Using machine learning to evaluate patients and control a clinical trial

A computer platform for monitoring a clinical trial comprising:

• at least one processor;
• and
• at least one memory connected with the at least one processor, wherein the at least one processor is configured to perform:
• receiving study design information for the clinical trial, the study design information including a set of parameters and corresponding parameter values related to travel constraints of a subject for the clinical trial;
• creating first training data including parameter values of actual clinical trials and actual travel scores related to participants of the clinical trials;
• training a first machine learning model using the first training data, wherein the first machine learning model includes a first neural network and is trained to predict a travel score indicative of a travel burden, and wherein the travel burden indicates a level of difficulty of travel;
• applying the study design information and the corresponding parameter values to the trained first machine learning model to calculate a predicted travel score indicative of the travel burden for the subject;
• when the predicted travel score fails to satisfy a travel score threshold:
• determining at least one suggestion for adjusting the predicted travel score, and
• outputting the at least one suggestion for adjusting the predicted travel score;
• training a second machine learning model using the first training data to produce a predicted retention rate for the clinical trial, wherein the second machine learning model includes a second neural network, and training the second machine learning model includes:
• creating second training data including at least some of the first training data:
• processing the second training data by the trained first machine learning model to produce training predicted travel scores;
• creating third training data including the training predicted travel scores from the trained first machine learning model and at least some of the second training data;
• and
• training the second machine learning model using the third training data to predict the retention rate;
• determining the predicted retention rate for the clinical trial by producing predicted travel scores for a plurality of subjects for the clinical trial by the trained first machine learning model and processing the predicted travel scores and parameter values for the plurality of subjects by the trained second machine learning model to produce the predicted retention rate;
• and
• outputting the predicted travel score and the predicted retention rate.

A non-transitory computer program product for monitoring a clinical trial, the computer program product comprising one or more computer readable storage media having program instructions collectively stored on the one or more computer readable storage media, the program instructions executable by at least one processor of a computer platform to cause the computer platform to:

• receive study design information for the clinical trial, the study design information including a set of parameters and corresponding parameter values related to travel constraints of a subject for the clinical trial;
• create first training data including parameter values of actual clinical trials and actual travel scores related to participants of the clinical trials;
• train a first machine learning model using the first training data, wherein the first machine learning model includes a first neural network and is trained to predict a travel score indicative of a travel burden, and wherein the travel burden indicates a level of difficulty of travel:
• apply the study design information and the corresponding parameter values to the trained first machine learning model to calculate a predicted travel score indicative of the travel burden for the subject;
• when the predicted travel score fails to satisfy a travel score threshold:
• determine at least one suggestion for adjusting the predicted travel score, and
• output the at least one suggestion for adjusting the predicted travel score;
• train a second machine learning model using the first training data to produce a predicted retention rate for the clinical trial, wherein the second machine learning model includes a second neural network, and training the second machine learning model includes:
• creating second training data including at least some of the first training data;
• processing the second training data by the trained first machine learning model to produce training predicted travel scores;
• creating third training data including the training predicted travel scores from the trained first machine learning model and at least some of the second training data;
• and
• training the second machine learning model using the third training data to predict the retention rate;
• determine the predicted retention rate for the clinical trial by producing predicted travel scores for a plurality of subjects for the clinical trial by the trained first machine learning model and process the predicted travel scores and parameter values for the plurality of subjects by the trained second machine learning model to produce the predicted retention rate;
• and
• output the predicted travel score and the predicted retention rate.