Two Revised Deep Neural Networks and Their Applications in Quantitative Analysis Based on Near-Infrared Spectroscopy
Small data sets make developing calibration models using deep neural networks difficult because it is easy to overfit the system. We developed two deep neural network architectures by revising two existing network architectures: the U-Net and the attention mechanism. The major changes were to use 1D...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-07-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/13/14/8494 |
_version_ | 1797590396290727936 |
---|---|
author | Hong-Hua Huang Jian-Fei Luo Feng Gan Philip K. Hopke |
author_facet | Hong-Hua Huang Jian-Fei Luo Feng Gan Philip K. Hopke |
author_sort | Hong-Hua Huang |
collection | DOAJ |
description | Small data sets make developing calibration models using deep neural networks difficult because it is easy to overfit the system. We developed two deep neural network architectures by revising two existing network architectures: the U-Net and the attention mechanism. The major changes were to use 1D convolutional layers to replace the fully connected layers. We also designed and combined average pooling and maximum pooling in our revised networks, respectively. We applied these revised network architectures to three publicly available data sets and the resulting calibration models can generate acceptable results for general quantitative analysis. It also generated rather good results for data sets that concern calibration transfer. It demonstrates that constructing network architectures by properly revising existing successful network architectures may provide additional choices in the exploration of the application of deep neural network in analytical chemistry. |
first_indexed | 2024-03-11T01:19:54Z |
format | Article |
id | doaj.art-b01d2737a57f4585bb52bcd533f59166 |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-11T01:19:54Z |
publishDate | 2023-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-b01d2737a57f4585bb52bcd533f591662023-11-18T18:13:53ZengMDPI AGApplied Sciences2076-34172023-07-011314849410.3390/app13148494Two Revised Deep Neural Networks and Their Applications in Quantitative Analysis Based on Near-Infrared SpectroscopyHong-Hua Huang0Jian-Fei Luo1Feng Gan2Philip K. Hopke3School of Chemistry, Sun Yat-Sen University, Guangzhou 510006, ChinaSchool of Chemistry, Sun Yat-Sen University, Guangzhou 510006, ChinaSchool of Chemistry, Sun Yat-Sen University, Guangzhou 510006, ChinaDepartment of Public Health Sciences, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642, USASmall data sets make developing calibration models using deep neural networks difficult because it is easy to overfit the system. We developed two deep neural network architectures by revising two existing network architectures: the U-Net and the attention mechanism. The major changes were to use 1D convolutional layers to replace the fully connected layers. We also designed and combined average pooling and maximum pooling in our revised networks, respectively. We applied these revised network architectures to three publicly available data sets and the resulting calibration models can generate acceptable results for general quantitative analysis. It also generated rather good results for data sets that concern calibration transfer. It demonstrates that constructing network architectures by properly revising existing successful network architectures may provide additional choices in the exploration of the application of deep neural network in analytical chemistry.https://www.mdpi.com/2076-3417/13/14/8494deep neural networksrevised U-Net DNNrevised attention mechanism DNNquantitative analysiscalibration transfer |
spellingShingle | Hong-Hua Huang Jian-Fei Luo Feng Gan Philip K. Hopke Two Revised Deep Neural Networks and Their Applications in Quantitative Analysis Based on Near-Infrared Spectroscopy Applied Sciences deep neural networks revised U-Net DNN revised attention mechanism DNN quantitative analysis calibration transfer |
title | Two Revised Deep Neural Networks and Their Applications in Quantitative Analysis Based on Near-Infrared Spectroscopy |
title_full | Two Revised Deep Neural Networks and Their Applications in Quantitative Analysis Based on Near-Infrared Spectroscopy |
title_fullStr | Two Revised Deep Neural Networks and Their Applications in Quantitative Analysis Based on Near-Infrared Spectroscopy |
title_full_unstemmed | Two Revised Deep Neural Networks and Their Applications in Quantitative Analysis Based on Near-Infrared Spectroscopy |
title_short | Two Revised Deep Neural Networks and Their Applications in Quantitative Analysis Based on Near-Infrared Spectroscopy |
title_sort | two revised deep neural networks and their applications in quantitative analysis based on near infrared spectroscopy |
topic | deep neural networks revised U-Net DNN revised attention mechanism DNN quantitative analysis calibration transfer |
url | https://www.mdpi.com/2076-3417/13/14/8494 |
work_keys_str_mv | AT honghuahuang tworeviseddeepneuralnetworksandtheirapplicationsinquantitativeanalysisbasedonnearinfraredspectroscopy AT jianfeiluo tworeviseddeepneuralnetworksandtheirapplicationsinquantitativeanalysisbasedonnearinfraredspectroscopy AT fenggan tworeviseddeepneuralnetworksandtheirapplicationsinquantitativeanalysisbasedonnearinfraredspectroscopy AT philipkhopke tworeviseddeepneuralnetworksandtheirapplicationsinquantitativeanalysisbasedonnearinfraredspectroscopy |