# US 11,896,382 B2 · Hyperspectral image-guided ocular imager for alzheimer's disease pathologies > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,896,382 B2 - Document ID: US11896382B2 - Status: Granted - Kind: B2 - Application number: 16/766,880 - Filed: 2018-11-27 - Published: 2020-10-15 - Granted: 2024-02-13 - Earliest claimed: 2017-11-27 - Assignees: Retispec Inc. - Applicants: Retispec Inc. - Inventors: Eliav Shaked; Grzegorz Dmochowski - Web page: https://patentgrep.com/patent/US11896382B2 ## CPC - A61B 3/12: for looking at the eye fundus, e.g. ophthalmoscopes (A61B3/13 takes precedence) - A61B 5/4088: {Diagnosing of monitoring cognitive diseases, e.g. Alzheimer, prion diseases or dementia} - A61B 3/0025: {characterised by electronic signal processing, e.g. eye models} - A61B 3/1225: {using coherent radiation} - A61B 3/14: Arrangements specially adapted for eye photography - A61B 5/0075: {by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy (A61B5/0071 takes precedence)} - A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems} - A61B 5/7267: {involving training the classification device} - G01N 1/286: {involving mechanical work, e.g. chopping, disintegrating, compacting, homogenising (microtomes G01N1/06; pulverising in general B02C; mixing in general B01F)} - G01N 1/30: Staining; Impregnating {; Fixation; Dehydration; Multistep processes for preparing samples of tissue, cell or nucleic acid material and the like for analysis} - G01N 21/55: Specular reflectivity - G01N 21/65: Raman scattering - G01N 33/6896: {Neurological disorders, e.g. Alzheimer's disease} - A61B 2576/02: specially adapted for a particular organ or body part - G01N 2001/2873: {Cutting or cleaving} - G01N 2333/4709: Amyloid plaque core protein - G01N 2800/2821: Alzheimer - G16H 30/40: for processing medical images, e.g. editing - G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q) - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA ## Abstract A non-invasive device for detecting Alzheimer's Disease-associated pathologies, comprising a fundus camera comprising a spectral reflectance imaging unit that includes a broadband light source and a light sensor and a lens assembly to focus light from the broadband light source onto a fundus of an eye, and one or more processors to detect reflected and/or backscattered light from the eye, illuminated by the broadband light source, using the light sensor for determining spectral reflectance information, generate a plurality of spectral reflectance maps comprising counts of reflected and/or backscattered light at a different wavelength or a different range of wavelengths, assign a weight to each of the spectral reflectance maps, and determine one or more regions of interest from the weighted spectral reflectance maps as being a potential Alzheimer's Disease-associated pathology based on a detection of one or more biomarkers indicative of Tauopathy. ## Claims ### Claim 1 A non-invasive in vivo ocular light-based detection system comprising: a fundus camera, comprising: a spectral reflectance imaging unit that includes a broadband light source and a light sensor; and a lens assembly configured to focus light from the broadband light source onto a fundus of an eye of a subject; and one or more processors configured to execute instructions stored in a memory to: control the spectral reflectance imaging unit to illuminate the fundus of the eye using the broadband light source, and detect resulting reflected and/or backscattered light from the eye using the light sensor for determining spectral reflectance information; generate a plurality of spectral reflectance maps, wherein each of the plurality of spectral reflectance maps comprise counts of reflected and/or backscattered light at a different wavelength or a different range of wavelengths; assign a weight to each of the plurality of spectral reflectance maps; determine one or more region of interest (ROI) from the weighted spectral reflectance maps as being a potential Alzheimer's Disease (AD)-associated pathology based on a detection of one or more biomarkers indicative of Tauopathy; and classify, using the spectral reflectance information, the subject as having one or more AD-associated pathologies, the one or more AD-associated pathologies associated with the one or more biomarkers indicative of Tauopathy, wherein the one or more biomarkers indicative of Tauopathy comprise Tau neurofibrillary tangles. ### Claim 2 (depends on 1) The detection system of claim 1 further comprising a Raman spectroscopy unit that includes a laser and a spectrometer, wherein the one or more processors are further configured to control the Raman spectroscopy unit to illuminate each of the one or more ROI using the laser, and detect Raman scattered light from the eye resulting from the laser and using the spectrometer for determining Raman spectroscopy information, such that the subject is further classified as having one or more AD-associated pathologies based on the Raman spectroscopy information. ### Claim 