Using the first-order, second-order, and third-order interpolation perturbation theory from quantum mechanics of modern physics to predict the CGM sensor device’s fasting plasma glucose value in the early morning, while analyzing their associated waveform shape similarities over a 14-month period based on GH-Method: math-physical medicine (No. 561)

Abstract

The author utilizes the first-order, second-order, and third-order interpolation perturbation equations from quantum mechanics of modern physics to his medical research work, which he has previously written a few medical articles on this topic. This equation is the simplest application using one selected “perturbation factor” to generate perturbed results with high prediction accuracy and waveform shape similarity. During November 2021, he has applied the statistics regression analysis model to analyze the relationship among many biomarkers where he wrote ~20 medical articles. In this study, he predicts his fasting plasma glucose (FPG) value from his body temperature (BT) and body weight (BW) in the early morning. As a comparison, he chose two separate perturbation values of 0.4 for BT perturbation factor and 0.6 for BW perturbation factor to calculate and predict his FPG value. He then compares the predicted FPG dataset against his measured sensor FPG dataset. This comparison study also contains the following two final measurement yardsticks to confirm the usefulness of the perturbed method. The first yardstick is to verify the prediction accuracies of the perturbed FPG dataset against his measured FPG dataset. The second yardstick is to examine the waveform shape similarity via the calculated correlation coefficient between the predicted or perturbed, FPG curve and the measured FPG curve. In summary, the purpose of this study is to investigate the prediction accuracy and the waveform shape similarities between a perturbed or predicted FPG waveform and his measured FPG waveform over a 14-month period, which is a total of 420 days from 10/1/2020 to 11/24/2021. He utilizes the first-order, second-order, and third-order of interpolation perturbation equations with two different perturbation factor values of 0.4 for BT and 0.6 for BW. The author has selected these two slightly different slopes, i.e., perturbation values, to study the sensitivity results. The two conclusions drawn from this research work are listed as follows: First, the 3 perturbation equations for the BT and BW cases offered an identical match for waveform shape similarity with a 100% correlation coefficient. Second, the 3 perturbation equations also provided extremely high prediction accuracies as outlined: Prediction Accuracy for BT Case (Perturbation Value 0.4) First-order: Accuracy = 96% Second-order: Accuracy = 95% Third-order: Accuracy = 94%Prediction Accuracy for BW case (Perturbation Value 0.6) First-order: Accuracy = 98% Second-order: Accuracy = 96% Third-order: Accuracy = 95% All 6 prediction accuracies are higher than 94%. It seems that the higher the slope (perturbation value), the higher prediction accuracy will be achieved.