基于GA-SVR的近红外无创检测智能算法研究

doi:10.3964/j.issn.1000-0593(2024)11-3020-09

摘要
参考文献
相关文章 (15)

全文: PDF (6642 KB)
输出: BibTeX | EndNote (RIS)

摘要：近年来，由于安全、无损、高效等因素，基于近红外光谱及人工智能算法的无创检测在医学、生物等领域内备受关注。如何获取智能回归模型的近红外光谱有效特征是关键问题之一。以血糖浓度检测为例，融合近红外光谱、遗传算法与支持向量回归(GA-SVR)，建立了近红外无创血糖浓度智能预测算法。首先，根据OGTT实验规则，采集志愿者无创动态血液近红外光谱及其对应的血糖浓度，进一步基于遗传算法确定最优近红外特征波长组合，最后建立支持向量机回归模型实现血糖浓度的回归预测。设计了对比实验，分别将遗传算法与多层感知机回归(GA-MLPR)、偏最小二乘回归(GA-PLSR)和随机森林回归(GA-RFR)结合，与本文提出的方法进行比较。实验结果表明，提出的GA-SVR模型预测效果最好，测试集相关系数相比GA-PLSR提高了44%，相关系数达到99.97%，均方误差为0.000 097。表明，提出的GA-SVR可以实现对近红外光谱数据的有效特征提取，验证了启发式智能算法对于近红外无创检测的可行性。

关键词：近红外光谱；无创检测；特征选择；遗传算法；支持向量机

Abstract：In recent years, non-invasive detection based on near-infrared spectroscopy and artificial intelligence algorithms has received much attention in medicine and biology due to its safety, non-invasiveness, and high efficiency. One key issue is selecting effective input features for intelligent regression models from wide-band near-infrared spectroscopy. This paper establishes a non-invasive near-infrared blood glucose concentration intelligent prediction model by combining near-infrared spectroscopy, genetic algorithm, and support vector regression (GA-SVR) using blood glucose concentration detection as an example. Firstly, according to the OGTT experimental rules, non-invasive dynamic blood near-infrared spectroscopy and corresponding blood glucose concentrations of volunteers were collected. The optimal near-infrared feature wavelength combination was further determined based on a genetic algorithm. Finally, the support vector machine regression model was established to achieve blood glucose concentration prediction. In this paper, comparative experiments were designed to compare the proposed method with the genetic algorithm and multi-layer perceptron regression (GA-MLPR), partial least squares regression (GA-PLSR), and random forest regression (GA-RFR). The experimental results show that the proposed GA-SVR model has the best prediction performance, and the correlation coefficient of the test set is increased by 44% compared with GA-PLSR, the correlation coefficient reaches 99.97%, and the mean square error is 0.000 97. The study shows that the proposed GA-SVR can achieve effective feature selection of near-infrared spectroscopy data, verifying the feasibility of intelligent algorithms for feature selection. The excellent performance of this feature selection model provides a new approach to non-invasive detection.

Key words：Near-infrared spectroscopy; Non-invasive detection; Feature Selection; Genetic algorithm; Support vector machine

收稿日期: 2023-08-17 修订日期: 2024-03-12

中图分类号:

O433

基金资助: 吉林省自然科学基金自由探索重点项目(YDZJ202201ZYTS585)，吉林省创新能力建设项目(2022C047-2)，国家自然科学基金项目(11401092)资助

通讯作者: 周林华 E-mail: zhoulh@cust.edu.cn

作者简介: 于欣冉, 女，1999年生，长春理工大学数学与统计学院硕士研究生 e-mail: 1339688841@qq.com
赵鹏，1981年生，长春理工大学物理学院讲师 e-mail: zhaop@cust.edu.cn
于欣冉，赵鹏：并列第一作者

引用本文:

于欣冉，赵鹏，宦克为，李野，姜志侠，周林华. 基于GA-SVR的近红外无创检测智能算法研究[J]. 光谱学与光谱分析, 2024, 44(11): 3020-3028.
YU Xin-ran, ZHAO Peng, HUAN Ke-wei, LI Ye, JIANG Zhi-xia, ZHOU Lin-hua. Research on Intelligent Algorithm of Near-Infrared Spectroscopy Non-Invasive Detection Based on GA-SVR Method. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3020-3028.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2024)11-3020-09 或 https://www.gpxygpfx.com/CN/Y2024/V44/I11/3020

[1] Awelisah Y M, Li G, Lin L. Infrared Physics & Technology, 2022, 121: 104049.
[2] Peng X, Yan Y X, Liu H. Optical Fiber Technology, 2022, 68: 102822.
[3] Heise H M, Delbeck S, Marbach R. Biosensors, 2021, 11(3): 64.
[4] Sun J, Pang R, Chen S, et al. Journal of Innovative Optical Health Sciences, 2021, 14(06): 2130006.
[5] Devezas M A M. Journal of Neuroimaging, 2021, 31(4): 641.
[6] Dahne C, Gross D. US Patent: 4655225, 1987.
[7] Morales G, Sheppard J W, Logan R D, et al. Remote Sensing. 2021; 13(18): 3649.
[8] YAN Hong-mei, HE Ming-yi(闫红梅，何明一). Journal of Signal Processing(信号处理), 2023, 39(1): 1.
[9] ZHANG Jin, HU Yun, ZHOU Luo-xiong, et al(张进, 胡芸, 周罗雄，等). Journal of Instrumental Analysis(分析测试学报), 2020, 39(10): 1196.
[10] Kasim S, Malek S, Ibrahim K S, et al. European Heart Journal, 2021, 42(Supplement_1): 3069.
[11] Qian W, Xiong Y, Yang J, et al. Information Sciences, 2022, 582: 38.
[12] Mostafa R R, Ewees A A, Ghoniem R M, et al. Knowledge-Based Systems, 2022, 246: 108743.
[13] Shao R, Zhang G, Gong X. Photonics Research, 2022, 10(8): 1868.
[14] El-Nemr M, Afifi M, Rezk H, et al. Mathematics, 2021, 9(5): 576.
[15] Wei Y, Chen Z, Zhao C, et al. Ocean Engineering, 2023, 270: 113659.
[16] Leardi R. Journal of Chemometrics, 2000, 14: 643.
[17] LI Hao-guang, YU Yun-hua, PANG Yan, et al(李浩光, 于云华, 逄燕，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2021, 41(8): 2437.
[18] An Y, Xue H. Pattern Recognition, 2022, 121: 108195.
[19] MA Shuang, PU Bao-ming(马爽, 蒲宝明). Computer Systems & Applications(计算机系统应用), 2016，25(8): 120.
[20] Oyehan T A, Alade I O, Bagudu A, et al. Computers in Biology and Medicine, 2018, 98: 85.
[21] LI Meng-ze, JI Zhong, CHENG Jin-xiu(李孟泽, 季忠, 程锦绣). Journal of Biomedical Engineering(生物医学工程学杂志), 2021，38 (2): 342.
[22] Yadav J, Rani A, Singh V, et al. Near-Infrared LED Based Non-Invasive Blood Glucose Sensor, 2014 International Conference on Signal Processing and Integrated Networks (SPIN). IEEE, 2014: 591.