Quantification system for pigment epithelial detachment (PED) and subretinal hyperreflective material (SHRM) reflectivity

Abstract

Systems and methods for quantifying biomarkers associated with ophthalmological elements identified via OCT imaging. A set of optical coherence tomography (OCT) volumes is received, each OCT volume corresponding to a different timepoint in a set of timepoints, and each OCT volume comprising a set of OCT B-scans. For each OCT volume of the set of OCT volumes, a set of element images is generated that visually identifies ophthalmological elements using ophthalmological element indicators. The ophthalmological element indicators assign a different group of pixels to each ophthalmological element of the plurality of ophthalmological elements. The ophthalmological elements include a target retinal element, a retinal pigment epithelial (RPE) layer, and a vitreous body. A reflectivity score is computed for the target retinal element identified in each OCT volume using the ophthalmological elements identified by the ophthalmological element indicators in each OCT volume to thereby form a set of reflectivity scores.

Description

QUANTIFICATION SYSTEM FOR PIGMENT EPITHELIAL DETACHMENT (PED) AND SUBRETINAL HYPERREFLECTIVE MATERIAL (SHRM) REFLECTIVITYInventors:Isabel Wilma BACHMEIER, Carl-Gustav Olsen GLITTENBERG, Ian Lloyd JONES, Sayedali SHETAB BOUSHEHRI, Siqing YU, Mahnaz AMIRI PARIAN, Andreas MAUNZCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application is related to and claims the benefit of the priority date of U.S. Provisional Application 63 / 566,349, filed March 17, 2024, entitled “Pigment Epithelial Detachment (PED) and Subretinal Hyperreflective Material (SHRM) Reflectivity Distribution and Dynamics,” which is incorporated herein by reference in its entirety.FIELD

[0002] The present application relates to the quantification of biomarkers using optical coherence tomography (OCT) images, and more particularly, to the quantification (e.g., one or more reflectivity scores) of one or more biomarkers (e.g., pigment epithelial detachment (PED) and / or subretinal hyperreflective material (SHRM)) identified using OCT images of a retina for use in evaluating a condition of the retina (e.g., neovascular age-related macular degeneration (nAMD)) over time (e.g., treatment response / treatment outcome).BACKGROUND

[0003] Age-related macular degeneration (AMD) is a leading cause of vision loss in subjects 50 years and older. AMD initially manifests as a dry type of AMD and can progress to a wet type of AMD. For the dry type, small deposits (drusen) form under the macula on the retina, causing the retina to progressively deteriorate. For the wet type, which may also be referred to as neovascular AMD (nAMD), abnormal blood vessels originating in the choroid layer of the eye grow into the retina and leak fluid from the blood into the retina. Upon entering the retina, the fluid may distort the vision of a subject immediately, and over time, can damage the retina itself,for example, by causing the loss of photoreceptors in the retina. The fluid can cause the macula to separate from its base, resulting in severe and rapid vision loss.

[0004] Optical coherence tomography (OCT) can provide a detailed scan of the macula to help detect macular degeneration, diabetic macular edema, and other macular and / or general retinal problems much earlier than was possible in the past. To investigate the extent of the deterioration in a retina with nAMD, OCT images (e.g., time domain optical coherence tomography (TD-OCT), spectral domain optical coherence tomography (SD-OCT), or swept-source optical coherence tomography (SS-OCT) images) of the retina may be obtained and used for identifying elements that may be associated with varying degenerative levels of nAMD. SD-OCT is an imaging technique in which light is directed at the retina at various optical frequencies and in which the reflected light is collected to capture two-dimensional or three-dimensional, high-resolution, cross- sectional images of the retina via interferometric signals detected as a function of frequencies. OCT imaging may be used to capture respective intensities of the identified elements of the retina.

[0005] OCT imaging data may visually identify or allow the identification of various biomarkers. Some currently available techniques used to quantify such biomarkers (e.g., reflectivity of retinal elements such as pigment epithelial detachment (PED) and / or subretinal hyperreflective material (SHRM)) may be unable to or at least find it difficult to adjust for noise in the OCT imaging data and / or adjust for the background reflectivity of other retinal elements. Thus, it may be desirable to have one or more methods and / or systems that recognize and consider the above-noted issues.SUMMARY

[0006] In one or more embodiments, a method for quantifying a target retinal element as a biomarker using reflectivity dynamics is provided. A set of optical coherence tomography (OCT) volumes for a retina of a subject may be received, wherein each OCT volume in the set of OCT volumes corresponds to a different timepoint in a set of timepoints and comprises a set of two- dimensional OCT B-scans. A set of element images may be generated for each OCT volume of the set of OCT volumes in which the set of element images visually identifies a plurality of ophthalmological elements using a plurality of ophthalmological element indicators; wherein the plurality of ophthalmological element indicators assigns a different group of pixels to each ophthalmological element of the plurality of ophthalmological elements; and wherein the plurality of ophthalmological elements includes a target retinal element, a retinal pigment epithelial (RPE)layer, and a vitreous body. A reflectivity score may be identified for the target retinal element identified in each OCT volume of the set of OCT volumes using the plurality of ophthalmological elements identified by the plurality of ophthalmological element indicators in each OCT volume of the set of OCT volumes to thereby form a set of reflectivity scores.

[0007] In one or more embodiments, a system comprises one or more data processors; and a non-transitory computer readable medium containing instructions, which when executed on the one or more data processors, cause the one or more data processors to perform part or all of one or more methods described herein or a portion thereof.

[0008] In one or more embodiments, a computer-program product is provided. The computerprogram product is tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause one or more data processors to perform part or all of one or more methods described herein or a portion thereof.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] For a more complete understanding of the principles disclosed herein, and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:

[0010] FIG. 1 is a block diagram of a biomarker quantification system, in accordance with various embodiments.[OU] FIG. 2 is a block diagram of a portion of the biomarker quantification system from Figure 1 described in further detail, in accordance with one or more embodiments.

[0012] FIG. 3 are example images used and / or generated by the biomarker quantification system from Figure 1, in accordance with one or more embodiments.

[0013] FIG. 4 is a flow chart for biomarker quantification, in accordance with various embodiments.

[0014] FIG. 5 is a graph showing an example reflectivity score distribution, in accordance with various embodiments.

[0015] FIGS. 6 and 7 are graphs showing the change in reflectivity scores computed for a set of patients receiving treatment over a set of timepoints, in accordance with various embodiments.

[0016] FIG. 8 is a block diagram of a computer system in accordance with various embodiments.

[0017] It is to be understood that the figures are not necessarily drawn to scale, nor are the objects in the figures necessarily drawn to scale in relationship to one another. The figures are depictions that are intended to bring clarity and understanding to various embodiments of apparatuses, systems, and methods disclosed herein. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts. Moreover, it should be appreciated that the drawings are not intended to limit the scope of the present teachings in any way.DETAILED DESCRIPTIONI. Overview[0181 Various types of ophthalmological diseases (or conditions) may be detected, diagnosed, and / or treated using a detailed scan of the retina to assess retinal elements which may change over time in response to disease progression and / or disease treatment. As one example, neovascular age-related macular degeneration (nAMD) may be detected, diagnosed, and / or treated using a detailed scan of the retina to assess the reflectivity of retinal elements (e.g., pigment epithelial detachment (PED) and / or subretinal hyperreflective material (SHRM)) where present in the retina. Fibrosis, or scar tissue in the retina, which can occur as a result of the blood vessel growth characteristic of nAMD, may also be detected, diagnosed, and / or treated in a similar manner. The embodiments described herein provide an improved technique for assessing the reflectivity of such retinal elements that is more accurate and more reliable than existing methods for assessing retinal element reflectivity. More accurate and more reliable assessment of retinal element reflectivity may help ensure more accurate and thorough diagnostic and / or treatment solutions for patients with ophthalmological diseases such as, for example, but not limited to, nAMD, and resulting scar tissue (e.g., fibrosis).

[0019] Retinal elements, such as SHRM, may be correlated with fibrosis development based on subtype. As one example, undefined SHRM, with low reflectivity and no clearly defined boundaries, may be correlated with immature blood vessels of the macular neovascular complex, which may be or have more solubility or more fluidic components, whereas well-defined SHRM, with higher reflectivity and more clearly defined borders, may be correlated with less fluidic and more fibrotic components.

[0020] A retinal element may be comprised of at least one of a retinal layer element or a retinal pathological element. Detection and identification of one or more retinal layer elements may be referred to as layer element (or retinal layer element) segmentation. Detection and identification of one or more retinal pathological elements may be referred to as pathological element (or retinal pathological element) segmentation.

[0021] In some cases, assessing retinal element reflectivity in a manner that accurately adjusts for noise in the OCT imaging data, background reflectivity of other retinal elements and / or retinal layers, or both using current methodologies and systems may be difficult. For example, some currently available methodologies and systems may be unable to adjust for noise in the OCTimaging data and / or the background reflectivity of other retinal elements and / or retinal layers, when analyzing the reflectivity of a selected retinal element (e.g., PED / SHRM), which may skew such analysis. Further, manual annotations of each SHRM subtype (e.g., undefined, well-defined, etc.) for each B-scan of an OCT volume is overly cumbersome for human graders; even more so for a set of OCT volumes over a set of time points.

[0022] Thus, the embodiments described herein provide improved methodologies and systems for quantifying reflectivity of ophthalmological elements, such as retinal elements, over one or more timepoints. For example, the embodiments may provide a way of quantifying reflectivity with a reflectivity score that adjusts for noise and / or the reflectivity of other ophthalmological elements, thereby leading to a more accurate and reliable quantification of reflectivity. Accordingly, this type of reflectivity score may be more reliably and confidently used as an indicator of the condition of the retina for the purposes of evaluating retinal health over time, treatment response, treatment outcome, etc.

[0023] Recognizing and taking into account the importance and utility of a methodology and system that can provide the improvements described above, the present disclosure describes various embodiments for quantifying reflectivity (e.g., computing a reflectivity score) for a selected ophthalmological element (e.g., PED, SHRM), which may include segmentation and identification of a plurality of selected ophthalmological elements, using a ML-based algorithm. The embodiments described herein enable more accurate and more reliable assessment of ophthalmological element reflectivity, which may improve the accuracy and reliability of any detection, diagnosis, and / or treatment methodologies that rely on the results of this reflectivity assessment. In some embodiments, the full automation results in reflectivity computations that are unbiased and unaffected or at least less biased and less affected by any underlying pathology compared to manual grading.

