Calibration of pulse-transit-time to blood pressure model using multiple physiological sensors and various methods for blood pressure variation

A blood pressure measurement system, comprising:

• one or more memories configured to store instructions;
• and
• one or more processors configured to execute the instructions to perform operations, the operations comprising:
• determining, based on inertial data obtained using one or more inertial sensors, whether a user associated with the blood pressure measurement system is stationary;
• in response to determining the user associated with the blood pressure measurement system is stationary based on the inertial data and determining a predetermined condition is satisfied, automatically initiating, by the blood pressure measurement system, a calibration operation during which a user holds a limb of the user at a plurality of positions, the calibration operation including:
• obtaining a plurality of blood pressure measurements of the user using one or more blood pressure devices which determine the plurality of blood pressure measurements independently of a pulse transit time,
• obtaining a plurality of hydrostatic pressure measurements based on altitude data obtained using the one or more inertial sensors and one or more altimeters,
• obtaining a plurality of pulse transit time (PTT) measurements based on proximal pulse wave data and the distal pulse wave data obtained using a plurality of biometric sensors,
• obtaining calibration data points corresponding to the plurality of blood pressure measurements, the plurality of pulse transit times measurements, and the plurality of hydrostatic pressure measurements, wherein
• at least one first calibration data point among the calibration data points is obtained while the user holds the limb at a first position among the plurality of positions and at least one second calibration data point among the calibration data points is obtained while the user holds the limb at a second position among the plurality of positions, and
• the one or more altimeters, the one or more inertial sensors, the plurality of biometric sensors, and the one or more blood pressure devices are attached to a same area of the limb of the user when the plurality of blood pressure measurements are obtained, and
• training a PTT-to-blood pressure machine-learned model based on the plurality of calibration data points to obtain a recalibrated PTT-to-blood pressure machine-learned model, the recalibrated PTT-to-blood pressure machine-learned model relating blood pressure to PTT and hydrostatic pressure, and
• subsequent to the calibration operation, implementing the recalibrated PTT-to-blood pressure machine-learned model to predict a blood pressure value in response to measuring a PTT, associated with the user, using the plurality of biometric sensors.

One or more non-transitory computer-readable media configured to store instructions that, when executed by one or more processors of a blood pressure measurement system, cause the one or more processors to perform operations, the operations comprising:

• determining based on inertial data obtained using one or more inertial sensors, whether a user associated with the blood pressure measurement system is stationary;
• in response to determining the user associated with the blood pressure measurement system is stationary based on the inertial data and determining a predetermined condition is satisfied, automatically initiating, by the blood pressure measurement system, a calibration operation during which a user holds a limb of the user at a plurality of positions, the calibration operation including the operations of:
• obtaining a plurality of blood pressure measurements of the user using one or more blood pressure devices which determine the plurality of blood pressure measurements independently of a pulse transit time;
• obtaining a plurality of hydrostatic pressure measurements based on altitude data obtained using the one or more inertial sensors and one or more altimeters;
• obtaining a plurality of pulse transit time (PTT) measurements based on proximal pulse wave data and the distal pulse wave data obtained using a plurality of biometric sensors;
• obtaining calibration data points corresponding to the plurality of blood pressure measurements, the plurality of pulse transit times measurements, and the plurality of hydrostatic pressure measurements, wherein
• at least one first calibration data point among the calibration data points is obtained while the user holds the limb at a first position among the plurality of positions and at least one second calibration data point among the calibration data points is obtained while the user holds the limb at a second position among the plurality of positions, and
• the one or more altimeters, the one or more inertial sensors, the plurality of biometric sensors, and the one or more blood pressure devices are attached to a same area of the limb of the user when the plurality of blood pressure measurements are obtained;
• and
• training a PTT-to-blood pressure machine-learned model based on the plurality of calibration data points to obtain a recalibrated PTT-to-blood pressure machine-learned model, the recalibrated PTT-to-blood pressure machine-learned model relating blood pressure to PTT and hydrostatic pressure, and
• subsequent to the calibration operation, implementing the recalibrated PTT-to-blood pressure machine-learned model to predict a blood pressure value in response to measuring a PTT, associated with the user, using the plurality of biometric sensors.