Influencing Factors of Mid-Infrared Spectrum Blood Glucose Detection

Yansong Yue; Zhushanying Zhang; Sicong Zhu; Huimin Cao; Dongyun Zheng; Qinlan Xie

doi:10.3788/LOP231035

Laser & Optoelectronics Progress, Volume. 60, Issue 24, 2430001(2023)

Influencing Factors of Mid-Infrared Spectrum Blood Glucose Detection

Yansong Yue^1,2,3, Zhushanying Zhang^1,2,3、*, Sicong Zhu^1,2,3, Huimin Cao^1,2,3, Dongyun Zheng^1,2,3, and Qinlan Xie^1,2,3

Author Affiliations

¹School of Biomedical Engineering, South-Central Minzu University, Wuhan 430074, Hubei, China

²Key Laboratory of Cognitive Science, State Ethnic Affairs Commission, South-Central Minzu University, Wuhan 430074, Hubei, China

³Hubei Key Laboratory of Medical Information Analysis and Tumor Diagnosis＆Treatment, Wuhan 430074, Hubei, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(9)

Fig. 1. Original measurement spectra. (a) Model 1; (b) model 2; (c) model 3; (d) model 4

Download full size

Fig. 2. Flow chart of mid-infrared spectroscopy glucose detection modeling and analysis

Download full size

Fig. 3. SG smooth spectra. (a) Model 1; (b) model 2; (c) model 3; (d) model 4

Download full size

Fig. 4. Clarke Error Grid based on four subset models. (a) Model 1; (b) model 2; (c) model 3; (d) model 4

Download full size

Fig. 5. Predicted values versus true values. (a) Model 1; (b) model 2; (c) model 3; (d) model 4

Download full size

Table 1. Mass concentration range of substances
View table
Table 1. Mass concentration range of substances
Index Glucose Albumin Urea Cholesterol
Mass concentration 0‒480 0‒6000 0‒90 0‒200
Mass concentration gradient 20 2000 30 50

Table 2. Composition of background solution samples

View table

Table 2. Composition of background solution samples

Sample number	Composition
1	4.00 mL 10% intralipid solutions
2	0.25 mL ethanol and 3.75 mL 10% intralipid solutions
3	0.50 mL ethanol and 3.50 mL 10% intralipid solutions
4	0.75 mL ethanol and 3.25 mL 10% intralipid solutions
5	1.00 mL ethanol and 3.00 mL 10% intralipid solutions

Table 3. Model performance metrics based on different subsets
View table
Table 3. Model performance metrics based on different subsets
Model RMSECV f R_p RMSEP
1 39.7370 10 0.9785 40.0187
2 17.7817 12 0.9042 175.7292
3 25.1170 12 0.9616 103.6627
4 53.6867 14 0.9742 38.6716

Table 4. Percentage of points for different models located in corresponding areas of the Clarke Error Grid
View table
Table 4. Percentage of points for different models located in corresponding areas of the Clarke Error Grid
Model A B C D E
1 85.7% — — 14.3% —
2 40.0% 40.0% 20.0% — —
3 42.9% 42.9% — 14.3% —
4 100.0% — — — —

Tools

Get Citation

Copy Citation Text

Yansong Yue, Zhushanying Zhang, Sicong Zhu, Huimin Cao, Dongyun Zheng, Qinlan Xie. Influencing Factors of Mid-Infrared Spectrum Blood Glucose Detection[J]. Laser & Optoelectronics Progress, 2023, 60(24): 2430001

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites