Closed-loop system for cardiopulmonary resuscitation (CPR)

A method comprising:

• receiving, by processing circuitry and from a plurality of physiological sensors, one or more sets of data representative of a plurality of patient parameters over a period of time, wherein the plurality of patient parameters comprises a right atrium pressure and an aortic blood pressure;
• generating, by the processing circuitry and using a deep learning model, output data sets over the period of time of predicted future trajectories of at least one patient parameter of the plurality of patient parameters, each predicted future trajectory of the predicted future trajectories representative of a predicted future state of the patient;
• iteratively determining, by the processing circuitry and based on at least one output data set of the predicted future trajectories, different sets of values for a plurality of control parameters, wherein the plurality of control parameters comprises an oscillation frequency, a duty cycle of compression, and compression depth;
• and
• controlling, by the processing circuitry and using the different sets of values for the plurality of control parameters over the period of time, a compression mechanism to apply pressure to a torso region of the patient, wherein the compression mechanism is configured to move according to at least one degree of freedom.

A system comprising:

• a memory comprising a deep learning model;
• and
• processing circuitry configured to:
• receive, from a plurality of physiological sensors, one or more sets of data representative of a plurality of patient parameters over a period of time, wherein the plurality of patient parameters comprises a right atrium pressure and an aortic blood pressure;
• generate, using the deep learning model, output data sets over the period of time of predicted future trajectories of at least one patient parameter of the plurality of patient parameters, each predicted future trajectory of the predicted future trajectories representative of a predicted future state of the patient;
• iteratively determine, based on at least one output data set of the predicted future trajectories, different sets of values for a plurality of control parameters, wherein the plurality of control parameters comprises an oscillation frequency, a duty cycle of compression, and compression depth;
• and
• control, using the different sets of values for the plurality of control parameters over the period of time, the compression mechanism to apply the pressure to the torso region of the patient.