Real-time estimation of human core body temperature based on non-invasive physiological measurements
Inventors
Reifman, Jaques • Laxminarayan, Srinivas • Rakesh, Vineet • Ramakrishnan, Sridhar • Liu, Jianbo
Assignees
United States Department of the Army
Publication Number
US-11517203-B2
Publication Date
2022-12-06
Expiration Date
2037-08-18
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
An embodiment of the invention provides a method of estimating a body temperature of an individual where physiological data is received from at least one sensor 510. Environmental data is received and the physiological data and the environmental data are input into a model. The model generates an estimated body temperature and an estimated physiological condition based on the physiological data and the environmental data. A processor 520 compares the estimated physiological condition to a measured physiological condition in the physiological data. A controller 530 modifies at least one parameter in the model when the difference between the estimated physiological condition and the measured physiological condition is above a threshold.
Core Innovation
The invention provides a method and system for real-time estimation of human core body temperature based on non-invasive physiological measurements. The method involves receiving physiological data from one or more sensors, including heart rate, skin temperature, and activity data, as well as receiving environmental data such as ambient temperature and humidity. These data inputs are processed using a physiological mathematical model combined with a Kalman filter to generate an estimated core body temperature and estimated physiological conditions in real time.
The physiological mathematical model relates physical activity to heart rate, heart rate to core body temperature, and core body temperature to skin temperature through differential equations using parameters such as rate constants and rates of heat loss or gain. The Kalman filter compares estimated physiological conditions with measured data and modifies model parameters when the difference exceeds a threshold, thereby customizing the model dynamically to fit individual physiological responses.
The problem being solved is the lack of practical solutions to alert individuals, particularly U.S. Armed Forces personnel, of impending heat injury by reliably estimating core body temperature non-invasively. Traditional methods such as ingestible thermometer pills are invasive and impractical for continuous monitoring of many individuals, and existing data-driven models may not account for individual physiological variations. This invention addresses these limitations by providing individualized, continual, and adaptive core temperature estimation using readily available non-invasive measurements and environmental data.
Claims Coverage
The patent contains several independent claims covering methods and systems for estimating body temperature using a model with physiological data and environmental data, incorporating a Kalman filter for continual model customization and prediction.
Method of estimating body temperature using a physiological model with Kalman filter adaptation
A method comprising receiving heart rate, skin temperature, activity data, ambient temperature, and humidity; inputting these into a model with a physiological mathematical model and Kalman filter; modifying model parameters based on errors between estimated and measured physiological conditions; estimating body temperature reflecting heat balance using updated model parameters; and continual customization of the model to the individual.
System for estimating body temperature with sensors, processor, controller, and physiological model
A system including at least one physiological sensor measuring activity, heart rate, skin temperature, at least one environmental sensor, a processor that inputs data into a model generating estimated physiological conditions, a controller that modifies model parameters using a Kalman filter based on errors, and the physiological mathematical model comprising three equations relating physical activity to estimated heart rate, heart rate to estimated body temperature, and body temperature to estimated skin temperature.
Method of predicting future body temperature using an autoregressive model
A method comprising predicting a future body temperature based on estimated body temperatures generated after modification of model parameters, where the prediction uses an autoregressive model assuming constancy in activity, heart rate, and skin temperature during the prediction horizon.
The independent claims cover the core innovative concept of combining physiological and environmental data into a model with a Kalman filter to adaptively and non-invasively estimate and predict individualized core body temperature, emphasizing continual customization and prediction capabilities.
Stated Advantages
Provides non-invasive, real-time, individualized estimation of core body temperature.
Adapts continuously to an individual's physiological response to environmental and exertional heat stress.
Reduces reliance on invasive techniques like ingestible thermometer pills, enabling practical continuous monitoring.
Improves accuracy over data-driven models by incorporating individual-specific thermoregulatory responses.
Is robust to sensor failures by relying on multiple physiological measurements.
Enables prediction of core body temperature ahead of time to allow intervention and prevention of heat injury.
Documented Applications
Continuous monitoring of core body temperature for U.S. Armed Forces personnel to prevent heat injuries during deployment in hot and humid climates.
Use in fitness-tracking devices such as wristbands or watches to monitor physiological and environmental data.
Implementation in mobile computing platforms (smartphones, tablets) in wireless communication with wearable devices to perform real-time estimation and prediction of core temperature.
Development of real-time heat-injury warning systems using predicted core body temperature values.
Interested in licensing this patent?