Supported by an unrestricted departmental funding grant from Research to Prevent Blindness. 

Disclosure: N. Wongchaisuwat, None; A. Amato, None; P. Yang, None; L. Everett, None; M.E. Pennesi, None; D. Huang, None; S. Chen, None