Intelligent ranking of clinical trials for a patient

1. A computer-implemented method comprising:

determining a plurality of clinical trials that a first patient is eligible to participate in;

determining a plurality of current attributes of the first patient;

determining a plurality of predicted attributes of the first patient for a future point in time;

providing inputs to a machine learning (ML) model based on the plurality of current attributes of the first patient, the plurality of predicted attributes of the first patient, and the plurality of clinical trials, wherein the ML model has been trained through a supervised learning process comprising:

generating a training data set comprising historical patient attributes associated with labels comprising characteristics of historical clinical trials; and

using the training data set to train the ML model to output fitness measures indicating strengths of association between given sets of patient attributes and given clinical trials associated with the characteristics;

receiving outputs from the ML model in response to the inputs, the outputs comprising a respective fitness measure for each respective clinical trial of the plurality of clinical trials, with respect to the first patient; and

ranking the plurality of clinical trials based on the outputs from the ML model.

2. The computer-implemented method of claim 1 , wherein determining the plurality of predicted attributes comprises:

identifying a plurality of patients that are clinically similar to the first patient;

determining, for each respective patient of the plurality of patients, a respective progression of a first disorder; and

determining a predicted progression of the first disorder for the first patient, based on the respective progressions for the plurality of patients.

3. The computer-implemented method of claim 1 , wherein determining the plurality of current attributes of the first patient comprises:

determining a treatment history of the first patient; and

determining whether the treatment history of the first patient includes one or more unexpected events.

4. The computer-implemented method of claim 1 , wherein determining the plurality of current attributes of the first patient comprises, for each respective clinical trial of the plurality of clinical trials:

determining a respective treatment to be used in the respective clinical trial; and

determining whether the first patient has already undergone the respective treatment.

5. The computer-implemented method of claim 1 , wherein the ML model is trained in part by:

identifying a corpus of prior trials;

determining a relative importance of a plurality of attributes for each of the prior trials in the corpus of prior trials; and

training the ML model based on the relative importance of the plurality of attributes.

6. The computer-implemented method of claim 1 , the method further comprising: upon determining that a fitness measure of a first trial of the plurality of clinical trials is below a predefined threshold, indicating that the first patient is not eligible for the first trial.

7. A system, comprising one or more processors and a computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by the one or more computer processors to perform a method, the method comprising:

determining a plurality of clinical trials that a first patient is eligible to participate in;

determining a plurality of current attributes of the first patient;

determining a plurality of predicted attributes of the first patient for a future point in time;

providing inputs to a machine learning (ML) model based on the plurality of current attributes of the first patient, the plurality of predicted attributes of the first patient, and the plurality of clinical trials, wherein the ML model has been trained through a supervised learning process comprising:

generating a training data set comprising historical patient attributes associated with labels comprising characteristics of historical clinical trials; and

using the training data set to train the ML model to output fitness measures indicating strengths of association between given sets of patient attributes and given clinical trials associated with the characteristics;

receiving outputs from the ML model in response to the inputs, the outputs comprising a respective fitness measure for each respective clinical trial of the plurality of clinical trials, with respect to the first patient; and

ranking the plurality of clinical trials based on the outputs from the ML model.

8. The system of claim 7 , wherein determining the plurality of predicted attributes comprises:

identifying a plurality of patients that are clinically similar to the first patient;

determining, for each respective patient of the plurality of patients, a respective progression of a first disorder; and

determining a predicted progression of the first disorder for the first patient, based on the respective progressions for the plurality of patients.

9. The system of claim 7 , wherein determining the plurality of current attributes of the first patient comprises:

determining a treatment history of the first patient; and

determining whether the treatment history of the first patient includes one or more unexpected events.

10. The system of claim 7 , wherein determining the plurality of current attributes of the first patient comprises, for each respective clinical trial of the plurality of clinical trials:

determining a respective treatment to be used in the respective clinical trial; and

determining whether the first patient has already undergone the respective treatment.

11. The system of claim 7 , wherein the ML model is trained in part by:

identifying a corpus of prior trials;

determining a relative importance of a plurality of attributes for each of the prior trials in the corpus of prior trials; and

training the ML model based on the relative importance of the plurality of attributes.

12. The system of claim 7 , the method further comprising:

upon determining that a fitness measure of a first trial of the plurality of clinical trials is below a predefined threshold, indicating that the first patient is not eligible for the first trial.

13. A computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by one or more computer processors to perform a method, the method comprising:

determining a plurality of clinical trials that a first patient is eligible to participate in;

determining a plurality of current attributes of the first patient;

determining a plurality of predicted attributes of the first patient for a future point in time;

providing inputs to a machine learning (ML) model based on the plurality of current attributes of the first patient, the plurality of predicted attributes of the first patient, and the plurality of clinical trials, wherein the ML model has been trained through a supervised learning process comprising:

generating a training data set comprising historical patient attributes associated with labels comprising characteristics of historical clinical trials; and

using the training data set to train the ML model to output fitness measures indicating strengths of association between given sets of patient attributes and given clinical trials associated with the characteristics;

receiving outputs from the ML model in response to the inputs, the outputs comprising a respective fitness measure for each respective clinical trial of the plurality of clinical trials, with respect to the first patient; and

ranking the plurality of clinical trials based on the outputs from the ML model.

14. The computer-readable storage medium of claim 13 , wherein determining the plurality of predicted attributes comprises:

identifying a plurality of patients that are clinically similar to the first patient;

determining, for each respective patient of the plurality of patients, a respective progression of a first disorder; and

determining a predicted progression of the first disorder for the first patient, based on the respective progressions for the plurality of patients.

15. The computer-readable storage medium of claim 13 , wherein determining the plurality of current attributes of the first patient comprises:

determining a treatment history of the first patient; and

determining whether the treatment history of the first patient includes one or more unexpected events.

16. The computer-readable storage medium of claim 13 , wherein determining the plurality of current attributes of the first patient comprises, for each respective clinical trial of the plurality of clinical trials:

determining a respective treatment to be used in the respective clinical trial; and

determining whether the first patient has already undergone the respective treatment.

17. The computer-readable storage medium of claim 13 , wherein the ML model is trained in part by:

identifying a corpus of prior trials;

determining a relative importance of a plurality of attributes for each of the prior trials in the corpus of prior trials; and

training the ML model based on the relative importance of the plurality of attributes.

18. The computer-readable storage medium of claim 13 , the method further comprising: upon determining that a fitness measure of a first trial of the plurality of clinical trials is below a predefined threshold, indicating that the first patient is not eligible for the first trial.