System and method of using body temperature logging patch
Continuous temperature monitors can be used to collect frequent temperature data over large periods of time. A machine learning system trained with collected temperature data can be used to predict future temperature data for patients. Such predictions can be clinically beneficial for disease states in which fever can be fatal, particularly for those fevers having fast onset. For example, patients undergoing chimeric antigen receptor T-cell (CAR-T) therapy may suffer fevers caused by cytokine release syndrome (CRS). Accordingly, a continuous temperature monitor is used to collect temperature data from CAR-T patients with high risk of CRS (or like patients with high fever risk), and the collected temperature data processed by a machine learning system to predict the patient's future temperature profiles. Results of analysis of the predicted temperature profiles may then be provided to the patient and/or clinician.
1. A method comprising:
receiving temperature data from a continuous temperature monitor worn by a patient, the continuous temperature monitor comprising a flexible patch configured to be attached to the patient's skin, the flexible patch comprising a temperature sensor that is configured to automatically collect at least one temperature of the patient every five minutes during a collection time period;
inputting the received temperature data to a machine learning system, the machine learning system being trained to output a predictive temperature profile of the patient based upon the received temperature data, the predictive temperature profile comprising predicted temperature values of the patient at a plurality of corresponding times in a future time period;
analyzing the predictive temperature profile for an indication of a fever during at least one of the plurality of times in the future time period; and
transmitting a notification of a result of the analysis to a personal device of the patient and/or to a clinician,
wherein the notification comprises an alert when the result of the analysis indicates the fever, and
wherein the notification is transmitted in substantially real time with the receipt of the temperature data and result of the analysis.
2. The method of claim 1 , wherein the flexible patch comprises a wireless transmitter, and wherein the machine learning system is remotely connected to the continuous temperature monitor and to the clinician via a network.
3. The method of claim 1 , wherein the machine learning system is trained with temperature data from subjects having a same medical condition as the patient.
4. The method of claim 3 , wherein the same medical condition is cytokine release syndrome.
5. The method of claim 1 , further comprising correcting received temperature data points having a value below a predetermined threshold prior to inputting the received temperature data to the machine learning system.
6. The method of claim 1 , wherein analyzing the predictive temperature profile for the fever comprises comparing the predicted temperature values to a predetermined threshold.
7. The method of claim 1 , wherein analyzing the predictive temperature profile for the fever comprises comparing a rate of increase of the predictive temperature profile to a predetermined threshold.
8. The method of claim 1 , wherein the collection time period is three hours long and the future time period is one hour long.
9. The method of claim 1 , wherein the notification further comprises an instruction to the patient to seek medical attention when the result of the analysis indicates the fever.
10. The method of claim 1 , wherein the machine learning system comprises a multi-step multi-layer perceptron network.
11. The method of claim 1 , wherein the machine learning system comprises a univariate multi-step convolutional neural network.
12. The method of claim 1 , wherein the machine learning system comprises a long-short-term memory model.
13. The method of claim 1 , wherein the machine learning system comprises a sequence to sequence model.
14. The method of claim 1 , further comprising:
when the result of the analysis indicates the fever, treating the medical condition of the patient prior to the corresponding time in the future time period at which the fever is predicted to occur.
15. The method of claim 1 , wherein analyzing the predictive temperature profile for the fever comprises determining a period of time the predictive temperature profile is greater than a predetermined threshold temperature.
16. The method of claim 1 , wherein the patient is undergoing or has undergone chimeric antigen receptor T-cell (CAR-T) therapy, and the fever is caused by cytokine release syndrome.
17. The method of claim 1 , wherein the notification further comprises a time series graph of the predicted temperature values at the plurality of corresponding times in the future time period.
18. The method of claim 1 , wherein the collection time period is thirty minutes long and the future time period is one hour long.