GH-METHODS

Math-Physical Medicine

NO. 049

Comparison of two Fasting Plasma Glucose measurements including Finger-piercing and continuous Sensor monitoring (Math-Physical Medicine)

Corresponding Author: Gerald C. Hsu, eclaireMD Foundation, USA.

Introduction
This paper discusses fasting plasma glucose (FPG) measurement results from two different methods, finger piercing and testing strip (Finger) and Libre’s continuous glucose monitoring device (Sensor).

Method
The author has been collecting a total of 1,825 FPG data by Finger in the early mornings from 1/1/2014 to 12/31/2018 (1,825 days) on a daily basis.

Recently, he has further collected 2,651 FPG data by applying a sensor on his upper arm to collect his glucose values continuously.  This sensor measurement is conducted in parallel with his routine finger measurement.  During the period of 5/5/2018 to 12/31/2018 (241 days), he has recorded his sensor fasting glucose values about 11 times per day (from 00:30am to 07:45 am).

Results
A standard FPG waveform can be clearly observed from 11 data points per day.  Generally speaking, FPG waves are calm and resemble an ocean wave, moving up and down with the trough occurring in the deep sleep stage (3am to 5am).  In comparison, PPG waves are more violent, reflecting like tsunami waves (Figure 1).

Figure 1: Comparison between synthesized FPG and synthesized PPG waveforms

Using time-series analysis of predicted FPG curve (based on math-physical medicine approach), finger-piercing and test strip (Finger), and continuous glucose monitor device (Sensor), the results reveal the following observed conclusions:

(1) The predicted FPG has 98.8% accuracy on predicting Finger FPG values, and their correlation coefficient (R) is extremely high 92% (Figures 2 & 3).

(2) The predicted FPG has 96.6% accuracy on predicting Sensor FPG values, but their correlation coefficient (R) is very low 16%.  (Figures 2 & 3).

Figure 2: Visual check of daily and 90-days average of three FPG curves (Prediction, Finger, Sensor)
Figure 3:Accuracy and Correlation between Predicted FPG and measured FPG (both Finger and Sensor)

(3) Correlation analyses of both Weight vs. Finger and Weight vs. Sensor show that their respective correlation coefficients are 80% (based on data from 5/5/2014 to 12/31/2018) and 81% (based on data from 5/5/2018 to 12/31/2018).  This shows that body weight is the primary contributing or influential factor of FPG formation (Figure 4).

Figure 4: Weight vs. FPG (both Finger and Sensor)

Conclusion
Both prediction accuracy and weight correlation are extremely high and as expected by the author, except for the low correlation existed between the predicted FPG and Sensor FPG. Nevertheless, in comparison with the author’s previous FPG findings, most of the conclusions regarding FPG characteristics are still holding true.

The sensor method is relatively new for type 2 diabetes (T2D) patients. There is limited information available regarding the difference between blood samples taken from finger vs. upper arm.  Besides, both Finger strip and arm Sensor device have a relatively high defect rate from the points of view of both mathematician and professional engineer.

The author’s present glucose prediction model is developed upon current mainstream glucose measurement device, i.e. finger-piercing and testing strip.  If and when some sort of continuous monitoring sensor device becomes a standard glucose monitoring device for majority of T2D patients, then another set of complete and rigorous data analysis and model tuning will be required in order to provide a practical and yet also highly accurate glucose prediction model.