DeepSpectra: An end-to-end deep learning approach for quantitative spectral analysis

• Published in: Analytica chimica acta Vol. 1058; pp. 48 - 57
• Main Authors: Zhang, Xiaolei, Lin, Tao, Xu, Jinfan, Luo, Xuan, Ying, Yibin
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 13-06-2019; Elsevier BV
• Subjects: Artificial neural networks; Calibration; Chemometrics; Convolutional neural network; Corn; Data analysis; Datasets; Deep learning; Inception; Learning theory; Model accuracy; Near infrared radiation; Neural networks; Preprocessing; Repeatability; Spectral analysis; Support vector machines; Tablets; Wheat; Repeatability; Inception; Convolutional neural network; Chemometrics; Model accuracy
• ISSN: 0003-2670; 1873-4324
• DOI: 10.1016/j.aca.2019.01.002

Learning patterns from spectra is critical for the development of chemometric analysis of spectroscopic data. Conventional two-stage calibration approaches consist of data preprocessing and modeling analysis. Misuse of preprocessing may introduce artifacts or remove useful patterns and result in worse model performance. An end-to-end deep learning approach incorporated Inception module, named DeepSpectra, is presented to learn patterns from raw data to improve the model performance. DeepSpectra model is compared to three CNN models on the raw data, and 16 preprocessing approaches are included to evaluate the preprocessing impact by testing four open accessed visible and near infrared spectroscopic datasets (corn, tablets, wheat, and soil). DeepSpectra model outperforms the other three convolutional neural network models on four datasets and obtains better results on raw data than in preprocessed data for most scenarios. The model is compared with linear partial least square (PLS) and nonlinear artificial neural network (ANN) methods and support vector machine (SVR) on raw and preprocessed data. The results show that DeepSpectra approach provides improved results than conventional linear and nonlinear calibration approaches in most scenarios. The increased training samples can improve the model repeatability and accuracy.
[Display omitted]
•DeepSpectra with the Inception module is developed for quantitative spectral analysis.•DeepSpectra outperforms other CNN approaches on raw spectra analysis.•Preprocessing strategies have little positive impact on DeepSpectra model performance.•DeepSpectra on raw data comparable to the best calibration approach on preprocessing.•The model repeatability and accuracy improved with increased sample sizes.