3 (depends on 2) The detection system of claim 2, wherein the classifying of the subject as having the one or more AD-associated pathologies is further based on previous spectral reflectance information and/or Raman spectroscopy information of the subject stored in the memory or in another device. ### Claim 4 (depends on 2) The detection system of claim 2, wherein the classifying of the subject as having the one or more AD-associated pathologies is further based on changes in the spectral reflectance information and/or the Raman spectroscopy information of the subject over time. ### Claim 5 (depends on 1) The detection system of claim 1, wherein the classifying of the subject as having the one or more AD-associated pathologies comprises classifying the subject as having a plurality of the AD-associated pathologies. ### Claim 6 (depends on 2) The detection system of claim 2, wherein the one or more processors are further configured to: determine a baseline ROI from the spectral reflectance information as being background tissue that does not contain the potential AD-associated pathology; control the Raman spectroscopy unit to illuminate the baseline ROI of the eye using the laser; and detect light from the eye resulting from the laser using the spectrometer for determining Raman spectroscopy information of the background tissue, wherein the classifying is further based on comparing the Raman spectroscopy information of the potential AD-associated pathology with the Raman spectroscopy information of the background tissue. ### Claim 7 (depends on 1) The detection system of claim 1, wherein the one or more AD-associated pathologies include two or more of the AD-associated pathologies associated with the one or more biomarkers indicative of Tauopathy. ### Claim 8 (depends on 1) The detection system of claim 1, wherein the one or more AD-associated pathologies include neuritic or glial cytopathology of the eye of the subject, or vascular characteristics of blood vessels or choroid of the eye of the subject. ### Claim 9 (depends on 2) The detection system of claim 2, wherein when the one or more AD-associated pathologies are associated with Amyloid Beta deposits, the classifying is based on analyzing Raman spectroscopy information at a wavenumber shift or shifts in a range of 1600 cm 1 to 1700 cm 1, which correspond to Raman vibrational resonances of Alpha-helix and Beta-pleated sheets. ### Claim 10 (depends on 2) The detection system of claim 2, wherein when the one or more AD-associated pathologies are associated with the Tau neurofibrillary tangles, wherein the classifying is based on analyzing Raman spectroscopy information at a wavenumber shift or shifts in a range of 1600 cm 1 to 1700 cm 1, which corresponds to Raman vibrational resonance of phosphorylated-Taus. ### Claim 11 (depends on 1) The detection system of claim 1, wherein the one or more processors use a machine learning algorithm for one or both of: the determining of the one or more ROI; or the classifying of the subject as having one or more AD-associated pathologies. ### Claim 12 (depends on 2) The detection system of claim 2, wherein the one or more processors are further configured to classify, from the Raman spectroscopy information, the subject as having: AD, or a precursor to AD, or a pre-screened classification for potential AD that requires further investigation, or responsiveness to treatment or intervention. ### Claim 13 (depends on 1) The detection system of claim 1, wherein exogenous fluorescing agents, dyes, or tracers are not required for the classifying of the one or more AD-associated pathologies. ### Claim 14 (depends on 2) The detection system of claim 2, further comprising one or more optical filters to filter out a wavelength of the laser prior to detection by the spectrometer. ### Claim 15 (depends on 2) The detection system of claim 2, wherein the one or more processors are configured to determine a respective size of each of the one or more ROI from the weighted spectral reflectance maps, and control of the Raman spectroscopy unit to emit the laser onto each of the ROI of the eye having the respective size. ### Claim 16 (depends on 2) The detection system of claim 2, wherein the Raman spectroscopy unit is controlled by the one or more processors to perform, for each of the one or more ROI, scanning of the respective ROI using the laser of the Raman spectroscopy unit for the determining of the Raman spectroscopy information. ### Claim 17 (depends on 1) The detection system of claim 1, wherein the light sensor comprises: a 2-dimensional array of light sensors, each light sensor sensitive to a range of wavelengths of light; and a 2-dimensional filter array that overlays the 2-dimensional array of light sensors, each individual filter selectively transmits light of a specific wavelength. ### Claim 18 (depends on 1) The detection system of claim 1, wherein the one or more biomarkers indicative of Tauopathy are a protein aggregate comprising Tau neurofibrillary tangles. ### Claim 