Accurate prediction of glucose concentration and identification of major contributing features from hardly distinguishable near-infrared spectroscopy. (May 2020)

• Record Type: Journal Article
• Authors: Mekonnen, Bitewulign Kassa; Yang, Webb; Hsieh, Tung-Han; Liaw, Shien-Kuei; Yang, Fu-Liang
• Is Part Of: Biomedical signal processing and control. Volume 59(2020)
• Publication Date: 2020-05
• Subjects: Near-infrared spectroscopy -- Partial least squares regression -- Support vector machine regression -- Random forest -- Extra trees regression -- Xgboost -- Principle component analysis -- Neural network; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2020.101923
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Performance of six machine learning methods for in vitro glucose concentration prediction were tested. SVMR was the most robust machine learning method for glucose concentration prediction. Major contributing features of NIR spectra data were identified and compared with different machine leaning methods. The major contributing features of NIR spectra data vary greatly for different machine learning models. Abstract: In this study, we tackle the accurate prediction of glucose aqueous concentration from hardly distinguishable near-infrared (NIR) spectroscopy. We adopted several machine learning approaches for the spectral analyses and identified important features learned by each model. The models we investigated include Partial Least Squares Regression (PLSR), Support Vector Machine Regression (SVMR), Random Forest Regression (RF), Extra Trees Regression (ETR), eXtreme Gradient Boosting (Xgboost), and hybrid Principal Component Analysis-Neural Network (PCA-NN) methods. From 47 different glucose aqueous concentrations which cover the range of 40–500 mg/dl, we measured 564 near-infrared (NIR) absorbance spectra samples with wavelength range 900 nm–2200 nm. Then the spectra samples were randomly split into 80% for the training set and 20% for the testing set. In our test, we found that the models SVMR, ETR, and PCA-NN reach extremely good performance, which had correlation coefficient R > 0.99 and determination of coefficient R 2 > 0.985. To explore the robustness of each machine learning approach, we extracted their high-weighting features and examine their distribution. We found that having large overlapping of the high-weighting features learned by the model when trained by different data sets may be an indication of model stability. In addition, our analysis came up with the essential region of features to disentangle the hardly distinguishable signal. Our study demonstrates a robust machine learning models for the prediction of glucose aqueous concentration in an in-vitro setup using near-infrared spectroscopy.