[0024] The embodiments described herein of quantifying reflectivity of ophthalmological elements also provide an improved technique for identifying three-dimensional measurements of such retinal elements that is more accurate and more reliable than existing methods. More accurate and more reliable three-dimensional measurements may help ensure more accurate and thorough diagnostic and / or treatment solutions for patients with ophthalmological diseases such as, for example, but not limited to, nAMD, and resulting scar tissue (e.g., fibrosis).II, Machine Learning (ML)-Based Reflectivity Score ComputationII.A. Example. Systems for Reflectivity Scare Computation[0251 FIG. 1 is a block diagram of a biomarker quantification system 100 in accordance with various embodiments. The biomarker quantification system 100 is used for automatically evaluating the condition of retinas of subjects by automatically quantifying reflectivity of ophthalmological elements as captured in the OCT imaging data of subjects with various ophthalmological diseases (or conditions) such as, for example, but not limited to, nAMD using the image input 102, which may be received or accessed via a network 104. In some embodiments, the retina is a healthy retina. In other embodiments, the retina is one that has been diagnosed with or is suspected of having a retinal disease. For example, the diagnosis may be one of age-related macular degeneration (AMD), diabetic macular edema (DME), or some other type of retinal disease.

[0026] Generally, the biomarker quantification system 100 is used to quantify biomarkers identified through OCT imaging data. For example, a biomarker may take the form of an OCT feature. An OCT feature may be a feature that can be visualized on an OCT image, identified from an OCT image, or otherwise measured using an OCT image to provide an indication of a normal biological processes, pathogenic processes, or pharmacologic response to a therapeutic intervention with respect to the retina of a subject.

[0027] As illustrated in FIG. 1, the biomarker quantification system 100 includes a computing platform 106 configured to store and execute an image processor 108, an ophthalmological element identification system 110, and a score generator 112. Generally, the image processor 108 receives or accesses the image input 102 and generates processed image(s) 114. The processed image(s) 114 are inputs to the ophthalmological element identification system 110, which uses the processed image(s) 114 to generate a set of element images 116. As illustrated, the ophthalmological element identification system 110 may include a retinal segmentation model 118 and an ophthalmological element algorithm 120. The retinal segmentation model 118 and the ophthalmological element algorithm 120 are used to generate the set of element images 116, which are sent to the score generator 112 to generate a set of reflectivity scores 122. The set of reflectivity scores 122 may be used to generate, using an output generator 124, an output 126, which may be sent to a remote device 128 via the network 104.

[0028] While the image processor 108, the ophthalmological element identification system 110, the score generator 112, and the output generator 124 are illustrated as being stored and executed using the same computing platform (i.e., the computing platform 106), in some embodiments, one or more of the image processor 108, the ophthalmological element identification system 110, the score generator 112, and the output generator 124 are stored and executed using a computing platform that is different from the computing platform 106. In some embodiments, the biomarker quantification system 100 also includes a data storage 130 and a display system 132. The data storage 130 and display system 132 are each in communication with the computing platform 106. In some examples, the data storage 130, display system 132, or both may be considered part of or otherwise integrated with the computing platform 106. Thus, in some examples, the computing platform 106, the data storage 130, and the display system 132 may be separate components in communication with each other, but in other examples, some combination of these components may be integrated together.

[0029] As illustrated, the image input 102 may include OCT imaging data 134, which may be generated using an OCT imaging system 136 or an OCT scanner. The OCT imaging system 136 can be a large tabletop configuration used in clinical settings, a portable or handheld dedicated system, or a “smart” OCT system incorporated into user personal devices such as smartphones. In some cases, the OCT imaging system 136 may include an image denoiser that is configured to remove noise and other artifacts from a raw OCT volume image to generate an OCT volume. In one or more embodiments, the OCT imaging data 134 includes a set of OCT volumes 138 for a retina of a subject. In some embodiments, the set of OCT volumes 138 corresponds to the retina of a patient at a given (i.e., one) timepoint. In one or more embodiments, the set of OCT volumes 138 corresponds to the retina of a patient at a set of timepoints. For example, each OCT volume 140 of the set of OCT volumes 138 may correspond to a particular timepoint of the set of timepoints (e.g., a baseline point in time, a number of hours, days, or months after a baseline point in time, or some other type of timepoint). Each OCT volume 140 of the set of OCT volumes 138 may be comprised of a set of OCT B-scans 142 of the retina of the subject. The set of OCT B- scans 142 may include, for example, without limitation, 10s, 100s, 1000s, 10,000s, or some other number of OCT B-scans. An OCT B-scan may also be referred to as an OCT slice image or a cross-sectional OCT image.

[0030] Although only one of each OCT imaging system 136 and the biomarker quantification system 100 is shown, there can be more than one of each in other embodiments. Further, although FIG.l shows the OCT imaging system 136 and the biomarker quantification system 100 as two separate components, in some embodiments, the OCT imaging system 136 and the biomarker quantification system 100 may be parts of the same system (e g., and maintained by the same entity such as a healthcare provider or clinical trial administrator). In some cases, a portion of the biomarker quantification system 100 may be implemented as part of OCT imaging system 136. For example, the biomarker quantification system 100 may be configured to run as a module implemented using a processor, microprocessor, or some other hardware component of OCT imaging system 136. In still other embodiments, the biomarker quantification system 100 may be implemented within a cloud computing system that can be accessed by or otherwise communicate with the OCT imaging system 136.

[0031] In one embodiment, the image processor 108 is configured or programmed to receive and perform a set of processing operations on the set of OCT B-scans 142 of the set of OCT volumes 138, which is the image input 102, to form the processed images 114. The set of OCT volumes 138 may be sent as input into the image processor 108, retrieved by the image processor 108 from storage, or accessed in some other manner. The set of processing operations may include, for example, without limitation, at least one of a normalization operation, a scaling operation, a resizing operation, a horizontal flipping operation, a vertical flipping operation, a cropping operation, a rotation operation, a noise filtering operation, or some other type of preprocessing operation.

[0032] The image processor 108 may be implemented using hardware, software, firmware, or a combination thereof. In one or more embodiments, the image processor 108 may be implemented within the computing platform 106 but in other embodiments at least a portion of (e.g., a module of) the image processor 108 is implemented within the OCT imaging system 136.

[0033] The ophthalmological element identification system 110 may be comprised of any number of models, algorithms, formulas, and / or equations that may be used to process the image input 102 and / or the processed image(s) 114. When the ophthalmological element identification system 110 processes the image input 102, the image input 102 may include the set of OCT volumes 138; may be formed by preprocessing the set of OCT volumes 138; may include the OCT volume 140 and its corresponding set of OCT B-scans 142; or may be formed by preprocessingthe OCT volume 140 and its corresponding set of OCT B-scans 142. In some embodiments, the ophthalmological element identification system 110 may process the image input 102 and / or the processed image(s) 114 to generate, for example, biomarker scores that can then be processed by the output generator 124 to form the output 126. The ophthalmological element identification system 110 processes the image input 102 and / or the processed images 114 to generate the set of element images 116 for each OCT B-scan in the set of OCT B-scans 142 for each OCT volume 140 in the set of OCT volumes 138 included in the image input 102 and / or the processed images 114.

[0034] In one or more embodiments, the ophthalmological element identification system 110 includes the retinal segmentation model 118 and the ophthalmological element algorithm 120. In some embodiments, the retinal segmentation model 118 may generate a first segmented image (e g., the first segmented image 144) for each OCT B-scan in the set of OCT B-scans 142 for each OCT volume 140 of the set of OCT volumes 138 in the image input 102 and / or the processed images 114. For example, one or more of first segmented images may be generated from OCT imaging data according to one or more OCT segmentation techniques as described in International Publication No. WO2023205511A1, which is incorporated by reference herein in its entirety. In some embodiments, the set of element images 116 visually identifies a plurality of ophthalmological elements using a plurality of ophthalmological element indicators. An ophthalmological element may be comprised of a retinal layer element or a retinal pathological element, or may be associated with at least one of a retinal layer element or a retinal pathological element.

[0035] A retinal layer element may be, for example, a retinal layer or a boundary associated with a retinal layer. Examples of retinal layers include, but are not limited to, an internal limiting membrane (ILM) layer, an external limiting membrane (ELM) layer, an outer plexiform layer- Henle fiber layer (OPL-HFL), a retinal pigment epithelial (RPE) layer, a layer of RPE detachment, a Bruch’s membrane (BM) layer, an ellipsoid zone (EZ), and other types of retinal layers. A boundary associated with a retinal layer may be, for example, an inner boundary of the retinal layer, an outer boundary of the retinal layer, a boundary associated with a pathological feature of the retinal layer (e.g., an inner or outer boundary of detachment of the retinal layer), or some other type of boundary. For example, a boundary may be an inner boundary of an RPE (IB-RPE)detachment layer, an outer boundary of the RPE (OB-RPE) detachment layer, or another type of boundary.[0361 A retinal pathological element may include, for example, fluid, cells, solid material, or a combination thereof that evidences a retinal pathology associated with an ophthalmological disease or condition. For example, the presence of certain retinal fluids may be a sign of leakage from retinal blood vessels, which may be a sign of nAMD. Examples of retinal pathological elements include, but are not limited to, intraretinal fluid (IRE), subretinal fluid (SRF), fluid associated with pigment epithelial detachment (PED), hyperreflective material (HRM), subretinal hyperreflective material (SHRM), intraretinal hyperreflective material (IHRM), hyperreflective foci (HRF), hard exudates (HE), a retinal fluid pocket, and a disruption. In some cases, a retinal pathological element may be a disruption (e.g., discontinuity, delamination, loss, etc.) of a retinal layer or retinal zone. For example, the disruption may be of the ellipsoid zone, of the ELM, of the RPE, or of another layer or zone. The disruption may represent damage to or loss of cells (e.g., photoreceptors) in the area of the disruption.

[0037] In some cases, a retinal layer element is associated with a retinal pathological element. For example, an RPE detachment layer, which is a retinal layer element, is associated with PED, which is a retinal pathological element. In some cases, an ophthalmological element is associated with a retinal layer element or a retinal pathological element. For example, a vitreous body, which is an ophthalmological element, is associated with the ILM layer. In some embodiments, the vitreous body may alternatively be referred to as the vitreous matrix, or the vitreous chamber.

[0038] In one or more embodiments, the retinal segmentation model 118 may additionally detect and identify ophthalmological elements further categorized as a subtype of the ophthalmological element. As a non-limiting example, the retinal segmentation model 118 may additionally detect and identify one or more subtypes of SHRM, including, for example without limitation, undefined SHRM, and / or well-defined SHRM. Undefined SHRM may be more common in early stages of retinal disease (e.g., nAMD, etc.) and can either fully resolve with treatment, or transform into well-defined SHRM.

