Methods and systems for recommending filters to apply to clinical trial search results using machine learning techniques

1. A method for cognitive clinical trial filtration, the method comprising:

receiving a first set of search filters that were applied to a first plurality of clinical trials with respect to a first patient;

determining one or more attributes of the first patient;

training a machine learning (ML) model based on the first set of filters and the one or more attributes of the first patient;

receiving a selection of a second patient;

determining one or more attributes of the second patient;

generating a base search result set for the second patient based on the determined one or more attributes of the second patient, the base search result set including a set of potentially relevant clinical trials for the second patient;

generating a second set of search filters by processing the one or more attributes of the second patient using the trained ML model, wherein the second set of search filters comprise search filters recommended for searching for relevant clinical trials for the second patient based on filters used to search for relevant clinical trials for historical patients having similar attributes to the determined one or more attributes of the second patient, and wherein the second set of search filters comprise search filters having a highest probability of relevance in a probability distribution over a universe of candidate search filters; and

generating a refined search result set by applying the generated second set of search filters to the base search result set.

2. The method of claim 1 , the method further comprising:

determining one or more characteristics of at least a first trial of the first plurality of clinical trials, wherein the one or more characteristics includes at least one of: a phase of the trial or a type of the trial; and

training the ML model based on the one or more characteristics of the first trial.

3. The method of claim 1 , the method further comprising:

determining operational data for at least a first trial of the first plurality of clinical trials, wherein the operational data includes at least one of: a site associated with the trial or one or more other trials that are also associated with the site; and

training the ML model based on the operational data for the first trial.

4. The method of claim 1 , wherein generating the refined search result set comprises automatically applying at least a first filter of the second set of search filters to a list of potential clinical trials for the second patient.

5. The method of claim 4 , the method further comprising:

determining a subset of the list of potential clinical trials that remain after application of the first filter;

selecting at least a second filter of the second set of search filters to apply to the subset of the list of potential clinical trials, based on determining that the second filter is most valuable to further divide the subset of the list of potential clinical trials.

6. The method of claim 5 , wherein determining that the second filter is most valuable to further divide the subset of the list of potential clinical trials comprises determining that the second filter will reduce a number of potential clinical trials remaining in the subset of the list of potential clinical trials to a minimum amount, as compared to other filters in the second set of search filters.

7. The method of claim 1 , the method further comprising:

determining, for at least a first potential trial of a list of potential clinical trials for the second patient, a likelihood that the second patient will complete the first potential trial.

8. The method of claim 7 , wherein determining the likelihood that the second patient will complete the first potential trial comprises:

identifying one or more prior studies that are clinically similar to the first potential trial;

identifying, for at least a first prior study of the one or more prior studies, one or more prior patients that are clinically similar to the second patient and participated in the first prior study; and

determining whether the one or more prior patients successfully completed the first prior study.

9. The method of claim 1 , wherein the first set of search filters were applied by a first healthcare professional at a first institution, and wherein training the ML model is further based on data collected from at least one other healthcare professional at the first institution.

10. The method of claim 1 , wherein the first set of search filters were applied by a first healthcare professional, and wherein training the ML model is further based on data collected from the first healthcare professional.

11. A system, comprising:

a processor; and

a memory having instructions stored thereon which, when executed by the processor, performs an operation for cognitive clinical trial filtration, the operation comprising:

receiving a first set of search filters that were applied to a first plurality of clinical trials with respect to a first patient;

determining one or more attributes of the first patient;

training a machine learning (ML) model based on the first set of filters and the one or more attributes of the first patient;

receiving a selection of a second patient;

determining one or more attributes of the second patient;

generating a base search result set for the second patient based on the determined one or more attributes of the second patient, the base search result set including a set of potentially relevant clinical trials for the second patient;

generating a second set of search filters by processing the one or more attributes of the second patient using the trained ML model, wherein the second set of search filters comprise search filters recommended for searching for relevant clinical trials for the second patient based on filters used to search for relevant clinical trials for historical patients having similar attributes to the determined one or more attributes of the second patient, and wherein the second set of search filters comprise search filters having a highest probability of relevance in a probability distribution over a universe of candidate search filters; and

generating a refined search result set by applying the generated second set of search filters to the base search result set.

12. The system of claim 11 , wherein the operation further comprises:

determining one or more characteristics of at least a first trial of the first plurality of clinical trials, wherein the one or more characteristics includes at least one of: a phase of the trial or a type of the trial; and

training the ML model based on the one or more characteristics of the first trial.

13. The system of claim 11 , wherein the operation further comprises:

determining operational data for at least a first trial of the first plurality of clinical trials, wherein the operational data includes at least one of: a site associated with the trial or one or more other trials that are also associated with the site; and

training the ML model based on the operational data for the first trial.

14. The system of claim 11 , wherein generating the refined search result set comprises: automatically applying at least a first filter of the second set of search filters to a list of potential clinical trials for the second patient.

15. The system of claim 11 , wherein the operation further comprises:

determining, for at least a first potential trial of a list of potential clinical trials for the second patient, a likelihood that the second patient will complete the first potential trial by:

identifying one or more prior studies that are clinically similar to the first potential trial;

identifying, for at least a first prior study of the one or more prior studies, one or more prior patients that are clinically similar to the second patient and participated in the first prior study; and

determining whether the one or more prior patients successfully completed the first prior study.

16. A non-transitory computer-readable storage medium having instructions stored thereon which, when executed by a processor, performs an operation for cognitive clinical trial filtration, the operation comprising:

receiving a first set of search filters that were applied to a first plurality of clinical trials with respect to a first patient;

determining one or more attributes of the first patient;

training a machine learning (ML) model based on the first set of filters and the one or more attributes of the first patient;

receiving a selection of a second patient;

determining one or more attributes of the second patient;

generating a base search result set for the second patient based on the determined one or more attributes of the second patient, the base search result set including a set of potentially relevant clinical trials for the second patient;

generating a second set of search filters by processing the one or more attributes of the second patient using the trained ML model, wherein the second set of search filters comprise search filters recommended for searching for relevant clinical trials for the second patient based on filters used to search for relevant clinical trials for historical patients having similar attributes to the determined one or more attributes of the second patient, and wherein the second set of search filters comprise search filters having a highest probability of relevance in a probability distribution over a universe of candidate search filters; and

generating a refined search result set by applying the generated second set of search filters to the base search result set.

17. The computer-readable storage medium of claim 11 , wherein the operation further comprises:

determine one or more characteristics of at least a first trial of the first plurality of clinical trials, wherein the one or more characteristics includes at least one of: a phase of the trial or a type of the trial; and

train the ML model based on the one or more characteristics of the first trial.

18. The computer-readable storage medium of claim 16 , wherein the operation further comprises:

determine operational data for at least a first trial of the first plurality of clinical trials, wherein the operational data includes at least one of: a site associated with the trial or one or more other trials that are also associated with the site; and

train the ML model based on the operational data for the first trial.

19. The computer-readable storage medium of claim 16 , wherein generating the refined search set comprises: automatically applying at least a first filter of the second set of search filters to a list of potential clinical trials for the second patient.

20. The computer-readable storage medium of claim 16 , wherein the operation further comprises:

determine, for at least a first potential trial of a list of potential clinical trials for the second patient, a likelihood that the second patient will complete the first potential trial by:

identifying one or more prior studies that are clinically similar to the first potential trial;

identifying, for at least a first prior study of the one or more prior studies, one or more prior patients that are clinically similar to the second patient and participated in the first prior study; and

determining whether the one or more prior patients successfully completed the first prior study.