Hyperspectral image-guided ocular imager for Alzheimer's Disease pathologies

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 hyperspectral reflectance imaging unit to illuminate a wide field-of-view of a fundus of the eye using a broadband light source;
• detecting light from the eye resulting from the broadband light source using a hyperspectral camera for determining hyperspectral reflectance information;
• determining, using one or more processors, a location of one or more region of interest (ROI) from the hyperspectral reflectance information as being a potential AD-associated pathology;
• controlling a Raman spectroscopy unit to illuminate each of the one or more ROI using a laser;
• detecting Raman scattered light from the eye resulting from the laser using a spectrometer for determining Raman spectroscopy information;
• and
• classifying, using the one or more processors, using the hyperspectral reflectance information and the Raman spectroscopy information, the subject as having one or more AD-associated pathologies, the one or more AD-associated pathologies including protein aggregates, the protein aggregates including at least one of Tau neurofibrillary tangles, Amyloid Beta deposits, soluble Amyloid Beta aggregates, or Amyloid precursor protein.

A computer program product by a machine learning training process, the computer program product comprising instructions stored in a non-transitory computer readable medium which, when executed by a computer, causes the computer to carry out non-invasive in vivo detection of one or more Alzheimer's Disease (AD)-associated pathologies from an eye of a subject, the machine learning training process comprising:

• training, using one or more processors, the computer program using verified training data, the verified training data obtained by:
• slicing an ex vivo tissue sample from a subject into tissue slices;
• placing the tissue slices onto slides;
• staining a first tissue slice of a first slide;
• providing a second slide having a second tissue slice that was adjacent to the first tissue slice in the tissue sample and is unstained;
• verifying that the stained first tissue slice has one or more of the AD-associated pathologies using histology;
• performing at least one imaging modality on the second slide to obtain imaging information;
• and
• classifying the imaging information as being indicative of one or more of the AD-associated pathologies, the one or more AD-associated pathologies including protein aggregates, the protein aggregates including at least one of Tau neurofibrillary tangles, Amyloid Beta deposits, soluble Amyloid Beta aggregates, or Amyloid precursor protein.

A method for machine learning training of a computer program stored in a memory which, when executed by a computer, causes the computer to carry out non-invasive in vivo detection of one or more Alzheimer's Disease (AD)-associated pathologies from an eye of a subject, the method comprising:

• training, using one or more processors, the computer program using verified training data, the verified training data obtained by:
• slicing an ex vivo tissue sample from a subject into tissue slices;
• placing the tissue slices onto slides;
• staining a first tissue slice of a first slide;
• providing a second slide having a second tissue slice that was adjacent to the first tissue slice in the tissue sample and is unstained;
• verifying that the stained first tissue slice has one or more of the AD-associated pathologies using histology;
• performing at least one imaging modality on the second slide to obtain detection information;
• and
• classifying the detection information as being indicative of one or more of the AD-associated pathologies, the one or more AD-associated pathologies including protein aggregates, the protein aggregates including at least one of Tau neurofibrillary tangles, Amyloid Beta deposits, soluble Amyloid Beta aggregates, or Amyloid precursor protein;
• and
• storing the trained computer program to the memory.