[0039] In some embodiments, the retinal segmentation model 118 may include, but is not limited to, a convolutional neural network, such as a U-net model, that has been trained to recognize regions of intraretinal fluid (IRF), subretinal fluid (SRF), SHRM and PED on OCT B- scans at a pixel-level, using OCT training data. In one or more embodiments, the retinalsegmentation model 1 18 was trained using training data that included a plurality of OCT B-scan images, which had been annotated by human graders to classify IRF, SRF, SHRM, and PED within the OCT image. In one or more embodiments, the retinal segmentation model 118 was trained using training data that included the plurality of OCT B-scan images, which had been annotated by human graders to identify contours of IRF, SRF, PED, and SHRM. OCT B-scans included in the training data that included a SHRM annotation were further manually labeled as including undefined SHRM or well-defined SHRM, based on if <50% or >50% of SHRM was delineable from the neurosensory retina. The labels “undefined SHRM” and “well-defined SHRM,” together with the background label, were used for training. IRF, SRF, and PED annotations were also included to allow learning of more generalized features.

[0040] Thus, the retinal segmentation model 118 may be used to perform layer element segmentation to detect and identify retinal layer elements and / or retinal pathological element segmentation to detect and identify ophthalmological elements. In some embodiments, ophthalmological elements may be subtypes of a retinal layer element or retinal pathological element.

[0041] In some embodiments, the ophthalmological element algorithm 120 may generate a second segmented image (e.g., the second segmented image 146), using the first segmented image generated by the retinal segmentation model 118 (e.g., the first segmented image 144), to identify ophthalmological element(s) that are associated with the retinal elements (i.e., retinal layer elements and / or retinal pathological elements) identified in the first segmented image. The second generated image may include a plurality of ophthalmological element indicators that identify additional ophthalmological element(s) relative to those identified in the first generated image, by assigning a group of pixels to the additional ophthalmological elements.

[0042] In one or more embodiments, the ophthalmological element algorithm 120 may use one or more ophthalmological elements identified on the first segmented image to identify the additional ophthalmological element(s). For example, without limitation, the ophthalmological element algorithm 120 may use the ILM layer as identified by the retinal segmentation model 118 on the first segmented image, in order to identify the vitreous body as an additional ophthalmological element.

[0043] The outputs of the retinal segmentation model 118 and the ophthalmological element algorithm 120 (i.e., the first and second segmented images) are used to generate the set of elementimages 116 (e.g., the first element image 148 and / or the second element image 150). In one or more embodiments, the set of element images 116 includes at least a first element image, that identifies at least the target retinal element (e.g., PED or SHRM), and a second element image that identifies at least the RPE layer and the vitreous body. In some embodiments, an element image of the set of element images 116 is postprocessed by the image processor 108, by performing a pixel erosion operation. In one or more embodiments, the pixel erosion operation may include, for example, without limitation, eroding a number of pixels from every direction. For example, without limitation, 1, 2, 3, 5, 8, 10, 20, or some other number of pixels may be removed from every direction in a pixel erosion operation. In some embodiments, the pixel erosion operation may be followed by a secondary cleaning operation. For example, without limitation, 1, 2, 3, 5, 8, 10, 20, or some other number of pixels may be removed from just one, two, or three directions. The set of element images 116 is discussed in further detail below with respect to FIG. 2.

[0044] As illustrated, the set of element images 116 is sent to the score generator 112, which uses at least one element image in the set of element images 116 to generate one or more scores for each corresponding OCT B-scan. For example, the score generator 112 may generate a reflectivity score for the target retinal element identified in each OCT B-scan in the set of OCT B- scans 142 to thereby form the set of reflectivity scores 122. Each reflectivity score in the set of reflectivity scores 122 is a quantification of reflectivity of the target retinal element that accounts for and has been adjusted for noise in the OCT B-scan and the reflectivity of the RPE layer and the vitreous body.

[0045] The output generator 124 may use the set of reflectivity scores 122 to generate the output 126. The output 126 may include a classification for a condition of the retina at a given timepoint, or set of timepoints, based on the set of reflectivity scores 122. The output 126 may include a report based on the set of reflectivity scores 122. The report may be an indication of disease progression, treatment response, or treatment outcome.

[0046] FIG. 2 is a block diagram of a portion of the biomarker quantification system 100 of FIG. 1, described in further detail in accordance with one or more embodiments. After the image input 102 has been received by the biomarker quantification system 100 and has been used to generate the processed image(s) 114, the biomarker quantification system 100 processes the processed image(s) 114. As noted above and in one or more embodiments, the image input 102and / or the processed image(s) 114 may include the set of OCT B-scans 142 or may be formed by preprocessing the set of OCT B-scans 142.[0471 As previously discussed, the biomarker quantification system 100 may include the ophthalmological element identification system 110, which may include the retinal segmentation model 118 and the ophthalmological element algorithm 120, which are used to generate the first segmented image 144 and the second segmented image 146, respectively.

[0048] As illustrated, the first segmented image 144, as generated by the retinal segmentation model 118, may include a plurality of ophthalmological element indicators which identify a plurality of ophthalmological elements (e.g., ophthalmological elements 202, 204, and 206) by assigning a group of pixels to each ophthalmological element of the plurality of ophthalmological elements. In one or more embodiments, the plurality of ophthalmological elements includes at least a target retinal element, such as pigment epithelial detachment (PED) or subretinal hyperreflective material (SHRM), an inner limiting membrane (ILM) layer, and a retinal pigment epithelium (RPE) layer. In FIG. 2, the first segmented image 144 includes an ILM layer (the ophthalmological element 202), PED (the ophthalmological element 204), and an RPE layer (the ophthalmological element 206). In some embodiments, the entire RPE across each B-scan as opposed to across the RPE in the region of SHRM in each B-scan is used because RPE may be better preserved peripheral to a macular neovascularization (MNV) lesion.

[0049] As further illustrated, the second segmented image 146 may be generated by the ophthalmological element algorithm 120 using the first segmented image 144 to identify an additional ophthalmological element using a plurality of ophthalmological element indicators by assigning a group of pixels to the additional ophthalmological element. In FIG. 2, the second segmented image 146 identifies the vitreous body as an additional ophthalmological element (the ophthalmological element 208) using the ILM layer (the ophthalmological element 202) as identified by the retinal segmentation model 118 on the first segmented image 144.

[0050] The first and second segmented images 144 and 146 are then used to generate the set of element images 116, which includes the first element image 148 and the second element image 150. The first element image 148 identifies at least the target retinal element (e.g., PED or SHRM), while the second element image 150 identifies at least the RPE layer and the additional ophthalmological element as identified by the ophthalmological element algorithm 120. In FIG. 2, the first element image 148 identifies PED as the target retinal element (the ophthalmologicalelement 204), while the second element image 150 identifies the RPE layer (the ophthalmological element 206) and the vitreous body (the ophthalmological element 208).[0511 The set of element images 116 (e.g., at least one element image of the set of element images 116) generated for each OCT B-scan of the image input 102 may be used by the score generator 112 to generate one or more scores for each corresponding OCT volume in the image input 102. For example, the score generator 112 may generate a reflectivity score for the target retinal element identified in each OCT volume to thereby form the set of reflectivity scores 122. Each reflectivity score of the set of reflectivity scores 122 is a quantification of reflectivity of target retinal element in an OCT volume at a timepoint, that accounts for and has been adjusted for noise in the OCT volume and the reflectivity of the RPE layer and the vitreous body.

[0052] A reflectivity score may be computed in various ways, for example, by quantifying an intensity metric for the target retinal element, and adjusting the intensity metric to account for the reflectivity of the RPE layer and the vitreous body (e.g., intensity metrics of the RPE layer and the vitreous body).

[0053] In some embodiments, an intensity metric for an ophthalmological element may be quantified by first identifying pixel intensity values for the group of pixels assigned to the ophthalmological element as identified in the set of element images 116 using one or more ophthalmological element indicators. For example, an intensity metric for the PED as identified in the first element image 148 (the ophthalmological element 204) may be quantified by first identifying pixel intensity values for the group of pixels assigned to the ophthalmological element 204.

[0054] In some embodiments, pixel intensity values for an ophthalmological element may be identified for each OCT B-scan for which the set of element images 116 has been generated. In other embodiments, pixel intensity values for an ophthalmological element may be further processed by the score generator 112 to normalize the pixel intensity value for the ophthalmological element in each OCT B-scan of the OCT volume for which the set of element images 116 has been generated.

[0055] An intensity metric for an ophthalmological element for an OCT volume may be identified by identifying a pixel intensity value of the ophthalmological element that meets a selected criterion across each OCT B-scan of the OCT volume. In some embodiments, the selected criterion may be the median pixel intensity value. In certain embodiments, the selected criterionmay be a pixel intensity value at a selected percentile. The selected percentile may be any percentile between a given range, such as, for example, between the 90th- 99thpercentiles.

[0056] In one or more embodiments, the score generator 112 computes intensity metrics for a plurality of ophthalmological elements, that are then used as references for adjusting the intensity metric of a target retinal element (e.g., when identifying a reflectivity score for the target retinal element) to account for a first reflectivity of a reference darker than the target retinal element, and to account for a second reflectivity of a reference lighter than the target retinal element.

[0057] The reflectivity score for the target retinal element may be identified using the equation below, where lMTarge tis the intensity metric identified for the target retinal element, IMDark_ reference is the intensity metric identified for the reference darker than the target retinal element, and IMLight referenceis the intensity metric identified for the reference lighter than the target retinal element:

[0058] For example, the score generator 112 may compute intensity metrics for at least a target retinal element (e.g., PED or SHRM), the RPE layer, and the vitreous body. These intensity metrics may then be used to compute a reflectivity score of the target retinal element that accounts for reflectivity of the RPE layer as the reference lighter than the target retinal element and the vitreous body as the reference darker than the target retinal element.

[0059] For example, a reflectivity score for a target retinal element may be identified using the equation below, wherein IMTargetis the intensity metric identified for the target retinal element, IMy reous is the intensity metric identified for the vitreous body, and 1MRPEis the intensity metric identified for the RPE layer:

[0060] In one or more embodiments, the IMTarge tmay be computed as a median intensity of the target retinal element identified on each OCT B-scan of an OCT volume of a set of OCT volumes, the IMvitreousmay be computed as a median intensity of the vitreous body identified oneach OCT B-scan of an OCT volume of a set of OCT volumes, and the IMRPEmay be computed as a pixel intensity value of the RPE at the 95thpercentile on each OCT B-scan of an OCT volume of a set of OCT volumes.

