Systems and methods for using a wearable health monitor
Inventors
MacEachern, Stuart P. • Jenkel, Martin W.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A system and method for using algorithms and acoustic input to control, monitor, annotate, and configure a wearable patch that monitors physiological signals. The method comprises the receiving a transmission at a wearable device wherein the transmission can identify a mode of operation, an annotation to the data in the device, a request for data from the device, a modification of the configuration of parameters, and signals in the device.
Core Innovation
The invention relates to a wearable health monitor that stores biometric data and a contextual record of events that includes predetermined keywords uttered by a patient. The keywords are picked up by at least one audio source and recognized as keywords descriptive of a physical or mental condition affecting the patient, and the wearable health monitor triggers keyword and biometric data to be written to the contextual record rather than treating audio and biometric measurements as separate streams.
The contextual record of events includes timestamp data specifically identifying a time at which a predetermined keyword was uttered as well as the time biometric data was measured. This allows correlation of biometric data and predetermined keywords so that spoken condition-descriptive events can be temporally aligned with corresponding biometric measurements. The architecture further supports enriching the contextual record with additional contextual information such as social determinant of health data, also correlated with timestamps.
The invention further includes analyzing patient/environmental sounds with acoustic input together with biometric inputs, using a neural network processor and machine learning model to make inferences from signals corresponding to biometric data and acoustic data. The system also describes a telemedicine architecture in which a telemedicine provider initiates a telemedicine session with at least bidirectional audio and manipulates the wearable health monitor using voice commands. Voice keywords can also be used to control which monitoring actions occur, including triggering inferences, alarms, configuration, and external event signaling.
Claims Coverage
The partial content includes two independent claims, covering a wearable health monitor that correlates predetermined voice keywords with measured biometric data using a contextual record with timestamps, and a method of monitoring biometric data using a network-connected wearable within a telemedicine session with bidirectional audio and provider voice-command manipulation.
Voice-keyword contextual event record correlated with biometric measurements
A wearable health monitor stores biometric data from at least one biometric sensor and stores a contextual record of events comprising predetermined keywords uttered by a patient, picked up by an audio source and recognized as keywords descriptive of a physical or mental condition affecting the patient, wherein utterance of a predetermined keyword triggers the keyword and biometric data to be written to the contextual record.
Timestamped correlation between keyword utterance time and biometric measurement time
The contextual record of events includes timestamp data specifically identifying a time at which a predetermined keyword was uttered as well as the time biometric data was measured, allowing correlation of biometric data and predetermined keywords.
Network-connected wearable with acoustic input and telemedicine bidirectional audio controlled by voice commands
A method of monitoring biometric data provides a network-connected, wearable health monitor configured to accept voice commands and comprising at least one acoustic input and at least one biometric sensor, attaches the wearable health monitor to a patient, and at a second location has a telemedicine provider initiate a telemedicine session comprising at least bidirectional audio with the patient, the telemedicine provider manipulating the wearable health monitor using voice commands.
Together, the independent claims define a wearable that recognizes predetermined spoken keywords and records them in a contextual event record with biometric data and timestamps to enable correlation, and they define a monitoring method that uses a network-connected wearable in a telemedicine session where a provider uses bidirectional audio and voice commands to manipulate the wearable.
Stated Advantages
Allows correlation of biometric data and predetermined keywords via timestamped contextual record of events.
Enables telemedicine provider manipulation of the wearable health monitor using voice commands during a telemedicine session with bidirectional audio.
Documented Applications
Telemedicine use case where a telemedicine provider initiates a telemedicine session with at least bidirectional audio with the patient and manipulates a network-connected wearable health monitor using voice commands.
Interested in licensing this patent?