Clinical trial support apparatus, operation method of clinical trial support apparatus, and operation program of clinical trial support apparatus

An operation method of a clinical trial support apparatus, the method comprising:

• predicting a progression speed of a target disease for subjects of a clinical trial;
• dividing the subjects into a plurality of groups according to a prediction result of the progression speed of the target disease;
• and
• allocating the subjects to a treatment group to which a test drug is administered and a placebo group to which a placebo is administered for each of the plurality of groups,
• wherein a processor predicts the progression speed of the target disease by using a trained model,
• the trained model is a trained model that has been trained by using a medical image showing an organ of a target disease patient at a first time point as learning input data and a numerical value indicating a degree of progression of the target disease of the target disease patient at a second time point after a lapse of a period equivalent to a clinical trial period from the first time point as correct answer data, and
• the processor
• inputs a medical image showing an organ of the subject at a start time point of the clinical trial to the trained model, and
• outputs a numerical value indicating a degree of progression of the target disease of the subject at an end time point of the clinical trial as the prediction result from the trained model.

A non-transitory computer-readable storage medium storing an operation program of a clinical trial support apparatus, the program causing a computer to execute:

• predicting a progression speed of a target disease for subjects of a clinical trial;
• dividing the subjects into a plurality of groups according to a prediction result of the progression speed of the target disease;
• and
• allocating the subjects to a treatment group to which a test drug is administered and a placebo group to which a placebo is administered for each of the plurality of groups,
• wherein the predicting the progression speed of the target disease is performed using a trained model,
• the trained model is a trained model that has been trained by using a medical image showing an organ of a target disease patient at a first time point as learning input data and a numerical value indicating a degree of progression of the target disease of the target disease patient at a second time point after a lapse of a period equivalent to a clinical trial period from the first time point as correct answer data,
• a medical image showing an organ of the subject at a start time point of the clinical trial is inputted to the trained model, and
• a numerical value indicating a degree of progression of the target disease of the subject at an end time point of the clinical trial as the prediction result is outputted from the trained model.