[0061] In one or more embodiments, the set of reflectivity scores 122 is a set of biomarkers for disease progression. In some embodiments, the set of reflectivity scores 122 is a set of biomarkers for at least treatment outcome or treatment response with respect to a selected treatment of the retina. For example, without limitation, the set of reflectivity scores 122 may be a set of biomarkers for at least treatment outcome or treatment response with respect to an anti angiogenic therapy of the retina, such as ranibizumab, aflibercept, bevacizumab, faricimab, or brolucizumab.

[0062] In some embodiments, a clinical trial may be altered, redesigned, replicated, or assessed, in response to the set of reflectivity scores 122. For example, without limitation, a clinical trial studying the effects of a treatment for a retinal disease (e.g., nAMD or fibrosis) may be altered, redesigned, replicated, or assessed, in response to the set of reflectivity scores generated for a target retinal element that is a biomarker for the retinal disease.

[0063] As discussed above, the set of reflectivity scores 122 may then be used by the output generator 124 to generate the output 126. The output 126 may include an indication of a classification for a condition of the retina, disease progression, treatment response, or treatment outcome, at a timepoint in the set of timepoints. In one or more embodiments, a clinical trial may be altered, redesigned, replicated, or assessed, in response to the output 126. For example, without limitation, a clinical trial studying the effects of a treatment for a retinal disease (e.g., nAMD or fibrosis) may be altered, redesigned, replicated, or assessed, in response to the output 126 generated using the set of reflectivity scores 122 generated for a target retinal element that is a biomarker for the retinal disease. In some embodiments, using the image processor 108 and / or the ophthalmological element identification system 110 to process the image input 102 may reduce the overall computing resources that would be otherwise needed to generate such reflectivity scores. In some embodiments, the ophthalmological element identification system 110 generating the set of element images 116 allows for the score generator 112 to more efficiently generate a more accurate output, saving computing resources.

[0064] In some embodiments, the biomarker quantification system 100 assesses and characterizes ophthalmological elements with better accuracy and consistency than expert human graders. As such, and in some embodiments, the biomarker quantification system 100 provides atechnical effect of improving accuracy, reducing the overall computing resources, and / or providing a fully automated biomarker quantification that was previously not automatable.[0651 Insome embodiments, the biomarker quantification system 100 provides a technical improvement to the field of three-dimensional SHRM investigations. For example, using the biomarker quantification system 100 may allow or enable for the investigation of SHRM and its subtypes in a three-dimensional manner, which allows for not only spot measurements but also for en face area and volume measurements. Further, using the biomarker quantification system 100 may allow or enable for the restrictions of measurements to specific ETDRS areas of interest.II.B. Example Images for Biomarker Quantification

[0066] FIG. 3 includes example images 300 used and / or generated by the biomarker quantification system 100 in accordance with various embodiments. Image 302 is an example OCT B-scan (e.g., an OCT B-scan of set of OCT B-scans 142) that may be used as an image input 102 by the biomarker quantification system 100. Image 304 is an example of a first generated image 144 as generated by the retinal segmentation model 118 of ophthalmological element identification system 110. On image 304, undefined and well-defined SHRM have been detected and identified as ophthalmological elements. In some embodiments, the set of reflectivity scores 122 may be computed separately by the biomarker quantification system 100 for undefined SHRM and well-defined SHRM as identified by the retinal segmentation model 118.ILC. Example Methodologies for Reflectivity Score Computation

[0067] FIG. 4 is a flowchart of a process 400 for quantifying a target retinal element as a biomarker using reflectivity dynamics, in accordance with various embodiments. The target retinal element may be a biomarker associated with retinal disease such as, for example, nAMD or fibrosis. The target retinal element may be quantified by identifying reflectivity dynamics, such as, for example, a set of reflectivity scores over a set of timepoints. In various embodiments, the process 400 can be implemented using the biomarker quantification system 100 described in FIG. 1.

[0068] Step 402 of process 400 includes receiving a set of optical coherence tomography (OCT) volumes for a retina of a subject, wherein each OCT volume in the set of OCT volumes corresponds to a different timepoint in a set of timepoints and comprises a set of two-dimensionalOCT B-scans. The set of OCT volumes may be, for example, the set of OCT volumes 138 in FIG. 1. The set of two-dimensional OCT B-scans may be, for example, the set of OCT B-scans 142 in FIG. 1. The set of OCT volumes may be used as, or may be preprocessed to form, the image input for process 400. For example, the set of OCT volumes 138 and the corresponding set of OCT B- scans 142 may be used or preprocessed as image input 102 for the biomarker quantification system 100 as described in FIG. 1. An example of an image input 102 is discussed further above with respect to FIG. 3.

[0069] Step 404 of process 400 includes generating a set of element images for each OCT volume of the set of OCT volumes in which the set of element images visually identifies a plurality of ophthalmological elements using a plurality of ophthalmological element indicators. The plurality of ophthalmological element indicators may assign a different group of pixels to each ophthalmological element of the plurality of ophthalmological elements, and the plurality of ophthalmological elements may include at least a target retinal element, a retinal pigment epithelial (RPE) layer, and a vitreous body. The set of element images may be, for example, the set of element images 116 in FIG. 1.

[0070] In one or more embodiments, generating the set of element images for each OCT volume of the set of OCT volumes may include generating, by a retinal segmentation model, a first segmented image for each OCT B-scan of the set of OCT B-scans for each OCT volume in the set of OCT volumes. The first segmented image may include the plurality of ophthalmological element indicators that identify the plurality of ophthalmological elements by assigning a group of pixels to each ophthalmological element of the plurality of ophthalmological elements. The first segmented image may be, for example, the first segmented image 144 generated by the retinal segmentation model 118 in FIG. 1. An example of the first segmented image 144 is discussed further above with respect to FIG. 3.

[0071] In some embodiments, a first element image that at least identifies the target retinal element is generated using the first segmented image. In some embodiments, a second element image that identifies at least the RPE layer and an additional ophthalmological element is generated using the first segmented image. The first and second element images may be, for example, the first element image 148 and the second element image 150, respectively, in FIGS. 1 and 2. The target retinal element may be a retinal layer element or a retinal pathological element, as described above with respect to FIG. 1. For example, the target retinal element may be a retinalpathological element, such as pigment epithelial detachment (PED) or subretinal hyperreflective material (SHRM). The target retinal element may be, for example, the ophthalmological element 204 in FIG. 2 (i.e., PED). The RPE layer may be, for example, the ophthalmological element 206 in FIG. 2. In some embodiments, the target retinal element may be one or more subtypes of a retinal element, such as for example without limitation, well-defined and / or undefined SHRM (i.e., subtypes of SHRM).

[0072] In one or more embodiments, generating the second element image includes generating a second segmented image using the first segmented image and an ophthalmological element algorithm to identify the additional ophthalmological element. The second segmented image may include the plurality of ophthalmological element indicators that identify the additional ophthalmological element by assigning a group of pixels to the additional ophthalmological element. The second segmented image may be, for example, the second segmented image 146 generated by the ophthalmological element algorithm 120 in FIG. 1.

[0073] Generating the second segmented image may include using one or more ophthalmological elements identified on the first segmented image to identify the additional ophthalmological element. As a non-limiting example, in one or more embodiments the additional ophthalmological element may be the vitreous body, and may be identified using an internal limiting membrane (ILM) layer identified on the first segmented image. The ILM layer may be, for example, the ophthalmological element 208 in FIG. 2. The vitreous body may be, for example, the ophthalmological element 202 in FIG. 2.

[0074] In one or more embodiments, the set of element images may be postprocessed by performing one or more pixel erosion operation to remove a number of pixels from each element image of the set of element images from every direction, removing a number of pixels from each element image of the set of element images from one or more directions, or a combination thereof. In some embodiments, one or more pixel erosion operations may be performed to mitigate overlap of the target retinal element (as identified by ophthalmological element identification system 100), with the RPE layer.

[0075] For example, without limitation, a pixel erosion operation may be performed on the first element image (e.g., the element image identifying the target retinal element) to remove one pixel from every direction. A secondary pixel erosion operation may be performed on the first element image to remove three pixels above the target retinal element (e.g., where the target retinalelement is PED), or to remove eight pixels below the target retinal element (e g., where the target retinal element is SHRM).[0761 Step 406 of process 400 includes identifying a reflectivity score for the target retinal element identified in each OCT volume of the set of OCT volumes using the plurality of ophthalmological elements identified by the plurality of ophthalmological element indicators in each OCT volume of the set of OCT volumes to thereby form a set of reflectivity scores. In some embodiments, identifying the reflectivity score for the target retinal element includes quantifying an intensity metric for the target retinal element, and adjusting the intensity metric for the target retinal element to account for a first reflectivity of the vitreous body and a second reflectivity of the RPE layer. The set of reflectivity scores may be, for example, set of reflectivity scores 122 generated by the score generator 112 in FIG. 1.

[0077] In one or more embodiments, the set of reflectivity scores for a target retinal element includes a baseline reflectivity score for a baseline timepoint in the set of timepoints, and a threshold for separating reflectivity scores that indicate low reflectivity and reflectivity scores that indicate high reflectivity may be selected based on analyzing a distribution of a plurality of baseline reflectivity scores computed for a plurality of subjects.

[0078] In some embodiments, the set of reflectivity scores may include a baseline reflectivity score associated with a baseline timepoint and a first reflectivity score associated with a first timepoint after the baseline timepoint. In some embodiments, the set of reflectivity scores may include a change in reflectivity between a baseline reflectivity score associated with the baseline timepoint and a first reflectivity score associated with a first timepoint after the baseline timepoint.

[0079] In one or more embodiments, the set of reflectivity scores is a set of biomarkers for disease progression. In some embodiments, the set of reflectivity scores is a set of biomarkers for at least treatment outcome or treatment response with respect to a selected treatment of the retina. For example, without limitation, the set of reflectivity scores may be a set of biomarkers for at least treatment outcome or treatment response with respect to an antiangiogenic therapy of the retina, such as ranibizumab, aflibercept, bevacizumab, faricimab, or brolucizumab.

[0080] In some embodiments, a clinical trial may be altered, redesigned, replicated, or assessed, in response to the set of reflectivity scores. For example, without limitation, a clinical trial studying the effects of a treatment for a retinal disease (e.g., nAMD or fibrosis) may be altered,redesigned, replicated, or assessed, in response to the set of reflectivity scores generated for atarget retinal element that is a biomarker for the retinal disease.[0811 Process 400 may optionally include the step 408 of generating an output using the set of reflectivity scores. The output may include an indication of a classification for a condition of the retina, disease progression, treatment response, or treatment outcome, at a timepoint in the set of timepoints. In one or more embodiments, a clinical trial may be altered, redesigned, replicated, or assessed, in response to the generated output. For example, without limitation, a clinical trial studying the effects of a treatment for a retinal disease (e.g., nAMD or fibrosis) may be altered, redesigned, replicated, or assessed, in response to the output generated using the set of reflectivity scores generated for a target retinal element that is a biomarker for the retinal disease. The generated output may be, for example, the output 126 generated by the output generator 124 in FIG. 1.

