Techniques for detecting atrial fibrillation

A system, comprising:

• a wearable ring device comprising a ring-shaped housing configured to be worn on a finger of a user, the wearable ring device comprising:
• an inner curved surface and an outer curved surface, wherein at least a portion of the inner curved surface is configured to contact a tissue of the finger of the user;
• one or more sensors configured to acquire physiological data from the user, the physiological data comprising temperature data and motion data, wherein the one or more sensors include one or more temperature sensors arranged on the inner curved surface of the ring-shaped housing and one or more photoplethysmogram PPG) sensors arranged on the inner curved surface of the ring-shaped housing;
• and
• one or more processors disposed at least partially within the ring-shaped housing and electrically coupled with the one or more sensors, wherein the one or more processors are individually or collectively configured to:
• generate physiological data associated with the user based at least in part on light received by the one or more PPG sensors;
• identify that a measurement trigger condition for atrial fibrillation has been satisfied, wherein the measurement trigger condition is satisfied based on one or more values of the temperature data being above a temperature threshold and one or more values of the motion data being below a motion threshold;
• and
• supply a first power level to one or more PPG sensors of the wearable ring device based on the measurement trigger condition being satisfied, wherein the one or more PPG sensors emit light using the supplied first power level to sample PPG data of the user for one or more time periods;
• a curved battery disposed at least partially within the ring-shaped housing, the curved battery electrically coupled with the one or more temperature sensors, the one or more PPG sensors, and the one or more processors;
• a communication module electrically coupled with the one or more processors, the communication module configured to transmit the physiological data generated by the one or more processors;
• a user device communicatively coupled with the wearable ring device;
• and
• one or more additional processors communicatively coupled with the wearable ring device and the user device, wherein the one or more additional processors are individually or collectively configured to:
• input, to one or more first machine learning classifiers, a set of PPG pulses associated with the sampled PPG data;
• classify, by the one or more first machine learning classifiers, one or more atrial fibrillation states based at least in part on one or more characteristics of the set of PPG pulses, the one or more atrial fibrillation states indicative of a relative risk of atrial fibrillation occurrences for the user;
• receive, from the user device, an indication of one or more behaviors of the user associated with triggering of the atrial fibrillation occurrences;
• predict, by one or more second machine learning identifiers, one or more atrial fibrillation occurrences, wherein the one or more second machine learning identifiers are trained according to the indication of the one or more behaviors of the user;
• and
• transmit a signal configured to cause a graphical user interface of the user device to display an indication of the one or more atrial fibrillation states, an indication of the one or more atrial fibrillation occurrences, or both.

A non-transitory computer-readable medium storing code, the code comprising instructions executable by one or more processors to:

• acquire, via one or more sensors of a wearable ring device, physiological data from a user, the physiological data comprising temperature data and motion data, wherein the wearable ring device comprises:
• a ring-shaped housing having an inner curved surface and an outer curved surface, wherein at least a portion of the inner curved surface is configured to contact a tissue of a finger of the user;
• one or more temperature sensors arranged on the inner curved surface of the ring-shaped housing;
• one or more photoplethysmogram (PPG) sensors arranged on the inner curved surface of the ring-shaped housing;
• one or more processors disposed at least partially within the ring-shaped housing, the one or more processors electrically coupled with the one or more temperature sensors and the one or more PPG sensors, the one or more processors configured to generate physiological data associated with the user based at least in part on light received by the one or more PPG sensors;
• a curved battery disposed at least partially within the ring-shaped housing, the curved battery electrically coupled with the one or more temperature sensors, the one or more PPG sensors, and the one or more processors;
• and
• a communication module electrically coupled with the one or more processors, the communication module configured to transmit the physiological data generated by the one or more processors;
• identify, by the one or more processors of the wearable ring device, that a measurement trigger condition for atrial fibrillation has been satisfied, wherein the measurement trigger condition is satisfied based on one or more values of the temperature data being above a temperature threshold and one or more values of the motion data being below a motion threshold;
• supply, by the one or more processors, a first power level to one or more PPG sensors of the wearable ring device based on the measurement trigger condition being satisfied, wherein the one or more PPG sensors emit light using the supplied first power level to sample PPG data of the user for one or more time periods;
• input, to one or more first machine learning classifiers, a set of PPG pulses associated with the sampled PPG data;
• classify, by the one or more first machine learning classifiers, one or more atrial fibrillation states based at least in part on one or more characteristics of the set of PPG pulses, the one or more atrial fibrillation states indicative of a relative risk of atrial fibrillation occurrences for the user;
• receive, via a user device, an indication of one or more behaviors of the user associated with triggering of the atrial fibrillation occurrences;
• predict, by one or more second machine learning identifiers, one or more atrial fibrillation occurrences, wherein the one or more second machine learning identifiers are trained according to the indication of the one or more behaviors of the user;
• and
• cause a graphical user interface of the user device to display an indication of the one or more atrial fibrillation states, an indication of the one or more atrial fibrillation occurrences, or both.