Predicting Hyperglycemia Using NIR Spectrum of Spent Fluid in Hemodialysis Patients
Authorized Users Only
2021
Authors
Matović, ValentinaTrbojević-Stanković, Jasna
Matija, Lidija
Šarac, Dušan
Vasić-Milovanović, Aleksandra
Petrović, A.
Article (Published version)
Metadata
Show full item recordAbstract
We aimed to assess the near infrared spectroscopy as a method for non-invasive on-line detection of hyperglycemia from spent hemodialysis effluent. We used partial least squares regression and several machine learning algorithms: random forest (RF), logistic regression, K-nearest neighbor (KNN), support vector machine (SVM), decision tree classifier, and Gaussian naive Bayes (NB) to classify normoglycemia from hyperglycemia. These classifier methods were used on the same dataset and evaluated by the area under the curve. The serum glucose levels were presented in the form of a binomial variable, where 0 indicated a glucose level within reference range and 1 a glucose level beyond the normal limit. For this reason, the methods of machine learning were applied as more specific methods of classification. RF and SVM have shown the best classification accuracy in predicting hyperglycemia, while decision tree and NB showed average accuracy.
Keywords:
spent dialysate / near infrared spectroscopy / machine learning / hemodialysisSource:
Journal of Applied Spectroscopy, 2021, 88, 3, 662-667Publisher:
- Springer, New York
Funding / projects:
- Functionalization of Nanomaterials for obtaining new contact lenses, and early diagnostics of diabetes (RS-MESTD-Integrated and Interdisciplinary Research (IIR or III)-45009)
- Development of methods and techniques for early diagnostic of cervical, colon, oral cavity cancer and melanoma based on a digital image and excitation-emission spectrum in visible and infrared domain (RS-MESTD-Integrated and Interdisciplinary Research (IIR or III)-41006)
DOI: 10.1007/s10812-021-01222-3
ISSN: 0021-9037
WoS: 000673203200009
Scopus: 2-s2.0-85110439390
Collections
Institution/Community
Mašinski fakultetTY - JOUR AU - Matović, Valentina AU - Trbojević-Stanković, Jasna AU - Matija, Lidija AU - Šarac, Dušan AU - Vasić-Milovanović, Aleksandra AU - Petrović, A. PY - 2021 UR - https://machinery.mas.bg.ac.rs/handle/123456789/3521 AB - We aimed to assess the near infrared spectroscopy as a method for non-invasive on-line detection of hyperglycemia from spent hemodialysis effluent. We used partial least squares regression and several machine learning algorithms: random forest (RF), logistic regression, K-nearest neighbor (KNN), support vector machine (SVM), decision tree classifier, and Gaussian naive Bayes (NB) to classify normoglycemia from hyperglycemia. These classifier methods were used on the same dataset and evaluated by the area under the curve. The serum glucose levels were presented in the form of a binomial variable, where 0 indicated a glucose level within reference range and 1 a glucose level beyond the normal limit. For this reason, the methods of machine learning were applied as more specific methods of classification. RF and SVM have shown the best classification accuracy in predicting hyperglycemia, while decision tree and NB showed average accuracy. PB - Springer, New York T2 - Journal of Applied Spectroscopy T1 - Predicting Hyperglycemia Using NIR Spectrum of Spent Fluid in Hemodialysis Patients EP - 667 IS - 3 SP - 662 VL - 88 DO - 10.1007/s10812-021-01222-3 ER -
@article{ author = "Matović, Valentina and Trbojević-Stanković, Jasna and Matija, Lidija and Šarac, Dušan and Vasić-Milovanović, Aleksandra and Petrović, A.", year = "2021", abstract = "We aimed to assess the near infrared spectroscopy as a method for non-invasive on-line detection of hyperglycemia from spent hemodialysis effluent. We used partial least squares regression and several machine learning algorithms: random forest (RF), logistic regression, K-nearest neighbor (KNN), support vector machine (SVM), decision tree classifier, and Gaussian naive Bayes (NB) to classify normoglycemia from hyperglycemia. These classifier methods were used on the same dataset and evaluated by the area under the curve. The serum glucose levels were presented in the form of a binomial variable, where 0 indicated a glucose level within reference range and 1 a glucose level beyond the normal limit. For this reason, the methods of machine learning were applied as more specific methods of classification. RF and SVM have shown the best classification accuracy in predicting hyperglycemia, while decision tree and NB showed average accuracy.", publisher = "Springer, New York", journal = "Journal of Applied Spectroscopy", title = "Predicting Hyperglycemia Using NIR Spectrum of Spent Fluid in Hemodialysis Patients", pages = "667-662", number = "3", volume = "88", doi = "10.1007/s10812-021-01222-3" }
Matović, V., Trbojević-Stanković, J., Matija, L., Šarac, D., Vasić-Milovanović, A.,& Petrović, A.. (2021). Predicting Hyperglycemia Using NIR Spectrum of Spent Fluid in Hemodialysis Patients. in Journal of Applied Spectroscopy Springer, New York., 88(3), 662-667. https://doi.org/10.1007/s10812-021-01222-3
Matović V, Trbojević-Stanković J, Matija L, Šarac D, Vasić-Milovanović A, Petrović A. Predicting Hyperglycemia Using NIR Spectrum of Spent Fluid in Hemodialysis Patients. in Journal of Applied Spectroscopy. 2021;88(3):662-667. doi:10.1007/s10812-021-01222-3 .
Matović, Valentina, Trbojević-Stanković, Jasna, Matija, Lidija, Šarac, Dušan, Vasić-Milovanović, Aleksandra, Petrović, A., "Predicting Hyperglycemia Using NIR Spectrum of Spent Fluid in Hemodialysis Patients" in Journal of Applied Spectroscopy, 88, no. 3 (2021):662-667, https://doi.org/10.1007/s10812-021-01222-3 . .