[0082] In some embodiments, the process 400 measures and characterizes ophthalmological elements with better accuracy and consistency than expert human graders. In some embodiments, the process 400 provides a technical effect of improving accuracy, reducing the overall computing resources, and / or providing a fully automated biomarker quantification that conventional systems cannot provide.

[0083] In some embodiments, the process 400 provides a technical improvement to the field of three-dimensional SHRM investigations. For example, using the biomarker quantification system 100 may allow or enable for the investigation of SHRM and its subtypes in a three- dimensional manner, which allows for not only spot measurements but also for en face area and volume measurements. Further, using the biomarker quantification system 100 may allow or enable for the restrictions of measurements to specific ETDRS areas of interest. In some embodiments, the process 400 includes a new combination of steps that results in the technical improvement over conventional SHRM grading and characterization approaches.

[0084] In some embodiments, the process 400 provides a technical improvement to the field of fibrosis detection and / or prediction. Traditionally, fibrosis is graded using color and fluorescein angiography (FA) characteristics, which are difficult for manual graders to grade consistently, causing variability in fibrosis detection and / or prevalence rates. In some embodiments, the process 400 provides an automated and objective method of detecting fibrosis by using well-defined SHRM as a proxy for the existence of fibrosis and its development. For example, and as discussedbelow, well-defined SHRM has higher internal reflectivity compared to undefined SHRM, and a higher reflectivity and well-defined borders of SHRM may indicate that fibrotic processes exist. As such, and in some embodiments, the biomarker quantification system 100 and / or the process 400 may identify, using the image input 102 associated with a subject, whether the subject has, or has a high risk of, development of fibrosis. In some embodiments, the process 400 further includes administering a treatment or preventative therapy to the subject identified of having, or having a high risk of, fibrotic processes existing, to prevent or minimize further fibrosis development.

[0085] In some embodiments, the process 400 provides a technical improvement to the field of SHRM subtypes with respect to their distribution, characteristics and associations with visual function in nAMD eyes. In some embodiments, the process 400 may enable an automated, volumetric, and objective comparison of different mechanisms of action in nAMD treatments in the future and may inform better treatment strategies due to a more comprehensive understanding of the disease.II.D. Example Results of Reflectivity Score Computation

[0086] FIG. 5 is a graph 500 illustrating an example distribution of the reflectivity scores computed in the manner described above using biomarker quantification system 100 in FIG. 1, or process 400 of FIG. 4, to automatically evaluate the condition of retinas by automatically quantifying reflectivity of target retinal elements. In one or more embodiments, a distribution graph such as the one depicted in FIG. 5 may be used to evaluate a set of OCT volumes for a plurality of patients at a given timepoint. For example, a distribution graph at a baseline timepoint may be used to analyze subsequent reflectivity distributions of the target retinal elements at later timepoints.

[0087] FIGS. 6 and 7 are graphs 600 and 700, respectively, illustrating the change in reflectivity scores computed for PED and SHRM, respectively, as target retinal elements in a set of OCT volumes of patients diagnosed with nAMD (for a total of 108 study eyes), who received treatment with aflibercept (i.e., an anti-VEGF treatment used to treat nAMD), at weeks 0, 4, 8, 16, and 24. OCT volumes were generated for the patients at weeks 0, 8, and 24 to determine a baseline timepoint and two additional timepoints to examine the correlation of PED and SHRM volume and reflectivity following two and six months of treatment. The reflectivity scores may beexamples of the set of reflectivity scores 122 computed in the manner described above using biomarker quantification system 100 in FIG. 1, or process 400 of FIG. 4.

[0088] FIG. 6 illustrates that PED reflectivity increased over time from baseline, while PED volume decreased over time from baseline. Similarly, FIG. 7 illustrates that SHRM reflectivity increased over time from baseline, while SHRM volume decreased over time.II. E. Example A nalysis Using the Example System far Reflectivity Score Computation

[0089] In some embodiments and during an example post-hoc analysis using the biomarker quantification system 100, at least some medical data was also received for patients involved in a clinical trial. In some embodiments, the medical data includes a treatment-naive status that provides an indication of whether the subject has been previously treated for nAMD or not (i.e., is treatment-naive). In some embodiments, the medical data includes demographic data such as for example, age, sex, one or more other types of demographic variables, or a combination thereof. In some embodiments, the medical data also includes clinical data such as for example a visual acuity measurement or another type of clinical parameter or measurement. The visual acuity measurement may be, for example, a BCVA at baseline (with respect to treatment start within a trial) and at one or more reference points in time. In some cases, the reference point in time is a point in time after the baseline and / or a first dose of treatment such as, for example, 4 weeks, 6 weeks, 18 weeks, 24 weeks, 3 months, 6 months, 9 months, 1 year, 2 years, or some other amount of time after the first dose of treatment. Generally, the clinical data corresponding to the reference point in time may include data generated at the reference point in time, data generated within a selected range (e.g., within a selected number of days, weeks, or months) of the reference point in time, or both.

[0090] In one example analysis, a total of 1094 eligible treatment-naive study eyes from 1094 patients (out of all 1326 TENAYA / LUCERNE study eyes) were included. Of these, there were 655 females and 439 males, the mean age was 75.1 years (range: 50 to 89; SD: 7.9) and the mean BCVA was 60.3 ETDRS letters (range: 24 to 78; SD: 13.1). Based on the segmentation performed and reflectivity scores identified by the biomarker quantification system 100, the majority of eyes (i.e., 79%) had some SHRM; volume, en face area, and maximum height within the ETDRS 3 mm diameter, including eyes with zero values for the respective SHRM subtype, were significantly higher for undefined compared to well-defined SHRM; and there were also fewer zero values forundefined than for well-defined SHRM (12% vs 25%;). The reflectivity per SHRM subtype was also assessed by including only eyes with non-zero segmentation for each subtype. The mean reflectivity of well-defined SHRM volumes was 0.59 (n=794), ranging from 0.28 to 0.89, and significantly higher than that of undefined SHRM, which was 0.55 (n=952) and ranged from 0.21 to 0.98 (difference: 0.04, 95% confidence interval (CI) 0.03 to 0.04, P < .001). The heterogeneity of reflectivity, measured as the standard deviation across all pixels in SHRM volumes, was 0.11 in well-defined SHRM and 0.12 in undefined SHRM, ranging from 0 to 0.28 and from 0 to 0.26, respectively (difference: -0.007, 95% CI -0.01 to -0.004, p < 0.001). Both variables, mean and SD reflectivity, exhibited a normal distribution.

[0091] When differentiating between MNV classes as graded by the central reading centers on fluorescein angiography (FA), the median volumes for both, undefined and well-defined SHRM, were highest in classic MNV and predominantly classic MNV, with a gradual decrease towards more occult MNV classes. In the classic and predominantly classic MNV groups, the proportion of well-defined SHRM was higher than in the more occult classes. With respect to a detailed summary statistics for all dimensions of undefined and well-defined SHRM per MNV class, a trend was observed for volumes repeating for en face area and height.

[0092] Using least squares regression analyses, a negative statistically significant association was observed between BCVA and SHRM in the EDTRS 1 mm diameter for both undefined and well-defined SHRM and for all SHRM dimensions at baseline. The associations were stronger for well-defined compared to undefined SHRM. As an example, for the volumes in the central 1 mm diameter, an additional 100 nl of undefined SHRM was associated with a loss of 6.0 ETDRS letters vs. a loss of 18.3 ETDRS letters for well-defined SHRM. The ETDRS 1 mm diameter measurements showed the strongest associations with BCVA, especially for volumes and en face areas. Of note, the ETDRS 1 mm ring volumes are affected by a ceiling effect (resulting from the smaller radius limiting the maximum possible SHRM volume or en face area) which can make regression slopes steeper than they would be without such ceilings. A standardized multiple regression analysis showed that for each of the well-defined SHRM dimension features, the 3 or 6 mm ETDRS diameters were more strongly correlated with BCVA than the ETDRS 1 mm diameter. This standardized regression analysis was performed in order to compare partial correlations between BCVA and SHRM subtypes across different dimensions and ETDRS diameters. Partial correlations again were stronger for well-defined (-0.14 to -0.34) than for undefined SHRM (-0.03to -0.11 ). Within well-defined SHRM, within each ETDRS diameter, en face area had the highest partial correlations, followed by maximum height and volume. Within undefined SHRM, partial correlations were closer to each other without a clear preference for any of the dimensions or ETDRS areas.

[0093] In this example, it was demonstrated that in treatment-naive eyes: (1) undefined SHRM is the predominant subtype when assessed volumetrically using the biomarker quantification system 100, (2) well-defined SHRM has a higher internal reflectivity than undefined SHRM, as identified using the biomarker quantification system 100 (3) classic and predominantly classic MNVs show larger amounts of SHRM with a higher proportion of well-defined SHRM compared to other MNV classes, (4) both SHRM subtypes have a negative association with BCVA, with a more pronounced effect for well-defined SHRM, and (5) the enface area has the most adverse impact on BCVA, with maximal effect sizes seen in the 3 and 6 mm ETDRS rings.

[0094] In some embodiments, the biomarker quantification system lOOand / or the process 400 allows the investigation of SHRM and its subtypes in a three-dimensional manner, which allows not only for spot measurements but also for en face area and volume measurements. This is a technical effect over other systems and processes that either restrict SHRM measurements to individual scans or the central foveal scan, or to only assessing the SHRM height and width. In addition to being able to also measure en face areas and volumes, the biomarker quantification system 100 and / or the process 400 allows for the restriction of measurements to specific ETDRS areas of interest.

[0095] In an example analysis and when comparing the population-wide means for the various dimensions between undefined and well-defined SHRM, higher baseline values and fewer zero values for undefined SHRM were observed compared to well-defined SHRM. Despite fibrosis at enrollment being an exclusion criterion in TENAYA / LUCERNE and only having been detected in six eyes by the reading centers, a high prevalence was observed of well-defined SHRM, which is often considered as a proxy for fibrosis if persistent. Therefore, in this example analysis, it was expected to see a much lower prevalence of well-defined SHRM. However, the large variability in fibrosis prevalence rates reported in the literature may have arisen as this feature is traditionally graded on color and FA characteristics, which are far more difficult to grade consistently. Nonetheless, it is possible that even well-defined SHRM may resolve or reduce with treatment, and in some embodiments, the biomarker quantification system 100 is configured to investigateSHRM evolution patterns in the example dataset, which in turn should allow a better understanding of the dynamic changes within the subtypes.[0961 Further, and as illustrated in the example analysis, well-defined SHRM has higher internal reflectivity compared to undefined SHRM and a higher reflectivity and well-defined borders of SHRM may indicate that fibrotic processes exist. As such, the biomarker quantification system 100, and / or the process 400 may identify, using the image input 102 associated with a subject, whether the subject has, or has a high risk of, fibrotic processes existing.

[0097] The example analysis also applied the quantitative assessment of undefined and well- defined SHRM to differentiate between MNV classes. The analysis showed that SHRM sizes for both subtypes were largest in classic MNV and predominantly classic MNV, which is in accordance with SHRM representing the OCT-correlate for subretinally localized neovascularization complexes. At least one study using DL segmentation of overall SHRM on Cirrus images of the HARBOR trial came to the same conclusion that MNV type 2 lesions demonstrated the largest volumes of SHRM, and others had also previously shown that hyperreflective material (HRM) was most commonly observed in type 2 and mixed choroidal neovascularization (CNV). The example analysis further shows that there was a higher proportion of well-defined SHRM vs. undefined SHRM in eyes with classic or predominantly classic MNV than in the other MNV classes before treatment initiation. This supports the hypothesis that in classic MNV with subretinal neovascular complexes, their natural course is a partial transformation into more mature components. In contrast, in the other MNV classes this happens less frequently because in these subtypes, SHRM represents more fluidic and resolvable components. The example analysis findings differ somewhat from the findings of other analyses (based on SHRM that had not been volumetrically quantified, classified eyes using human graders, and made no attempt to determine the relative proportions of well-defined or undefined SHRM), and which demonstrated that at baseline, the largest proportion of eyes with well-defined SHRM were classified as CNV type 1, polypoidal choroidal vasculopathy (PCV) or mixed.

[0098] In the example analysis, both, undefined and well-defined SHRM, had negative associations with baseline BCVA. In the example analysis, it was found that the detrimental impact on vision to be more pronounced for well-defined SHRM on a per-unit basis, which supports the view that in well-defined SHRM fibrotic transformations have likely taken place. Nevertheless, undefined SHRM also exhibited a negative association with vision, as it is abiomarker for exudation and active MNV. Importantly, the overall visual burden related to undefined SHRM is considerable despite the small regression coefficient given the overall larger undefined SHRM volumes in treatment-naive eyes. However, longitudinally, undefined SHRM plays a negligible role as full resolution is possible on commencing anti angiogenic treatment. The example analysis revealed that all SHRM dimensions (height, en face area and volume) were negatively associated with baseline BCVA, which contrasts with other studies, where only width but not height, or height but not width were found to have a detrimental effect on vision. This contrast may be a result of the more robust and unbiased nature of measurement resulting from the biomarker quantification system 100 versus human reads, some of which even used only measurements of median-threshold rather than continuous ones, based on individually selected scans. The en face area was the measure with the greatest detrimental impact on vision in our example analysis in line with the view that the horizontal extent of the lesion is most relevant as it determines the extent of photoreceptor affected. On a per-unit basis, the measurements within the central 1 mm had the greatest association with BCVA, as would be expected from biological considerations, with that area covering the fovea. In some embodiments, the impact of the 3 mm and 6 mm diameters was even larger when making the comparison using standardized regressions. This is a surprising finding, but may on the one hand reflect that BCVA is not only a measure of central vision with patients being able to develop eccentric fixation. On the other hand, due to the inclusion criteria in the TENAYA / LUCERNE trials only central lesions are present in this dataset and not solely extrafoveal lesions that would be associated with better vision but only be captured within the larger ETDRS rings.[0991 The example analysis presents several notable strengths in characterizing SHRM in nAMD. Firstly, the biomarker quantification system 100 provides a technical improvement and technical effect in its application of deep learning to differentiate between undefined and well- defined SHRM. By distinguishing between these two subtypes of SHRM and identifying their reflectivities, the biomarker quantification system 100 provides a more nuanced understanding of their respective roles in visual impairment and potentially in fibrosis development. Employing an automated analysis allows for consistent, objective and scalable analysis, overcoming the limitations of manual segmentation, which is both time-consuming and subject to intergrader variability. In some embodiments, the continuous quantification, via the biomarker quantification system 100, utilizing volumetric and three-dimensional measurements rather than traditionalbinary reads or thresholded measurements allows for a more comprehensive and accurate representation of SHRM, which is crucial for a comprehensive understanding of its clinical significance. Also, the ability of the biomarker quantification system 100 to continuously quantify SHRM reflectivity enables the identification of subtle differences in SHRM composition that may be clinically relevant and may help distinguish between different underlying biological components. Secondly, the example analysis leverages a large and well-characterized dataset from the TENAYA and LUCERNE phase 3 clinical trials, which enhances the generalizability of our findings and provides robust statistical power for our analyses.

[0100] The results of this example analysis deepen the understanding of the different subtypes of SHRM with respect to their distribution, characteristics and associations with visual function in treatment-naive nAMD eyes. The results of this example analysis deliver further evidence for undefined SHRM to be a biomarker for exudation and active MNV, and well-defined SHRM being a biomarker for fibrotic changes in the MNV. The ability of the biomarker quantification system 100 to characterize SHRM allowed for the investigation of a large phase 3 dataset and overcame the limitations of human binary or semiquantitative gradings.

[0101] In some embodiments, the biomarker quantification system 100 and / or the process 400 is applied to longitudinal data to further establish the dynamics and relevance of undefined (UD) and well-defined (WD) SHRM. For example and in one embodiment, the biomarker quantification system, which included a U-NET model, was used in a retrospective study to examine the dynamics of UD and WD SHRM under anti angiogenic therapy and associations with best- corrected visual acuity (BCVA). In this example, the retinal segmentation model 110 was trained on AVENUE (NCT02484690) phase 2 Spectralis SD-OCT scans to segment SHRM subtypes and applied on all available 97-line Spectralis SD-OCT volume scans from the TENAYA / LUCERNE (NCT03823287 / NCT03823300) phase 3 trials in a treatment-agnostic manner. In this example, pixel- level annotations were performed on 1,499 B-scans across 334 volumes, and scans with a SHRM annotation were subsequently classified as SHRM-UD or SHRM-WD based on their delineation from the neurosensory retina (<50% or >50% of SHRM delineable, respectively). The U-Net model was trained with 1,337 scans over 150 epochs. Dice scores (sensitivity, specificity) for segmentation on the hold-out set (162 scans) were 0.711 (0.860, 0.998) and 0.839 (0.976, 0.999) for SHRM-UD and SHRM-WD, respectively. SHRM-UD and SHRM-WD volumes were computed within the ETDRS 3-mm diameter circle for morphological and the ETDRS 1-mmdiameter circle for BCVA analyses. For summary statistics, medians and interquartile range (IQR) were used due to skewed distribution. Associations with BCVA were analyzed using least squares regression. With respect to example results, on day 1 (n eyes = 1095), median volumes (Q1-Q3) of SHRM-UD and SHRM-WD were 72.5 (3.9-237.4) nl and 2.2 (0-20.3) nl, respectively. By week 12 (n eyes = 1072), SHRM-UD decreased to 0 (0-10) nl and remained low until week 48 (0, 0-2.2 nl; n eyes = 970) and week 112 (0, 0-1.9 nl; n eyes = 923), while SHRM-WD remained relatively stable (week 12: 1, 0-11.3 nl; week 48: 0.9, 0-11.8 nl; week 112: 0.9, 0-10.0 nl). On day 1, an additional 100 nl in SHRM volumes was associated with a 6.0-letter (95% CI: -7.7; -4.4; P < 001) decrease in BCVA for SHRM-UD and an 18.3-letter (95% CI: -22.6; -14.0; P <.001) decrease for SHRM-WD. In these example results, BCVA declines per SHRM subtype volume increase intensified over time. With respect to an example conclusion based on the example results, the trained biomarker quantification system 100 robustly segments and quantifies SHRM-UD and SHRM-WD in nAMD. SHRM- UD shows larger initial volumes and treatment-induced changes, while SHRM-WD has stronger negative BCVA associations. This supports the hypothesis that SHRM-UD indicates active neovascular membranes, whereas SHRM-WD suggests fibrotic processes. In some embodiments, the results of this example analysis may enable an automated, volumetric, and objective comparison of different mechanisms of action in nAMD treatments in the future and may inform better treatment strategies due to a more comprehensive understanding of the disease.III. Computer Implemented System

[0102] FIG. 8 is a block diagram of a computer system in accordance with various embodiments. The computer system 800 may be an example of one implementation for the computing platform 106 described above in FIG. 1. In one or more examples, the computer system 800 can include a bus 802 or other communication mechanism for communicating information, and a processor 804 coupled with the bus 802 for processing information. In various embodiments, the computer system 800 can also include a memory, which can be a random-access memory (RAM) 806 or other dynamic storage device, coupled to the bus 802 for determining instructions to be executed by the processor 804. Memory also can be used for storing temporary variables or other intermediate information during execution of instructions to be executed by the processor 804. In various embodiments, the computer system 800 can further include a read only memory(ROM) 808 or other static storage device coupled to the bus 802 for storing static information and instructions for the processor 804. A storage device 810, such as a magnetic disk or optical disk, can be provided and coupled to the bus 802 for storing information and instructions.

[0103] In various embodiments, the computer system 800 can be coupled via the bus 802 to a display 812, such as a cathode ray tube (CRT) or liquid crystal display (LCD), for displaying information to a computer user. An input device 814, including alphanumeric and other keys, can be coupled to the bus 802 for communicating information and command selections to the processor 804. Another type of user input device is a cursor control 816, such as a mouse, a joystick, a trackball, a gesture input device, a gaze-based input device, or cursor direction keys for communicating direction information and command selections to the processor 804 and for controlling cursor movement on the display 812. This input device 814 typically has two degrees of freedom in two axes, a first axis (e.g., x) and a second axis (e.g., y), that allows the device to specify positions in a plane. However, it should be understood that the input devices 814 allowing for three-dimensional (e.g., x, y and z) cursor movement are also contemplated herein.

[0104] Consistent with certain implementations of the present teachings, results can be provided by the computer system 800 in response to the processor 804 executing one or more sequences of one or more instructions contained in the RAM 806. Such instructions can be read into the RAM 806 from another computer-readable medium or computer-readable storage medium, such as the storage device 810. Execution of the sequences of instructions contained in the RAM 806 can cause processor 804 to perform the processes described herein. Alternatively, hard-wired circuitry can be used in place of or in combination with software instructions to implement the present teachings. Thus, implementations of the present teachings are not limited to any specific combination of hardware circuitry and software.

