PREDICTION OF CHRONIC KIDNEY DISEASE BASED ON SCINTIGRAPHIC IMAGES USING MACHINE LEARNING METHODS
DOI:
https://doi.org/10.24867/29BE25VajdaKeywords:
Machine learning, scintigraphy, chronic kidney diseaseAbstract
This paper investigates the application of machine learning data for predicting the stage of chronic kidney disease based on examinee data and images obtained by scintigraphy. Clinical and demographic data, as well as data obtained from images, were analyzed, and the following models were applied: k-nearest neighbors, decision tree, random forest and gradient boosting. The results showed that the k-nearest neighbors method had the lowest accuracy of 24%, while the decision tree and random forest methods showed significantly better performance with an accuracy of 96%. The gradient boosting model performed the best, achieving an accuracy of 100%.
References
[1] T. Nosek, B. Brkljač, D. Despotović, M. Sečujski, T. Lončar-Turukalo, Praktikum iz mašinskog učenja, 1st ed., Novi Sad: Univerzitet u Novom Sadu, 2020, pp. 122.
[2] P. Cunningham, S. J. Delany., ”k-Nearest neighbour classifiers.”, ACM Computing Surveys, vol. 54(6), pp. 1-25, April, 2007.
[3] M. Nikolić, A. Zečević, Маšinsko učenje, Beograd: Matematički fakultet Univerziteta u Beogradu, 2019.
[4] https://dynamicrenalstudy.org/ (pristupljeno u maju 2024.)
[5] M. Sezgin, B. Sankur., "Survey over image thresholding techniques and quantitative performance evaluation." Journal of Electronic Imaging, vol. 13(1), pp. 146-165, 2004.
[6] N. Otsu., (1979). "A Threshold Selection Method from Gray-Level Histograms", IEEE Transactions on Systems, Man, and Cybernetics, vol. 9(1), pp. 62-66,1979.
[7] D. Vidaković, Tehnologije i alati u mašinskom učenju, Novi Sad: Fakultet tehničkih nauka Univerziteta u Novom Sadu, 2023.
[8] https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html (pristupljeno u junu 2024.)
[2] P. Cunningham, S. J. Delany., ”k-Nearest neighbour classifiers.”, ACM Computing Surveys, vol. 54(6), pp. 1-25, April, 2007.
[3] M. Nikolić, A. Zečević, Маšinsko učenje, Beograd: Matematički fakultet Univerziteta u Beogradu, 2019.
[4] https://dynamicrenalstudy.org/ (pristupljeno u maju 2024.)
[5] M. Sezgin, B. Sankur., "Survey over image thresholding techniques and quantitative performance evaluation." Journal of Electronic Imaging, vol. 13(1), pp. 146-165, 2004.
[6] N. Otsu., (1979). "A Threshold Selection Method from Gray-Level Histograms", IEEE Transactions on Systems, Man, and Cybernetics, vol. 9(1), pp. 62-66,1979.
[7] D. Vidaković, Tehnologije i alati u mašinskom učenju, Novi Sad: Fakultet tehničkih nauka Univerziteta u Novom Sadu, 2023.
[8] https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html (pristupljeno u junu 2024.)
Downloads
Published
2024-11-05
Issue
Section
Electrotechnical and Computer Engineering
