Comparative Analysis of Classification Algorithms in the Analysis of Medical Images From Speckle Tracking Echocardiography Video Data

Authors

DOI:

https://doi.org/10.20535/ibb.2021.5.3.234990

Keywords:

classification algorithms, medical image analysis, speckle-tracking echocardiography, heart condition, coronary heart disease, harmonic structures, м, group method of data handling, random forest, adaptive boosting

Abstract

Background. Machine learning allows applying various intelligent algorithms to produce diagnostic and/or prognostic models. Such models can be used to determine the functional state of the heart, which is diagnosed by speckle-tracking echocardiography. To determine the patient's heart condition in detail, a classification approach is used in machine learning. Each of the classification algorithms has a different performance when applied to certain situations. Therefore, the actual task is to determine the most efficient algorithm for solving a specific task of classifying the patient's heart condition when applying the same speckle-tracking echocardiography data set.

Objective. We are aimed to evaluate the effectiveness of the application of prognostic models of logistic regression, the group method of data handling (GMDH), random forest, and adaptive boosting (AdaBoost) in the construction of algorithms to support medical decision-making on the diagnosis of coronary heart disease.

Methods. Video data from speckle-tracking echocardiography of 40 patients with coronary heart disease and 16 patients without cardiac pathology were used for the study. Echocardiography was recorded in B-mode in three positions: long axis, 4-chamber, and 2-chamber. Echocardiography frames that reflect the systole and diastole of the heart (308 samples in total) were taken as objects for classification. To obtain informative features of the selected objects, the genetic GMDH approach was applied to identify the best structure of harmonic textural features. We compared the efficiency of the following classification algorithms: logistic regression method, GMDH classifier, random forest method, and AdaBoost method.

Results. Four classification models were constructed for each of the three B-mode echocardiography positions. For this purpose, the data samples were divided into 3: training sample (60%), validation sample (20%), and test sample (20%). Objective evaluation of the models on the test sample showed that the best classification method was random forest (90.3% accuracy on the 4-chamber echocardiography position, 74.2% on the 2-chamber, and 77.4% on the long axis). This was also confirmed by ROC analysis, wherein in all cases, the random forest was the most effective in classifying cardiac conditions.

Conclusions. The best classification algorithm for cardiac diagnostics by speckle-tracking echocardiography was determined. It turned out to be a random forest, which can be explained by the ensemble approach of begging, which is inherent in this classification method. It will be the mainstay of further research, which is planned to be performed to develop a full-fledged decision support system for cardiac diagnostics.

References

Voigt JU, Pedrizzetti G, Lysyansky P, Marwick TH, Houle H, Baumann R, et al. Definitions for a common standard for 2D speckle tracking echocardiography: consensus document of the EACVI/ASE/Industry Task Force to standardize deformation imaging. Eur Heart J Cardiovasc Imaging. 2015;16(1):1-11. DOI: 10.1093/ehjci/jeu184

Lazoryshynets VV, Kovalenko VM, Rudenko AV, Ivaniv YA, Beshlyaga VM, Potashev SV, et al. Definition for a common standard for 2D speckle-tracking echocardiography (The Association of Cardiovascular Surgeons of Ukraine and the Ukrainian Society of Cardiology Working Group Draft Consensus). Cardiol Cardiac Surg Contin Profes Develop. 2019;2(2):105-29. DOI: 10.30702/ccs.201905.02.2dst105129

Nastenko Ie, Maksymenko V, Potashev S, Pavlov V, Babenko V, Rysin S, et al. Random forest algorithm construction for the diagnosis of coronary heart disease based on echocardiography video data streams. Innov Biosyst Bioen. 2021;5(1):61-9. DOI: 10.20535/ibb.2021.5.1.225794

Nastenko Ie, Maksymenko V, Potashev S, Pavlov V, Babenko V, Rysin S, et al. Group method of data handling application in constructing of coronary heart disease diagnosing algorithms. Biomedichna Injeneriya Technologiya. 2021;5:1-9. DOI: 10.20535/2617-8974.2021.5.227141

