Patient-specific glucose prediction systems and methods

A non-transitory computer-readable medium having instructions stored thereon that are executable by a processing system to perform a method comprising:

• obtaining current measurement data for a physiological condition of a patient provided by a sensing arrangement;
• obtaining a user input indicative of one or more future events associated with the patient;
• and
• in response to the user input:
• determining a prediction of the physiological condition of the patient in the future based at least in part on the current measurement data and the one or more future events using one or more prediction models associated with the patient, wherein determining the prediction comprises:
• determining hourly forecast values for the physiological condition using an hourly forecasting model associated with the patient based at least in part on the current measurement data and the one or more future events, the hourly forecasting model comprising a neural network including a plurality of cells, wherein each cell corresponds to a respective hourly interval and at least one of the plurality of cells is configured to output an average value for the physiological condition during the respective hourly interval in the future based at least in part on a subset of the one or more future events predicted to occur within the respective hourly interval in the future;
• determining a second plurality of predicted values for the physiological condition of the patient in the future based at least in part on the current measurement data using a second prediction model different from the hourly forecasting model;
• determining, for each of the hourly forecasting model and the second prediction model, a weighting factor associated with the respective prediction model based at least in part on the one or more future events;
• and
• determining an ensemble prediction for the physiological condition of the patient with respect to time in the future based at least in part on the hourly forecast values, the second plurality of predicted values, and respective weighting factors associated with the hourly forecasting model and the second prediction model;
• and
• displaying a graphical representation of the ensemble prediction with respect to time on a display device.

An electronic device comprising:

• a communications interface to receive current measurement data for a physiological condition of a patient from a sensing arrangement;
• a display device having a graphical user interface display presented thereon;
• a user interface to obtain a user input indicative of one or more future events;
• and
• a control system coupled to the communications interface, the display device, and the user interface to determine hourly forecast values for the physiological condition using an hourly forecasting model associated with the patient based at least in part on the current measurement data and the one or more future events, determine a second plurality of predicted values for the physiological condition of the patient in the future based at least in part on the current measurement data using a second prediction model different from the hourly forecasting model, determine, for each of the hourly forecasting model and the second prediction model, a weighting factor associated with the respective prediction model based at least in part on the one or more future events, determine an ensemble prediction of the physiological condition of the patient in the future based at least in part on the hourly forecast values, the second plurality of predicted values and respective weighting factors associated with the hourly forecasting model and the second prediction model, and display a graphical representation of the ensemble prediction within the graphical user interface display on the display device, wherein:
• the hourly forecasting model comprises a neural network including a plurality of cells;
• each cell corresponds to a respective hourly interval;
• and
• at least one of the plurality of cells is configured to output an average value for the physiological condition during the respective hourly interval in the future based at least in part on a subset of the one or more future events predicted to occur within the respective hourly interval in the future.