Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Nov 16:2021:9979560.
doi: 10.1155/2021/9979560. eCollection 2021.

Keratoconus Severity Classification Using Features Selection and Machine Learning Algorithms

Mustapha Aatila  1 , Mohamed Lachgar  1 , Hrimech Hamid  2 , Ali Kartit  1
Affiliations

Keratoconus Severity Classification Using Features Selection and Machine Learning Algorithms

Mustapha Aatila et al. Comput Math Methods Med. .

Abstract

Keratoconus is a noninflammatory disease characterized by thinning and bulging of the cornea, generally appearing during adolescence and slowly progressing, causing vision impairment. However, the detection of keratoconus remains difficult in the early stages of the disease because the patient does not feel any pain. Therefore, the development of a method for detecting this disease based on machine and deep learning methods is necessary for early detection in order to provide the appropriate treatment as early as possible to patients. Thus, the objective of this work is to determine the most relevant parameters with respect to the different classifiers used for keratoconus classification based on the keratoconus dataset of Harvard Dataverse. A total of 446 parameters are analyzed out of 3162 observations by 11 different feature selection algorithms. Obtained results showed that sequential forward selection (SFS) method provided a subset of 10 most relevant variables, thus, generating the highest classification performance by the application of random forest (RF) classifier, with an accuracy of 98% and 95% considering 2 and 4 keratoconus classes, respectively. Found classification accuracy applying RF classifier on the selected variables using SFS method achieves the accuracy obtained using all features of the original dataset.

PubMed Disclaimer

Conflict of interest statement

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Figures

Figure 1
Figure 1
Normal eye (a) and keratoconus eye (b) [6].
Figure 2
Figure 2
Feature selection model.
Figure 3
Figure 3
Keratoconus classification model.
Figure 4
Figure 4
Classification accuracy of different models using all features (a) and applying mutual information (b), ANOVA (c), embedded (d), embedded with a filter (e), and filter with RFE (f) feature selection algorithms.
Figure 5
Figure 5
Classification accuracy of different models using RFE (a), filter with HRFA (b), filter with SFS (c), SFS (d), genetic (e), and filter with SBS (f) feature selection algorithms.
Figure 6
Figure 6
Classification accuracy of different models using SBS feature selection algorithms.
Figure 7
Figure 7
Comparison of classification performance based on the accuracy of different models using all features (a) and applying mutual information (b), ANOVA (c), embedded (d), embedded with a filter (e), and filter with RFE (f) feature selection algorithms.
Figure 8
Figure 8
Comparison of classification performance based on the accuracy of different models using RFE (a), filter with HRFA (b), filter with SFS (c), SFS (d), genetic (e), and filter with SBS (f) feature selection algorithms.
Figure 9
Figure 9
Comparison of classification performance based on the accuracy of different models using SBS feature selection algorithms.
Figure 10
Figure 10
Classification accuracy of different models using all features (a) and applying mutual information (b), ANOVA (c), embedded (d), embedded with a filter (e), and filter with RFE (f) feature selection algorithms.
Figure 11
Figure 11
Classification accuracy of different models using RFE (a), filter with HRFA (b), filter with SFS (c), SFS (d), genetic (e), and filter with SBS (f) feature selection algorithms.
Figure 12
Figure 12
Comparison of classification performance based on the accuracy of different models using all features (a) and applying mutual information (b), ANOVA (c), embedded (d), embedded with a filter (e), and filter with RFE (f) feature selection algorithms.
Figure 13
Figure 13
Comparison of classification performance based on the accuracy of different models using RFE (a), filter with HRFA (b), filter with SFS (c), SFS (d), genetic (e), and filter with SBS (f) feature selection algorithms.
Figure 14
Figure 14
Comparison of ROC curves of RF, LDA, and LR algorithms with respect to the normal eyes class (a), forme fruste keratoconus class (b), mild keratoconus class (c), and advanced keratoconus class (d) using the selected variables by applying SFS method.
Algorithm 1
Algorithm 1
Random forest algorithm.
Algorithm 2
Algorithm 2
KNN algorithm.
Algorithm 3
Algorithm 3
LDA steps.
Algorithm 4
Algorithm 4
CART algorithm.
See this image and copyright information in PMC

References

    1. Adnan K., Akbar R. An analytical study of information extraction from unstructured and multidimensional big data. Journal of Big Data . 2019;6(1) doi: 10.1186/s40537-019-0254-8. - DOI
    1. Reddy G. T., Reddy M. P. K., Lakshmanna K., et al. Analysis of dimensionality reduction techniques on big data. IEEE Access . 2020;8:54776–54788. doi: 10.1109/ACCESS.2020.2980942. - DOI
    1. Obaid H. S., Dheyab S. A., Sabry S. S. The impact of data pre-processing techniques and dimensionality reduction on the accuracy of machine learning. 2019 9th Annual Information Technology, Electromechanical Engineering and Microelectronics Conference (IEMECON); March 2019; Jaipur, India. pp. 279–283. - DOI
    1. Xu X., Liang T., Zhu J., Zheng D., Sun T. Review of classical dimensionality reduction and sample selection methods for large-scale data processing. Neurocomputing . 2019;328:5–15. doi: 10.1016/j.neucom.2018.02.100. - DOI
    1. Yousefi E., Yousefi E., Takahashi H., et al. Keratoconus severity identification using unsupervised machine learning. PLoS One . 2018;13(11, article e0205998) doi: 10.1371/journal.pone.0205998. - DOI - PMC - PubMed

Publication types

LinkOut - more resources