Glucose monitoring system
Disclosed herein are systems and methods for non-invasively monitoring blood glucose levels with a level of accuracy that is high enough to replace invasive methods such as finger prick devices and others. In some examples, glucose values are determined using a patient's electrocardiogram (ECG) signals. Additionally, glucose values may additionally be determined using an impedance spectroscopy based method and then combined with glucose values determined using an ECG waveform to output a blood glucose value.
1. A system for monitoring glucose values of a patient, the system comprising:
at least two electrodes configured to output electrical data;
a display;
a memory;
a control system coupled to the memory comprising one or more processors, the control system configured to execute a machine executable code to cause the control system to:
receive a set of electrical data output from the at least two electrodes;
process the set of electrical data to output a set of electrocardiogram (ECG) features; and
process the set of ECG features to output a set of clean ECG features;
process the set of clean ECG features to output an ECG-based blood glucose level;
store the ECG-based blood glucose level in the memory; and
process an impedance-based glucose value and the ECG-based blood glucose level to output an adapted glucose value;
wherein process the set of ECG features to output a set of clean ECG features comprises eliminating ECG features in the set of ECG features that are outside a threshold; and
wherein process the set of clean ECG features to output the ECG-based blood glucose level comprises determining the ECG-based blood glucose level based on a function of amplitude difference ratios of the set of clean ECG features and time difference ratios of the set of clean ECG features.
2. The system of claim 1 , wherein the control system is further configured to display the ECG-based blood glucose level and/or the adapted glucose value on the display.
3. The system of claim 1 , wherein the set of ECG features comprises amplitude differences between ECG points.
4. The system of claim 1 , wherein the set of ECG features comprises a QS amplitude difference, a QR amplitude difference, a TR amplitude difference, and an ST amplitude difference; wherein the QS amplitude difference is a difference between Q and S ECG points; wherein the QR amplitude difference is a difference between Q and R ECG points; wherein the TR amplitude difference is a difference between T and R ECG points; and wherein the ST amplitude difference is a difference between S and T ECG points.
5. The system of claim 4 , wherein the set of ECG features further comprises a Q−T interval and a Q−S interval.
6. The system of claim 5 , wherein process the set of ECG features further comprises process the ECG features according to an equation C1*(the QS Amplitude difference/the QR amplitude difference+the TR Amplitude difference/the ST Amplitude difference)+C2*(the Q−T interval/the Q−S interval); wherein C1 and C2 are determined by experimentation using a data set output from an invasive glucose meter reading.
7. The system of claim 1 , further comprising an accelerometer configured to output acceleration data and wherein the control system is further configured to:
receive a set of acceleration data output from the accelerometer; and determine an orientation of the patient.
8. The system of claim 1 , further comprising an electronic control system connected to the at least two electrodes and configured to control a voltage applied to the at least two electrodes.
9. The system of claim 8 , wherein the voltage comprises a sinusoidal voltage.
10. The system of claim 8 , wherein the control system is further configured to:
provide instructions to the electronic control system to apply a sinusoidal voltage to the at least two electrodes;
determine a set of impedance measurements based on electrical data output from the at least two electrodes; and
process the set of impedance measurements to output the impedance-based glucose value.
11. The system of claim 10 , wherein apply the sinusoidal voltage comprises applying a set of sinusoidal frequencies in a range of 10 Mhz-200 Mhz.
12. The system of claim 10 , wherein process the set of impedance measurements to output the impedance-based glucose value further comprises calibration with temperature values, amplitudes of QRS complex values, and heart rate values using a linear equation.
13. The system of claim 10 , wherein apply the sinusoidal voltage comprises applying a set of sinusoidal frequencies in a range of 1 Mhz-160 MHz.
14. The system of claim 1 , wherein process the impedance-based glucose value and the ECG-based blood glucose level to output the adapted glucose value comprises combining the impedance-based glucose value and the ECG-based glucose level with a linear formula.
15. The system of claim 14 , wherein the linear formula comprises summing the impedance-based glucose value multiplied by a first coefficient and the ECG-based blood glucose level multiplied by a second coefficient.
16. The system of claim 1 , wherein the process the impedance-based glucose level and the ECG-based blood glucose level to output the adapted glucose level is performed using linear regression.
17. A method for monitoring glucose values of a patient, the method comprising:
acquiring, via at least two electrodes, a set of electrical data;
receiving, via a control system, the set of electrical data from the at least two electrodes;
processing, via the control system, the set of electrical data to output a set of ECG features;
processing, via the control system, the set of ECG features to output a set of clean ECG features;
processing, via the control system, the set of clean ECG features to output an ECG-based blood glucose level;
storing the ECG-based blood glucose level in a memory coupled to the control system; and
processing an impedance-based glucose value and the ECG-based blood glucose level to output an adapted glucose value;
wherein processing the set of ECG features to output a set of clean ECG features comprises eliminating ECG features in the set of ECG features that are outside a threshold; and
wherein processing the set of clean ECG features to output the ECG-based blood glucose level comprises determining the ECG-based blood glucose level based on a function of amplitude difference ratios of the set of clean ECG features and time difference ratios of the set of clean ECG features.
18. The method of claim 17 , wherein the set of ECG features comprises a QS amplitude difference, a QR amplitude difference, a TR amplitude difference, and an ST amplitude difference; wherein the QS amplitude difference is a difference between Q and S ECG points; wherein the QR amplitude difference is a difference between Q and R ECG points; wherein the TR amplitude difference is a difference between T and R ECG points; and wherein the ST amplitude difference is a difference between S and T ECG points.
19. The method of claim 18 , wherein processing the set of ECG features further comprises processing the ECG features according to an equation C1*(the QS Amplitude difference/the QR amplitude difference+the TR Amplitude difference/the ST Amplitude difference)+C2*(a Q−T interval/a Q−S interval); wherein C1 and C2 are determined by experimentation using a data set output from an invasive glucose meter reading.