Nastenko Ie, Maksymenko V, Galkin A, Pavlov V, Nosovets O, Dykan I, et al. Liver pathological states identification with self-organization models based on ultrasound images texture features. In: Shakhovska N, Medykovskyy MO, editors. Advances in intel¬ligent systems and computing V. Cham: Springer International Publishing; 2021. p. 401-18. DOI: 10.1007/978-3-030-63270-0_26

Nastenko Ie, Pavlov V, Nosovets O, Kruhlyi V, Honcharuk M, Karliuk A, et al. Texture analysis application in medical images classification task solving. Biomedichna Injeneriya Technologiya. 2020;4:69-82. DOI: 10.20535/2617-8974.2020.4.221876

Kumari R, Srivastava KS. Machine learning: a review on binary classification. Int J Comp Appl. 2017;160(7):11-5. DOI: 10.5120/ijca2017913083

Legua MP, Morales I, Sánchez Ruiz LM. The Heaviside step function and MATLAB. In: Proceedings of Computational Science and Its Applications – ICCSA 2008. Berlin, Heidelberg: Springer Berlin Heidelberg; 2008. p. 1212-21. Available from: DOI: 10.1007/978-3-540-69839-5_93

Venetis J. An analytic exact form of heaviside step function. Adv Appl Discrete Math. 2019 Nov 15;22(2):153-9. DOI: 10.17654/DM022020153

Ramola A, Shakya AK, Van Pham D. Study of statistical methods for texture analysis and their modern evolutions. Eng Rep. 2020;2(4):1-24. DOI: 10.1002/eng2.12149

Osapoetra LO, Chan W, Tran W, Kolios MC, Czarnota GJ. Comparison of methods for texture analysis of QUS parametric images in the characterization of breast lesions. PLoS One. 2020;15(12):e0244965. DOI: 10.1371/journal.pone.0244965

Ramola A, Shakya AK, Vidyarthi A. Applications and approaches for texture analysis and their modern evolution. In: Lecture Notes in Electrical Engineering. Singapore: Springer; 2020. p. 273-81. DOI: 10.1007/978-981-15-4932-8_30

Benco M, Hudec R, Kamencay P, Zachariasova M, Matuska S. An advanced approach to extraction of colour texture features based on GLCM. Int J Adv Robotic Syst. 2014;11(7):104. DOI: 10.5772/58692

Sohail ASM, Bhattacharya P, Mudur SP, Krishnamurthy S. Local relative GLRLM-based texture feature extraction for classifying ultrasound medical images. In: Proceedings of 2011 24th Canadian Conference on Electrical and Computer Engineering. IEEE; 2011. p. 001092-5. DOI: 10.1109/CCECE.2011.6030630

Babenko V. Ultrasound images classification by the genetic forest of optimal complexity trees [master’s thesis]. Kyiv: Igor Sikorsky Kyiv Polytechnic Institute; 2021.

Jiřina M, Jiřina M jr. GMDH Method with Genetic Selection Algorithm and Cloning. Neural Network World. 2013;23(5):451-64. DOI: 10.14311/NNW.2013.23.028

Madala HR, Ivakhnenko AG. Inductive learning algorithms for complex systems modeling. Inductive Learning Algorithms for Complex Systems Modeling. CRC Press; 2019. 380 p. DOI: 10.1201/9781351073493-2

Katoch S, Chauhan SS, Kumar V. A review on genetic algorithm: past, present, and future. Multimedia Tools Applications. 2021;80(5):8091-126. DOI: 10.1007/s11042-020-10139-6

Glumov NI, Kolomiyetz EI, Sergeyev VV. Detection of objects on the image using a sliding window mode. Optics Laser Technol. 1995;27(4):241-9. DOI: 10.1016/0030-3992(95)93752-D

Lian R, Huang L. DeepWindow: Sliding window based on deep learning for road extraction from remote sensing images. IEEE J Select Topics Appl Earth Observ Remote Sens. 2020;13:1905-16. DOI: 10.1109/JSTARS.2020.2983788

