Machine-learning system for diagnosing disorders and diseases and determining drug responsiveness
Described are platforms, systems, and methods for screening patients. In one aspect, a computer-implemented method comprises: receiving, from a cellular imaging device, image data comprising calcium kinetic features of neuronal cultures derived from a patient; processing the image data through a machine-learning model to determine a diagnosis for the patient based on the calcium kinetic features, the machine-learning model trained using neuronal calcium data; and providing the diagnosis a user interface.
1. A classification system, comprising:
a user interface;
a cellular imaging device;
one or more processors; and
a non-transitory computer-readable storage device coupled to the one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, from the cellular imaging device, image data comprising calcium kinetic features of neuronal cultures derived from a patient;
processing the image data through a machine-learning model to determine a diagnosis for the patient based on the calcium kinetic features, wherein the machine-learning model is trained using neuronal calcium data; and
wherein the neuronal calcium data comprises (i) basal calcium level, peak calcium transience, calcium event frequency, and calcium event amplitude, and (ii) at least one of calcium event influx and calcium event efflux; and
providing the diagnosis to the user interface, wherein the diagnosis is for bipolar disorder (BPD), Alzheimer's disease or Parkinson's disease.
2. The classification system of claim 1 , wherein the operations comprise:
processing the image data through the machine-learning model to determine if the patient will respond to a treatment.
3. The classification system of claim 2 , wherein the treatment comprises lithium carbonate for bipolar disorder (BPD).
4. The classification system of claim 1 , wherein the diagnosis comprises whether the patient is lithium responsive or lithium non-responsive.
5. The classification system of claim 1 , wherein the neuronal calcium data is acquired from in vitro neural cultures.
6. A computer-implemented method for patient screening, the method being executed by one or more processors and comprising:
receiving, from a cellular imaging device, image data comprising calcium kinetic features of neuronal cultures derived from a patient;
processing the image data through a machine-learning model to determine a diagnosis for the patient based on the calcium kinetic features, wherein the machine-learning model is trained using neuronal calcium data and wherein the neuronal calcium data comprises (i) basal calcium level, peak calcium transience, calcium event frequency, and calcium event amplitude, and (ii) at least one of calcium event influx and calcium event efflux; and
providing the diagnosis to a user interface, wherein the diagnosis is for bipolar disorder (BPD), Alzheimer's disease or Parkinson's disease.
7. The method of claim 6 , comprising:
processing the image data through the machine-learning model to determine if the patient will respond to a treatment.
8. The method of claim 6 , comprising:
processing the image data through the machine-learning model to determine a drug responsiveness of the patient.
9. The method of claim 6 , wherein the diagnosis comprises whether the patient is lithium responsive or lithium non-responsive.
10. The method of claim 6 , wherein the machine-learning model comprises a gradient boosting classifier to prevent over fitting and not to over bias the machine-learning model.
11. The method of claim 6 , wherein the neuronal calcium data is acquired from in vitro neural cultures.
12. The method of claim 6 , wherein a portion of the neuronal calcium data is employed to train the machine-learning model and a remaining portion of the neuronal calcium data is employed to test the machine-learning model once trained, and wherein i) the portion of the neuronal calcium data comprises 70 percent of the data for training, and wherein the remaining portion of the neuronal calcium data comprises 30 percent of the data for validation, or ii) the portion of the neuronal calcium data comprises 80 percent of the data for training, and wherein the remaining portion of the neuronal calcium data comprises 20 percent of the data for validation.
13. The method of claim 6 , wherein the image data comprises intracellular calcium level traces.
14. The method of claim 6 , wherein the neuronal cultures comprise Collapsin Response Mediator Protein-2 (CRMP2)-knock out (KO), CRMP2-knock in (KI), and WT E16.5 primary hippocampal neurons.
15. The method of claim 6 , wherein the neuronal cultures are derived from blood samples from the patient.