The Journal of the Korea institute of electronic communication sciences (한국전자통신학회논문지)

Korea Institute of Electronic Communication Science (한국전자통신학회)
권호 표지
DOI QR코드

DOI QR Code

Research on a Non-invasive Blood Glucose level Estimation Algorithm based on Near- infrared Spectroscopy

근적외선 분광법 기반 비침습식 혈당 수치 추정 알고리즘 연구

  • 강영만 (전남대학교 문화콘텐츠학부) ;
  • 한순희 (전남대학교 문화콘텐츠학부)
  • Received : 2023.10.11
  • Accepted : 2023.12.27
  • Published : 2023.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Various methods are being attempted to resolve the inconvenience of blood glucose meters used to check blood sugar levels. In this paper, we attempted to estimate blood sugar levels non-invasively using machine learning technology from spectral data acquired using a near-infrared sensor. The non-invasive blood glucose meter used in the study has a total of six near-infrared ray emitters, including visible rays, and a light receiver that receives them. It is a device created to collect spectral data on specific parts of the human body, such as the fingers. To verify whether there was a significant difference depending on blood sugar level, we attempted to estimate blood sugar level through machine learning algorithms. As a result of applying five machine learning algorithm techniques to the collected data and adjusting various hyper parameters, it was confirmed that the support vector regression algorithm showed the best performance.

혈당 수치를 확인하기 위해 사용하는 채혈식 혈당 측정기의 불편함을 해소하기 위한 다양한 방법이 시도되고 있다. 본 논문에서는 근적외선 센서를 이용하여 획득한 스펙트럼 데이터로부터 머신러닝 기술을 사용하여 비침습적으로 혈당 수치를 추정하고자 하였다. 연구에 사용한 비침습식 혈당측정기는 가시광선을 포함하여 총 6개로 구성된 근적외선 발광부와 이를 수신하는 수광부를 가지는데, 손가락과 같은 인체의 특정 부위에 대한 스펙트럼 데이터를 수집하기 위해 만든 기기이다. 혈당 수치에 따라 유의미한 차이가 있는지를 검증하기 위해 머신 러닝 알고리즘들을 통해 혈당 수치 추정을 시도하였다. 수집한 데이터에 5가지 머신 러닝 알고리즘 기법을 적용하면서 다양한 하이퍼 파라미터를 조정한 결과, 서포트 벡터 회귀 알고리즘이 가장 좋은 성능을 나타냄을 확인하였다.

