Author Affiliations
1College of Biomedical Engineering, South-Central University for Nationalities, Wuhan, Hubei 430074, China2Key Laboratory of Cognitive Science, State Ethnic Affairs Commission, South-Central University for Nationalities, Wuhan, Hubei 430074, China3Hubei Key Laboratory of Medical Information Analysis and Tumor Diagnosis & Treatment, Wuhan, Hubei 430074, China;4School of Information Engineering, Wuhan University of Technology, Wuhan, Hubei 430070, Chinashow less
Fig. 1. Physical image of the hemoglobin mimetic solution
Fig. 2. Raw spectra of the blood samples
Fig. 3. Spectra of blood after removing singular samples
Fig. 4. NIR obtained by single pretreatment method. (a) Mean centering; (b) scaling; (c) normalization; (d) first derivative; (e) second derivative; (f) MSC; (g) SNV; (h) MWA smoothing; (i) SG smoothing
Fig. 5. Optimization of MWA parameters of blood samples
Fig. 6. RMSECV obtained by different pretreatment methods(blood)
Fig. 7. Raw spectra of the phantom solution
Fig. 8. Spectra of the phantom solution after removing singular samples
Fig. 9. Spectra of the phantom solution obtained by a single pretreatment method. (a) Mean centering; (b) scaling; (c) normalization; (d) first derivative; (e) second derivative; (f) MSC; (g) SNV; (h) MWA smoothing; (i) SG smoothing
Fig. 10. RMSECV obtained by different pretreatment methods (phantom solution)
Order | Solution | Manufacturer |
---|
1 | ultrapure water | Thermo | 2 | bovine blood hemoglobin | Biosharp | 3 | intralipid | FRESENIUS KABI SSPC |
|
Table 1. Raw material of the hemoglobin mimetic solution
Sample | Data set | Number of samples | Maximum | Minimum | Average | Standard deviation |
---|
Blood | raw set | 185 | 173 | 103 | 137.08 | 16.49 | calibration set | 139 | 173 | 103 | 137.95 | 16.40 | validation set | 46 | 173 | 106 | 134.46 | 16.66 | Hemoglobin mimetic solution | raw set | 146 | 150 | 5 | 77.50 | 42.29 | calibration set | 110 | 150 | 5 | 77.34 | 42.53 | validation set | 36 | 148 | 8 | 78.00 | 42.14 |
|
Table 2. Hemoglobin contents of different data sets
Scale scaling | Baseline correction | Scatter correction | Smoothing |
---|
mean centering | 1st derivative | MSC | MWA | Pareto scaling | 2nd derivative | SNV | SG | normalization | | | |
|
Table 3. Classification of pretreatment methods
Method | Combination method | Method | Combination method | Method | Combination method |
---|
1 | unpretreated | 8 | SNV | 15 | 1st derivative and MSC | 2 | mean centering | 9 | MWA | 16 | Pareto scaling and 1st derivative | 3 | Pareto scaling | 10 | SG | 17 | 1st derivative, MSC and MWA | 4 | normalization | 11 | Pareto scaling and MWA | 18 | Pareto scaling, MSC and MWA | 5 | 1st derivative | 12 | 1st derivative and MWA | 19 | Pareto scaling, 1st derivative and MWA | 6 | 2nd derivative | 13 | MSC and MWA | 20 | Pareto scaling, 1st derivative and MSC | 7 | MSC | 14 | Pareto scaling and MSC | 21 | Pareto scaling, 1st derivative, MSC and MWA |
|
Table 4. Combination of pretreatment methods (blood)
Method | Combination method | Method | Combination method | Method | Combination method |
---|
1 | unpretreated | 8 | SNV | 15 | 1st derivative and SNV | 2 | mean centering | 9 | MWA | 16 | mean centering and 1st derivative | 3 | Pareto scaling | 10 | SG | 17 | 1st derivative, SNV and MWA | 4 | normalization | 11 | mean centering and MWA | 18 | mean centering, SNV and MWA | 5 | 1st derivative | 12 | 1st derivative and MWA | 19 | mean centering, 1st derivative and MWA | 6 | 2nd derivative | 13 | SNV and MWA | 20 | mean centering, 1st derivative and SNV | 7 | MSC | 14 | mean centering and SNV | 21 | mean centering, 1st derivative, SNV and MWA |
|
Table 5. Combination of pretreatment methods (phantom solution)