[0105] The term “computer-readable medium” (e.g., data store, data storage, storage device, data storage device, etc.) or "computer-readable storage medium" as used herein refers to any media that participates in providing instructions to the processor 804 for execution. Such a medium can take many forms, including but not limited to, non-volatile media, volatile media, and transmission media. Examples of non-volatile media can include, but are not limited to, optical, solid state, magnetic disks, such as the storage device 810. Examples of volatile media can include, but are not limited to, dynamic memory, such as the RAM 806. Examples of transmission mediacan include, but are not limited to, coaxial cables, copper wire, and fiber optics, including the wires that comprise the bus 802.

[0106] Common forms of computer-readable media include, for example, a floppy disk, a flexible disk, hard disk, magnetic tape, or any other magnetic medium, a CD-ROM, any other optical medium, punch cards, paper tape, any other physical medium with patterns of holes, a RAM, PROM, and EPROM, a FLASH-EPROM, any other memory chip or cartridge, or any other tangible medium from which a computer can read.

[0107] In addition to computer readable medium, instructions or data can be provided as signals on transmission media included in a communications apparatus or system to provide sequences of one or more instructions to the processor 804 of the computer system 800 for execution. For example, a communication apparatus may include a transceiver having signals indicative of instructions and data. The instructions and data are configured to cause one or more processors to implement the functions outlined in the disclosure herein. Representative examples of data communications transmission connections can include, but are not limited to, telephone modem connections, wide area networks (WAN), local area networks (LAN), infrared data connections, NFC connections, optical communications connections, etc.

[0108] It should be appreciated that the methodologies described herein, flow charts, diagrams, and accompanying disclosure can be implemented using the computer system 800 as a standalone device or on a distributed network of shared computer processing resources such as a cloud computing network.

[0109] The methodologies described herein may be implemented by various means depending upon the application. For example, these methodologies may be implemented in hardware, firmware, software, or any combination thereof. For a hardware implementation, the processing unit may be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to perform the functions described herein, or a combination thereof.

[0110] In various embodiments, the methods of the present teachings may be implemented as firmware and / or a software program and applications written in conventional programming languages such as C, C++, Python, etc. If implemented as firmware and / or software, theembodiments described herein can be implemented on a non-transitory computer-readable medium in which a program is stored for causing a computer to perform the methods described above. It should be understood that the various engines described herein can be provided on a computer system, such as the computer system 800, whereby the processor 804 would execute the analyses and determinations provided by these engines, subject to instructions provided by any one of, or a combination of, the memory components RAM 806, ROM 808, or storage device 810 and user input provided via the input device 814.IV. Example Definitions and Context[OUl] The disclosure is not limited to these example embodiments and applications or to the manner in which the example embodiments and applications operate or are described herein. Moreover, the figures may show simplified or partial views, and the dimensions of elements in the figures may be exaggerated or otherwise not in proportion.

[0112] Unless otherwise defined, scientific and technical terms used in connection with the present teachings described herein shall have the meanings that are commonly understood by those of ordinary skill in the art. Further, unless otherwise required by context, singular terms shall include pluralities and plural terms shall include the singular. Generally, nomenclatures utilized in connection with, and techniques of, chemistry, biochemistry, molecular biology, pharmacology, and toxicology are described herein are those well-known and commonly used in the art.

[0113] In addition, as the terms “on,” “attached to,” “connected to,” “coupled to,” or similar words are used herein, one element (e.g., a component, a material, a layer, a substrate, etc.) can be “on,” “attached to,” “connected to,” or “coupled to” another element regardless of whether the one element is directly on, attached to, connected to, or coupled to the other element or there are one or more intervening elements between the one element and the other element. In addition, where reference is made to a list of elements (e.g., elements a, b, c), such reference is intended to include any one of the listed elements by itself, any combination of less than all of the listed elements, and / or a combination of all of the listed elements. Section divisions in the specification are for ease of review only and do not limit any combination of elements discussed.

[0114] The term “subject” may refer to a subject of a clinical trial, a person undergoing treatment, a person undergoing anti-cancer therapies, a person being monitored for remission or recovery, a person undergoing a preventative health analysis (e.g., due to their medical history),or any other person or patient of interest. Tn various cases, “subject” and “patient” may be used interchangeably herein.

[0115] As used herein, “substantially” means sufficient to work for the intended purpose. The term “substantially” thus allows for minor, insignificant variations from an absolute or perfect state, dimension, measurement, result, or the like such as would be expected by a person of ordinary skill in the field but that do not appreciably affect overall performance. When used with respect to numerical values or parameters or classifications that can be expressed as numerical values, “substantially” means within ten percent.

[0116] As used herein, the term “about” used with respect to numerical values or parameters or classifications that can be expressed as numerical values means within ten percent of the numerical values. For example, “about 50” means a value in the range from 45 to 55, inclusive.

[0117] The term “ones” means more than one.

[0118] As used herein, the term “plurality” can be 2, 3, 4, 5, 6, 7, 8, 9, 10, or more.

[0119] As used herein, the term “set of’ means one or more. For example, a set of items includes one or more items.

[0120] As used herein, the phrase “at least one of,” when used with a list of items, means different combinations of one or more of the listed items may be used and only one of the items in the list may be used. The item may be a particular object, thing, step, operation, process, or category. In other words, “at least one of’ means any combination of items or number of items may be used from the list, but not all of the items in the list may be used. For example, without limitation, “at least one of item A, item B, or item C” means item A; item A and item B; item B; item A, item B, and item C; item B and item C; or item A and C. In some cases, “at least one of item A, item B, or item C” means, but is not limited to, two of item A, one of item B, and ten of item C; four of item B and seven of item C; or some other suitable combination.

[0121] As used herein, a “model” may include one or more algorithms, one or more mathematical techniques, one or more machine learning algorithms, or a combination thereof.

[0122] As used herein, “machine learning” may include the practice of using algorithms to parse data, learn from it, and then make a determination or prediction about something in the world. Machine learning may use algorithms that can learn from data without relying on rules-based programming. Deep learning may be one form of machine learning.

[0123] As used herein, an “artificial neural network” or “neural network” (NN) may refer to mathematical algorithms or computational models that mimic an interconnected group of artificial neurons that processes information based on a connectionistic approach to computation. Neural networks, which may also be referred to as neural nets, can employ one or more layers of nonlinear units to predict an output for a received input. Some neural networks may include one or more hidden layers in addition to an output layer. The output of each hidden layer may be used as input to the next layer in the network, i.e., the next hidden layer or the output layer. Each layer of the network generates an output from a received input in accordance with current values of a respective set of parameters. In the various embodiments, a reference to a “neural network” may be a reference to one or more neural networks.

[0124] A neural network may process information in two ways; when it is being trained it is in training mode and when it puts what it has learned into practice it is in inference (or prediction) mode. Neural networks may learn through a feedback process (e.g., backpropagation) which allows the network to adjust the weight factors (modifying its behavior) of the individual nodes in the intermediate hidden layers so that the output matches the outputs of the training data. In other words, a neural network may learn by being fed training data (learning examples) and eventually learns how to reach the correct output, even when it is presented with a new range or set of inputs. A neural network may include, for example, without limitation, at least one of a Feedforward Neural Network (FNN), a Recurrent Neural Network (RNN), a Modular Neural Network (MNN), a Convolutional Neural Network (CNN), a Residual Neural Network (ResNet), an Ordinary Differential Equations Neural Networks (neural-ODE), a U-Net, a fully convolutional network (FCN), a stacked FCN, a stacked FCN with multi-channel learning, a Squeeze and Excitation embedded neural network, a MobileNet, or another type of neural network.

[0125] As used herein, “deep learning” may refer to the use of multi-layered artificial neural networks to automatically learn representations from input data such as images, video, text, etc., without human provided knowledge, to deliver highly accurate predictions in tasks such as object detection / identification, speech recognition, language translation, etc.

[0126] As used herein, a “target retinal element” may refer to a retinal element (e.g., a retinal layer element or retinal pathological element) to be used as a biomarker by quantifying its reflectivity using methods and processes described herein.V. Recitation of Example Embodiments

[0127] Embodiment 1: A method comprising: receiving a set of optical coherence tomography (OCT) volumes for a retina of a subject, wherein each OCT volume in the set of OCT volumes corresponds to a different timepoint in a set of timepoints and comprises a set of two-dimensional OCT B-scans; generating a set of element images for each OCT volume of the set of OCT volumes in which the set of element images visually identifies a plurality of ophthalmological elements using a plurality of ophthalmological element indicators; wherein the plurality of ophthalmological element indicators assigns a different group of pixels to each ophthalmological element of the plurality of ophthalmological elements; and wherein the plurality of ophthalmological elements includes a target retinal element, a retinal pigment epithelial (RPE) layer, and a vitreous body; and identifying a reflectivity score for the target retinal element identified in each OCT volume of the set of OCT volumes using the plurality of ophthalmological elements identified by the plurality of ophthalmological element indicators in each OCT volume of the set of OCT volumes to thereby form a set of reflectivity scores. The foregoing embodiment may include one or more of the following embodiments, either alone or in combination with one another:

[0128] Embodiment 2: The method of embodiment 1, wherein generating the set of element images for each OCT volume of the set of OCT volumes comprises: generating, by a retinal segmentation model, a first segmented image for each OCT B-scan of the set of OCT B-scans for each OCT volume in the set of OCT volumes; wherein the first segmented image comprises the plurality of ophthalmological element indicators that identify the plurality of ophthalmological elements by assigning a group of pixels to each ophthalmological element of the plurality of ophthalmological elements; generating, using the first segmented image, a first element image that at least identifies the target retinal element; and generating, using the first segmented image, a second element image that identifies at least the RPE layer and an additional ophthalmological element.

[0129] Embodiment 3: The method of embodiment 2, wherein generating the second element image comprises: generating a second segmented image using the first segmented image and an ophthalmological element algorithm to identify the additional ophthalmological element, wherein the second segmented image comprises the plurality of ophthalmological element indicators that identify the additional ophthalmological element by assigning a group of pixels to the additional ophthalmological element.

[0130] Embodiment 4: The method of embodiment 3, wherein generating the second segmented image comprises using one or more ophthalmological elements identified on the first segmented image to identify the additional ophthalmological element.

[0131] Embodiment 5: The method of embodiment 4, wherein the additional ophthalmological element comprises a vitreous body.

[0132] Embodiment 6: The method of embodiment 5, wherein the one or more ophthalmological elements identified on the first segmented image comprises an internal limiting membrane (ILM) layer; and wherein the vitreous body is identified using the ILM layer as identified on the first segmented image.