Keywords

References

  1. L. Tang, S.-J. Chang, C.-J. Chen and J.-T. Liu, "Non-lnvasive Blood Glucose Monitoring Technology: A Review," Sensors, vol. 20, 2020, pp. 6925.
  2. A. Hina and W. Saadeh, "Noninvasive Blood Glucose Monitoring Systems Using Near-Infrared Technology-A Review," Sensors, vol. 22, 2022, pp. 4855.
  3. M. Syafrudin, G. Alfian, N. L. Fitriyani, I. Fahrurrozi, M. Anshari, and J. Rhee, "A Personalized Blood Glucose Prediction Model Using Random Forest Regression," In Proc. 2022 ASU Int. Conf. in Emerging Technologies for Sustainability and Intelligent Systems, Manama, Bahrain, June 2022, pp. 295-299.
  4. M.-M. Enric, "Non-invasive estimate of blood glucose and blood pressure from a photoplethysmograph by means of machine learning techniques," Artificial Intelligence in Medicine, vol. 53, no. 2, Oct. 2011, pp. 127-38. https://doi.org/10.1016/j.artmed.2011.05.001
  5. J. Baek, N. Kang, Y.Woo, and H. Kim, "Compensation of Variation from Long-Term Spectral Measurement for Non-invasive Blood Glucose in Mouse by Near-Infrared Spectroscopy," J. of the Pharmaceutical Society of Korea, vol. 48. no. 3, pp. 177-181.
  6. Y. Kang and S. Han, "Non-invasive Blood Glucose Detection Sensor System Based on Near-Infrared Spectroscopy," J. of the Korea Institute of Electronic Communication Sciences, vol. 16, no. 5, 2021, pp. 991-1000.
  7. R. Asmund, F. Berg, and S. Engelsen, "Review of the most common pre-processing techniques for near-infrared spectra," TrAC Trends in Analytical Chemistry, vol. 28, no. 10, 2009, pp. 1201-1222. https://doi.org/10.1016/j.trac.2009.07.007
  8. F. Jansson, "Comparing machine learning algorithms for detecting behavioural anomalies," Master's Thesis, Blekinge Institute of Technology, Faculty of Computing, 2023.
  9. U. Khaire and R. Dhanalakshmi, "Stability of feature selection algorithm: A review," J. of King Saud University-Computer and Information Sciences, vol. 34, no. 4, 2022, pp. 1060-1073. https://doi.org/10.1016/j.jksuci.2019.06.012
  10. K. Kim, S. Han, T. Kim, S. Jung, S. Kim, H. Ceong, and J. Park, "Pattern Analysis of Clinical Signs in Cultured Olive Flounder, Paralichthys Olivaceus, with Edwardsielosis using the Decision Tree Technique," J. of the Korea Institute of Electronic Communication Sciences, vol. 16, no. 4, 2021, pp. 661-674.
  11. A. E. Hoerl and R. W. Kennard. "Ridge Regression: Biased Estimation for Nonorthogonal Problems," Technometrics, vol. 42, no. 1, 2000, pp. 80-86. https://doi.org/10.1080/00401706.2000.10485983
  12. R. Tibshirani, "Regression Shrinkage and Selection via the Lasso," J. of the Royal Statistical Society. Series B (Methodological), vol. 58, no. 1, 1996, pp. 267-288. https://doi.org/10.1111/j.2517-6161.1996.tb02080.x
  13. L. Breiman, "Random Forests," Machine Learning, Springer Nature, vol. 45, Oct. 2001, pp. 5-32. https://doi.org/10.1023/A:1010933404324
  14. A. J. Smola and B. Scholkopf, "A tutorial on support vector regression," Statistics and Computing, vol. 14, 2004, pp. 199-222. https://doi.org/10.1023/B:STCO.0000035301.49549.88
  15. H. Kurt, S. Maxwell and W. Halbert, "Multilayer feedforward networks are universal approximators," Neural Networks, vol. 2, no. 5, 1989, pp. 359-366. https://doi.org/10.1016/0893-6080(89)90020-8
  16. G. Geladi and B. R. Kowalski, "Partial least-squares regression: a tutorial," Analytica Chimica Acta, vol. 185, 1986, pp. 1-17. https://doi.org/10.1016/0003-2670(86)80028-9
  17. J. Bergstra and Y. Bengio, "Random search for hyper-parameter optimization," J. of Machine Learning Research. vol. 13, 2012, pp. 281-305.
  18. D. Chicco, M.-J. Warrens, and G. Jurman, "The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation," PeerJ Computer Science. vol. 7, 2021, pp. e623.
  19. ISO/TC 212/SC, Int. Std. ISO 15197, In vitro diagnostic test systems, Geneva, Switzerland: ISO, 2003.
  20. G. Freckmann, J. Mende, S. Pleus, D. Waldenmaier, A. Baumstark, N. Jendrike, and C. Haug, "Mean Absolute Relative Difference of Blood Glucose Monitoring Systems and Relationship to ISO 15197," J. of diabetes science and technology, vol. 16, no. 5, 2022, pp. 1089-1095. https://doi.org/10.1177/19322968211001402
  21. D. Nimmagadda and B. Potluri, "Assessment of Parkes Error Grid through Machine learning techniques," Master's Thesis, Blekinge Institute of Technology, Faculty of Computing, 2020.