19 (depends on 1) The detection system of claim 1, wherein AD-associated pathologies include additional protein aggregates comprising one or more Amyloid Beta deposits, soluble Amyloid Beta aggregates, or Amyloid precursor protein. ### Claim 20 A non-invasive in vivo ocular light-based detection system comprising: a first imaging modality; a second imaging modality; and one or more processors configured to execute instructions stored in a memory to: control the first imaging modality to illuminate a wide field-of-view of a fundus of an eye of a subject, and detect light from the eye for determining a first imaging modality information; determine one or more region of interest (ROI) from the first imaging modality information as being a potential Alzheimer's Disease (AD)-associated pathology based on a detection of one or more biomarkers indicative of AD; control the second imaging modality to illuminate each of the one or more ROI, and detect light from the eye for determining second imaging modality information; and classify, using the first imaging modality information and the second imaging modality information, the subject as having one or more AD-associated pathologies, the one or more AD-associated pathologies associated with protein aggregates, the protein aggregates including Tau neurofibrillary tangles. ### Claim 21 (depends on 20) The detection system of claim 20, wherein the first imaging modality is a reflectance system comprising a hyperspectral or multispectral camera and the second imaging modality is a Raman spectroscopy device. ### Claim 22 A method of non-invasive in vivo detection of one or more Alzheimer's Disease (AD)-associated pathologies from an eye of a subject, comprising: controlling a spectral reflectance imaging unit to illuminate a fundus of the eye using a light source; detecting light from the eye resulting from the light source using a light sensor for determining spectral reflectance information; determining, using one or more processors utilizing a first neural network, a location of one or more region of interest (ROI) from the spectral reflectance information as being a potential AD-associated pathology based on a detection of one or more biomarkers indicative of Tauopathy; illuminating the fundus of the eye only in each of the one or more ROI, and detecting light from the one or more ROI for determining imaging information; and classifying, with one or more processors utilizing a second neural network and using at least the imaging information, the subject as having one or more AD-associated pathologies, the one or more AD-associated pathologies associated with the one or more biomarkers indicative of Tauopathy, wherein the one or more biomarkers indicative of Tauopathy comprise Tau neurofibrillary tangles. ### Claim 23 (depends on 22) The method of claim 22 further comprising controlling a Raman spectroscopy unit to illuminate each of the one or more ROI using a laser; and detecting Raman scattered light from the eye resulting from the laser using a spectrometer for determining Raman spectroscopy information; wherein the subject is further classified as having one or more AD-associated pathologies based on the Raman spectroscopy information. ### Claim 24 (depends on 1) The detection system of claim 1, wherein the one or more processors are further configured to execute instructions stored in a memory to: determine one or more ROI from the spectral reflectance information as being a potential AD-associated pathology based on a detection of a plurality of biomarkers indicative of AD; and classify, using the spectral reflectance information the subject as having one or more AD-associated pathologies, the one or more AD-associated pathologies including the plurality of biomarkers indicative of AD, wherein the plurality of biomarkers indicative of AD comprise Tau neurofibrillary tangles, Amyloid Beta deposits, soluble Amyloid Beta aggregates, and Amyloid precursor protein. ### Claim 25 (depends on 20) The detection system of claim 20, wherein the one or more processors are further configured to: generate a plurality of spectral reflectance maps, wherein each of the plurality of spectral reflectance maps comprise counts of reflected and/or backscattered light at a different wavelength or a different range of wavelengths; and select a subset of the plurality of spectral reflectance maps based on a pre-determined criteria, wherein: the first imaging modality information comprises the subset of the plurality of spectral reflectance maps. ### Claim 26 (depends on 22) The method of claim 22 further comprising: generating a plurality of spectral reflectance maps, wherein each of the plurality of spectral reflectance maps comprise counts of reflected and/or backscattered light at a different wavelength or a different range of wavelengths; and assigning a weight to each of the plurality of spectral reflectance maps, wherein: the location of the one or more ROI is determined from the plurality of weighted spectral reflectance maps.