[0133] Embodiment 7: The method of any one of embodiments 1-6, further comprising postprocessing the set of element images by performing a pixel erosion operation to remove a number of pixels from each element image of the set of element images from every direction, removing a number of pixels from each element image of the set of element images from one or more directions, or a combination thereof.

[0134] Embodiment 8: The method of any one of embodiments 1-7, wherein identifying the reflectivity score for the target retinal element further comprises: quantifying an intensity metric for the target retinal element; and adjusting the intensity metric for the target retinal element to account for a first reflectivity of the vitreous body and a second reflectivity of the RPE layer.

[0135] Embodiment 9: The method of embodiment 1, wherein the set of reflectivity scores for a target retinal element includes a baseline reflectivity score for a baseline timepoint in the set of timepoints; and wherein the method further comprises selecting a threshold for separating reflectivity scores that indicate low reflectivity and reflectivity scores that indicate high reflectivity based on analyzing a distribution of a plurality of baseline reflectivity scores computed for a plurality of subjects.

[0136] Embodiment 10: The method of embodiment 9, wherein the set of timepoints comprises the baseline timepoint and a first timepoint after the baseline timepoint; and wherein the set of reflectivity scores comprise a baseline reflectivity score associated with the baseline timepoint and a first reflectivity score associated with the first timepoint.

[0137] Embodiment 11 : The method of embodiment 9, wherein the set of timepoints comprises the baseline timepoint and a first timepoint after the baseline timepoint; and wherein the set ofreflectivity scores comprise a change in reflectivity between a baseline reflectivity score associated with the baseline timepoint and a first reflectivity score associated with the first timepoint.

[0138] Embodiment 12: The method of embodiment 1, wherein the set of reflectivity scores is a set of biomarkers for disease progression.

[0139] Embodiment 13: The method of embodiment 1, wherein the set of reflectivity scores is a set of biomarkers for at least treatment outcome or treatment response with respect to a selected treatment of the retina.

[0140] Embodiment 14: The method of embodiment 1, further comprising using the set of reflectivity scores to generate an output; wherein the output comprises an indication of a classification for a condition of the retina, disease progression, treatment response, or treatment outcome, at a timepoint in the set of timepoints.

[0141] Embodiment 15: The method of any one of embodiments 1-14, further comprising altering, redesigning, replicating, or assessing a clinical trial, in response to the set of reflectivity scores.

[0142] Embodiment 16: The method of any one of embodiments 1-15, wherein the target retinal element is a retinal layer element.

[0143] Embodiment 17 : The method of any one of embodiments 1-15 wherein the target retinal element is a retinal pathological element.

[0144] Embodiment 18: The method of any one of embodiments 1-15, wherein the target retinal element is a subtype of a retinal layer element or a subtype of a retinal pathological element.

[0145] Embodiment 19: The method of embodiment 17, wherein the retinal pathological element is fluid associated with pigment epithelial detachment (PED).

[0146] Embodiment 20: The method of embodiment 17 or embodiment 18, wherein the retinal pathological element is subretinal hyperreflective material (SEIRM).

[0147] Embodiment 21 : The method of embodiment 1, identifying the reflectivity score for the target retinal element identified in each OCT volume of the set of OCT volumes occurs automatically and in response to generating the set of element images for each OCT volume in the set of OCT volumes.

[0148] Embodiment 22: The method of claim 1, comprising: identifying a reflectivity score at a given timepoint for the target retinal element identified in an OCT volume of the set of OCT volumes using the plurality of ophthalmological elements identified by the plurality ofophthalmological element indicators in the OCT volume, wherein the OCT volume corresponds to the given timepoint.

[0149] Embodiment 23: A system comprising: one or more data processors; and a non- transitory computer readable storage medium containing instructions which, when executed on the one or more data processors, cause the one or more data processors to perform part or all of one or more methods disclosed in embodiments 1-22.

[0150] Embodiment 24: A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause one or more data processors to perform part or all of one or more methods disclosed in embodiments 1-22.VI. Additional Considerations

[0151] While the present teachings are described in conjunction with various embodiments, it is not intended that the present teachings be limited to such embodiments. On the contrary, the present teachings encompass various alternatives, modifications, and equivalents, as will be appreciated by those of skill in the art.

[0152] For example, the flowcharts and block diagrams described above illustrate the architecture, functionality, and / or operation of possible implementations of various method and system embodiments. Each block in the flowcharts or block diagrams may represent a module, a segment, a function, a portion of an operation or step, or a combination thereof. In some alternative implementations of an embodiment, the function or functions noted in the blocks may occur out of the order noted in the figures. For example, in some cases, two blocks shown in succession may be executed substantially concurrently. In other cases, the blocks may be performed in the reverse order. Further, in some cases, one or more blocks may be added to replace or supplement one or more other blocks in a flowchart or block diagram.

[0153] Thus, in describing the various embodiments, the specification may have presented a method and / or process as a particular sequence of steps. However, to the extent that the method or process does not rely on the particular order of steps set forth herein, the method or process should not be limited to the particular sequence of steps described, and one skilled in the art can readily appreciate that the sequences may be varied and still remain within the spirit and scope of the various embodiments.

Claims

CLAIMSWhat is claimed is:

1. A method comprising: receiving a set of optical coherence tomography (OCT) volumes for a retina of a subject, wherein each OCT volume in the set of OCT volumes corresponds to a different timepoint in a set of timepoints and comprises a set of two-dimensional OCT B- scans; generating a set of element images for each OCT volume of the set of OCT volumes in which the set of element images visually identifies a plurality of ophthalmological elements using a plurality of ophthalmological element indicators; wherein the plurality of ophthalmological element indicators assigns a different group of pixels to each ophthalmological element of the plurality of ophthalmological elements; and wherein the plurality of ophthalmological elements includes a target retinal element, a retinal pigment epithelial (RPE) layer, and a vitreous body; and identifying a reflectivity score for the target retinal element identified in each OCT volume of the set of OCT volumes using the plurality of ophthalmological elements identified by the plurality of ophthalmological element indicators in each OCT volume of the set of OCT volumes to thereby form a set of reflectivity scores.

2. The method of claim 1, wherein generating the set of element images for each OCT volume of the set of OCT volumes comprises: generating, by a retinal segmentation model, a first segmented image for each OCT B- scan of the set of OCT B-scans for each OCT volume in the set of OCT volumes; wherein the first segmented image comprises the plurality of ophthalmological element indicators that identify the plurality of ophthalmological elementsby assigning a group of pixels to each ophthalmological element of the plurality of ophthalmological elements; generating, using the first segmented image, a first element image that at least identifies the target retinal element; and generating, using the first segmented image, a second element image that identifies at least the RPE layer and an additional ophthalmological element.

3. The method of claim 2, wherein generating the second element image comprises: generating a second segmented image using the first segmented image and an ophthalmological element algorithm to identify the additional ophthalmological element, wherein the second segmented image comprises the plurality of ophthalmological element indicators that identify the additional ophthalmological element by assigning a group of pixels to the additional ophthalmological element.

4. The method of claim 3, wherein generating the second segmented image comprises using one or more ophthalmological elements identified on the first segmented image to identify the additional ophthalmological element.

5. The method of claim 4, wherein the additional ophthalmological element comprises a vitreous body.

6. The method of claim 5, wherein the one or more ophthalmological elements identified on the first segmented image comprises an internal limiting membrane (ILM) layer; and wherein the vitreous body is identified using the ILM layer as identified on the first segmented image.

7. The method of any one of claims 1-6, further comprising postprocessing the set of element images by performing a pixel erosion operation to remove a number of pixels from each element image of the set of element images from every direction, removing anumber of pixels from each element image of the set of element images from one or more directions, or a combination thereof.

8. The method of any one of claims 1-7, wherein identifying the reflectivity score for the target retinal element further comprises: quantifying an intensity metric for the target retinal element; and adjusting the intensity metric for the target retinal element to account for a first reflectivity of the vitreous body and a second reflectivity of the RPE layer.

9. The method of claim 1, wherein the set of reflectivity scores for a target retinal element includes a baseline reflectivity score for a baseline timepoint in the set of timepoints; and wherein the method further comprises selecting a threshold for separating reflectivity scores that indicate low reflectivity and reflectivity scores that indicate high reflectivity based on analyzing a distribution of a plurality of baseline reflectivity scores computed for a plurality of subjects.

10. The method of claim 9, wherein the set of timepoints comprises the baseline timepoint and a first timepoint after the baseline timepoint; and wherein the set of reflectivity scores comprise a baseline reflectivity score associated with the baseline timepoint and a first reflectivity score associated with the first timepoint.

11. The method of claim 9, wherein the set of timepoints comprises the baseline timepoint and a first timepoint after the baseline timepoint; and wherein the set of reflectivity scores comprise a change in reflectivity between a baseline reflectivity score associated with the baseline timepoint and a first reflectivity score associated with the first timepoint.

12. The method of claim 1, wherein the set of reflectivity scores is a set of biomarkers for disease progression.

13. The method of claim 1, wherein the set of reflectivity scores is a set of biomarkers for at least treatment outcome or treatment response with respect to a selected treatment of the retina.

14. The method of claim 1, further comprising using the set of reflectivity scores to generate an output; wherein the output comprises an indication of a classification for a condition of the retina, disease progression, treatment response, or treatment outcome, at a timepoint in the set of timepoints.

15. The method of any one of claims 1-14, further comprising altering, redesigning, replicating, or assessing a clinical trial, in response to the set of reflectivity scores.

16. The method of any one of claims 1-15, wherein the target retinal element is a retinal layer element.

17. The method of any one of claims 1-15, wherein the target retinal element is a retinal pathological element.

18. The method of any one of claims 1-15, wherein the target retinal element is a subtype of a retinal layer element or a subtype of a retinal pathological element.

19. The method of claim 17, wherein the retinal pathological element is fluid associated with pigment epithelial detachment (PED).

20. The method of claim 17 or claim 18, wherein the retinal pathological element is subretinal hyperreflective material (SHRM).

21. The method of claim 1, identifying the reflectivity score for the target retinal element identified in each OCT volume of the set of OCT volumes occurs automatically and in response to generating the set of element images for each OCT volume in the set of OCT volumes.

22. The method of claim 1, comprising: identifying a reflectivity score at a given timepoint for the target retinal element identified in an OCT volume of the set of OCT volumes using the plurality of ophthalmological elements identified by the plurality of ophthalmological element indicators in the OCT volume, wherein the OCT volume corresponds to the given timepoint.

23. A system comprising: one or more data processors; and a non-transitory computer readable storage medium containing instructions which, when executed on the one or more data processors, cause the one or more data processors to perform part or all of one or more methods disclosed in claims 1-22.

24. A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause one or more data processors to perform part or all of one or more methods disclosed in claims 1-22.

Images