Klymenko D, Nastenko Ie, Pavlov V. USI images classification by surface modeling method with genetic GMDH. In: Proceedings of XXII International Scientific and Practical Conference Theoretical Foundations for the Implementation and Adaptation of Scientific Achievements in Practice. Helsinki: International Science Group; 2020. p. 188-93. Available from: https://isg-konf.com/wp-content/uploads/2020/06/XXII-Conference-22-23-Helsinki-Finland-book.pdf

Luque A, Carrasco A, Martín A, Lama JR. Exploring symmetry of binary classification performance metrics. Symmetry (Basel). 2019;11(1):47. DOI: 10.3390/sym11010047

Dag O, Karabulut E, Alpar R. GMDH2: Binary classification via GMDH-type neural network algorithms—R package and web-based tool. Int J Comput Intell Syst. 2019;12(2):649. DOI: 10.2991/ijcis.d.190618.001

Xie L, Jia Y, Xiao J, Gu X, Huang J. GMDH-based outlier detection model in classification problems. J Syst Sci Complex. 2020 ;33(5):1516-32. DOI: 10.1007/s11424-020-9002-6

Sperandei S. Understanding logistic regression analysis. Biochem Medica. 2014;24(1):12-8. DOI: 10.11613/BM.2014.003

Liu H, Li T, Chen L, Zhan S, Pan M, Ma Z, et al. To set up a logistic regression prediction model for hepatotoxicity of Chinese herbal medicines based on traditional chinese medicine theory. Evid Based Complement Alternat Med. 2016;2016:1-9. DOI: 10.1155/2016/7273940

Poullis M, Pullan M, Chalmers J, Mediratta N. The validity of the original EuroSCORE and EuroSCORE II in patients over the age of seventy. Interact Cardiovasc Thorac Surg. 2015;20(2):172-7. DOI: 10.1093/icvts/ivu345

Fernández-Hidalgo N, Ferreria-González I, Marsal JR, Ribera A, Aznar ML, de Alarcón A, et al. A pragmatic approach for mortality prediction after surgery in infective endocarditis: optimizing and refining EuroSCORE. Clin Microbiol Infec. 2018;24(10):1102.e7-e15. DOI: 10.1016/j.cmi.2018.01.019

Nashef SAM, Roques F, Sharples LD, Nilsson J, Smith C, Goldstone AR, et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41(4):734-44. DOI: 10.1093/ejcts/ezs043

Wang R. Symptoms selection using random forest based on Chinese medicine diagnostic cases of stomachache. J Phys Conf Ser. 2019;1168(6):062025. DOI: 10.1088/1742-6596/1168/6/062025

Cafri G, Li L, Paxton EW, Fan J. Predicting risk for adverse health events using random forest. J Appl Stat. 2018;45(12):2279-94. DOI: 10.1080/02664763.2017.1414166

Thongkam J, Xu G, Zhang Y, Huang F. Breast cancer survivability via AdaBoost algorithms. In: Proceedings of the second Australasian workshop on Health data and knowledge management. Wollongong: Australian Computer Society; 2008. p. 55-64. Available from: http://dl.acm.org/citation.cfm?id=1385098

Zhong L, Wang JTL, Wen D, Aris V, Soteropoulos P, Shapiro BA. Effective classification of MicroRNA precursors using feature mining and AdaBoost algorithms. OMICS. 2013;17(9):486-93. DOI: 10.1089/omi.2013.0011

Hoo ZH, Candlish J, Teare D. What is an ROC curve? Emerg Med J. 2017;34(6):357-9. DOI: 10.1136/emermed-2017-206735

Published

2021-09-10

How to Cite

1.
Petrunina O, Shevaga D, Babenko V, Pavlov V, Rysin S, Nastenko I. Comparative Analysis of Classification Algorithms in the Analysis of Medical Images From Speckle Tracking Echocardiography Video Data. Innov Biosyst Bioeng [Internet]. 2021Sep.10 [cited 2024Dec.13];5(3):153-66. Available from: https://ibb.kpi.ua/article/view/234990

Issue

Section

Articles