Medical system for diagnosing cognitive disease pathology and/or outcome
A medical system useful in the determination of future disease progression in a subject. More specifically the present invention applies machine learning techniques to aid prediction of disease pathology and clinical outcomes in subjects presenting with symptoms of cognitive decline and to expedite clinical development of novel therapeutics.
1. An apparatus comprising:
memory storing instructions; and
a processor to execute the instructions to implement:
a first trained learning machine trained on first clinical data related to patient cognitive status to predict a brain disease pathology associated with a brain disease in a first set of subjects, the brain disease pathology including amyloid beta (AB) positivity in the brain, the first trained learning machine tuned for a clinical protocol to identify a second set of subjects from the first set of subjects, the second set of subjects having a first probability of the brain disease pathology greater than a first threshold; and
a second trained learning machine trained using a training data set created using image data and non-image data and deployed to predict a second probability of the brain disease pathology in the second set of subjects based on i) second clinical data related to patient cognitive status and ii) image-related data correlated with the second clinical data, the second trained learning machine tuned for the clinical protocol to select a third set of subjects from the second set of subjects, the third set of subjects having the second probability of the brain disease greater than a second threshold; and
a network interface to generate an output to trigger execution of the clinical protocol for the third set of subjects by arranging a computer system to display an indication and include the third set of subjects classified as a cohort for treatment, the treatment including administering a disease modifying drug (DMD) for the brain disease to the cohort via the clinical protocol, the brain disease including at least one of mild cognitive impairment (MCI) or Alzheimer's disease (AD).
2. The apparatus of claim 1 , wherein the clinical protocol is associated with a clinical trial for a drug.
3. The apparatus of claim 1 , wherein the clinical protocol is associated with a treatment plan for a patient.
4. The apparatus of claim 1 , wherein the image-related data includes at least one of a positron emission tomography (PET) image or a magnetic resonance (MR) image.
5. The apparatus of claim 1 , wherein the image-related data includes data related to quantified regions of at least one of a PET image or an MR image.
6. The apparatus of claim 1 , wherein the image-related data is first image-related data, and wherein the first trained learning machine further predicts the brain disease pathology using the first clinical data and second image-related data.
7. The apparatus of claim 1 , wherein the brain disease pathology is loss of dopamine-producing brain cells.
8. The apparatus of claim 1 , wherein at least one of the first clinical data or the second clinical data includes demographic data.
9. The apparatus of claim 1 , wherein the first trained learning machine includes at least a first machine-learned model, and wherein the second trained learning machine includes at least a second machine-learned model.
10. The apparatus of claim 9 , wherein at least one of the first machine-learned model or the second machine-learned model is rebuilt based on at least one of: i) a change in the respective first threshold or second threshold, ii) additional data related to at least one of the first clinical data or the second clinical data.
11. The apparatus of claim 10 , wherein at least one of the first machine-learned model or the second machine-learned model includes a plurality of folds, and wherein each of the plurality of folds is rebuilt to form a final model.
12. A non-transitory machine-readable medium including instructions that, when executed, cause a processor to at least:
process, using a first trained learning machine, a first set of subjects to identify a second set of subjects from the first set of subjects, the second set of subjects having a first probability of a brain disease pathology greater than a first threshold, the brain disease pathology including amyloid beta (AB) positivity in the brain, wherein the first trained learning machine is tuned for a clinical protocol and trained on first clinical data related to patient cognitive status to predict the brain disease pathology associated with a brain disease;
generate, using a second trained learning machine, a third set of subjects from the second set of subjects, the third set of subjects having a second probability of the brain disease pathology greater than a second threshold, wherein the second trained learning machine is tuned for the clinical protocol and trained using a training data set created using image data correlated with non-image data and deployed to predict the second probability of the brain disease in the second set of subjects based on i) second clinical data related to patient cognitive status and ii) image-related data; and
generate an output to trigger execution of the clinical protocol for the third set of subjects by arranging a computer system to display an indication and include the third set of subjects classified as a cohort for treatment, the treatment including administering a disease modifying drug (DMD) for the brain disease to the cohort via the clinical protocol, the brain disease including at least one of mild cognitive impairment (MCI) or Alzheimer's disease (AD).
13. The machine-readable medium of claim 12 , wherein the instructions, when executed, cause the processor to assess at least one of the first set of subjects or the second set of subjects with cognitive testing.
14. The machine-readable medium of claim 12 , wherein the instructions, when executed, cause the processor to obtain the image-related data from the second set of subjects.
15. The machine-readable medium of claim 12 , wherein the instructions, when executed, cause the processor to quantify regions in the image-related data.
16. The machine-readable medium of claim 12 , wherein the instructions, when executed, cause the processor to trigger patient treatment involving the clinical protocol.
17. The machine-readable medium of claim 12 , wherein the instructions, when executed, cause the processor to support a clinical trial using the clinical protocol.
18. The machine-readable medium of claim 12 , wherein the first trained learning machine includes at least a first machine-learned model and the second trained learning machine includes at least a second machine-learned model, and wherein the instructions, when executed, cause the processor to rebuild at least one of the first machine-learned model or the second machine-learned model based on at least one of: i) a change in the respective first threshold or second threshold, ii) additional data related to at least one of the first clinical data or the second clinical data.
19. An apparatus comprising:
memory storing instructions; and
a processor to execute the instructions to implement:
a first trained learning machine trained on first clinical data related to patient cognitive status to predict a brain disease pathology associated with a brain disease, the first set of subjects having a first probability of the brain disease pathology greater than a first threshold, the brain disease pathology including amyloid beta (AB) positivity in the brain; and
a second trained learning machine trained using a training data set created using image data and non-image data and deployed to predict a second probability of the brain disease pathology based on i) second clinical data related to patient cognitive status and ii) image-related data, the second set of subjects having the second probability of the brain disease greater than a second threshold,
the processor to form a third set of subjects based on at least one of the first set of subjects or the second set of subjects; and
a network interface to generate an output to trigger execution of a clinical protocol with the third set of subjects by arranging a computer system to display an indication and include the third set of subjects in the clinical protocol, the clinical protocol including treatment including administering a disease modifying drug (DMD) for the brain disease to the third set of subjects via the clinical protocol, the brain disease including at least one of mild cognitive impairment (MCI) or Alzheimer's disease (AD).
20. The apparatus of claim 1 , wherein the brain disease pathology is predicted using a radioactive diagnostic agent.
21. The apparatus of claim 20 , wherein the radioactive diagnostic agent includes at least one of Vizamyl™, Neuraceq™, or Amyvid™.