Keratoconus Classification Using Multimodal Imaging Strategy

Authors

  • Mustapha AATILA Cadi Ayyad University, Faculty of science Semlalia, Laboratory of Computer Science and Smart Systems https://orcid.org/0000-0002-4330-4579
  • Ali KARTIT LTI laboratory, ENSA, Chouaib Doukkali University, El Jadida 1166, Morocco.
  • El Mehdi RAOUHI LTI laboratory, ENSA, Chouaib Doukkali University, El Jadida 1166, Morocco.

Keywords:

Keratoconus Classification, Mumtimodal data fusion, Corneal topography, Machine learning, Deep learning

Abstract

Data fusion improves the accuracy and robustness of diagnostic models by combining different types of information. This study presents a multimodal framework for keratoconus classification. It uses numeric and textual features from Pentacam reports, extracted with OCR. These are combined with corneal topographic images processed by a dual-branch deep neural network. The method was tested on 2,924 labeled Pentacam scans. Of these, 1,900 were used for training and 1,024 for testing. Scans were labeled as normal, suspicious, or keratoconus. Results show that combining image and text features improves classification. Deep learning accuracy rose from 96.78% to 98.34%. SVM improved from 93.35% to 95.60%. LDA increased from 92.85% to 94.80%, and KNN from 90.50% to 93.94%. These gains, up to 1.56% for deep learning and 3.44% for KNN, show the value of multimodal data for more accurate keratoconus diagnosis.

Downloads

Download data is not yet available.

References

X. Zhang, S. Z. Munir, S. A. Sami Karim et W. M. Munir, « A review of imaging modalities for detecting early keratoconus », Eye, vol. 35, no 1, pp. 173–187, 2021.

J. Gao, P. Li, Z. Chen et J. Zhang, « A survey on deep learning for multimodal data fusion », Neural Computation, vol. 32, no 5, pp. 829–864, 2020.

W. Chango, J. A. Lara, R. Cerezo et C. Romero, « A review on data fusion in multimodal learning analytics and educational data mining », Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, vol. 12, p. e1458, 2022.

W. Nsengiyumva, S. Zhong, M. Luo, Q. Zhang et J. Lin, « Critical insights into the state-of-the-art nde data fusion techniques for the inspection of structural systems », Structural Control and Health Monitoring, vol. 29, p. e2857, 2022.

K. Srinivas, A. Gale et J. Dolby, « Merging datasets through deep learning », arXiv preprint arXiv:1809.01604, 2018.

A. Smolinska, J. Engel, E. Szymanska, L. Buydens et L. Blanchet, « Chapter 3 – General framing of low-, mid-, and high-level data fusion with examples in the life sciences », vol. 31, pp. 51–79, 2019.

K. Y. Tey, E. Z. K. Cheong et M. Ang, « Potential applications of artificial intelligence in image analysis in cornea diseases: a review », Eye and Vision, vol. 11, no 1, p. 10, 2024.

S. Gao, Y. Chen, F. Shi, J. Peng et al., « Lkg-net: lightweight keratoconus grading network based on corneal topography », Biomedical Optics Express, vol. 14, pp. 799–814, 2023.

A. H. Al-Timemy, L. Alzubaidi, J. Mosa et al., « A deep feature fusion of improved suspected keratoconus detection with deep learning », Diagnostics, vol. 13, p. 1689, 2023.

R. Feng, Z. Xu, X. Zheng, H. Hu, J. Jin et al., « Kernet: a novel deep learning approach for keratoconus and sub-clinical keratoconus detection based on raw data of the Pentacam HR system », IEEE Journal of Biomedical and Health Informatics, vol. 25, pp. 3898–3910, 2021.

M. M. Daud, W. M. D. W. Zaki, A. Hussain et H. A. Mutalib, « Keratoconus detection using the fusion features of anterior and lateral segment photographed images », IEEE Access, vol. 8, pp. 142 282–142 294, 2020.

S. Hashemi, H. Veisi, E. Jafarzadehpur, R. Rahmani et Z. Heshmati, « An image processing approach for rigid gas-permeable lens base-curve identification », Signal, Image and Video Processing, vol. 14, pp. 971–979, 2020.

A. Upadhyaya, S. Khan, P. Sahay, S. Goel et D. Kumawat, « Pentacam – A corneal tomography system », Delhi Journal of Ophthalmology, vol. 31, no 1, pp. 90–95, 2020.

C. Thorat, A. Bhat, P. Sawant, I. Bartakke et S. Shirsath, « A detailed review on text extraction using optical character recognition », ICT Analysis and Applications, pp. 719–728, 2022.

A. B. Abadi et S. Tahcfulloh, « Digital image processing for height measurement application based on Python OpenCV and regression analysis », JOIV: International Journal on Informatics Visualization, vol. 6, no 4, pp. 763–770, 2022.

R. K. Cho, K. Sood et C. S. C. Channapragada, « Image repair and restoration using deep learning », dans 2022 4th International Conference on Artificial Intelligence and Speech Technology (AIST). IEEE, 2022, pp. 1–8.

Downloads

Published

2025-06-30

How to Cite

[1]
M. AATILA, A. KARTIT, and E. M. RAOUHI, “Keratoconus Classification Using Multimodal Imaging Strategy”, IJCEDS, vol. 4, no. 2, pp. 1–9, Jun. 2025.

Issue

Vol. 4 No. 2 (2025): Volume 4. Issue 2

Section

Articles

ARK

https://n2t.net/ark:/32155/IJCEDS.v4i2.88

License

Copyright (c) 2025 Mustapha AATILA, Ali KARTIT , El Mehdi RAOUHI

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Copyright on any article published in the International Journal of Computer Engineering and Data Science (IJCEDS) is retained by the author(s). All articles are published under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0), which permits any non-commercial use, distribution, and reproduction in any medium, provided that the original work is properly cited.

License Agreement

By submitting and publishing their work in IJCEDS, the authors:

  • Grant IJCEDS the non-exclusive right to publish the article and to identify IJCEDS as the original publisher.

  • Authorize any third party to use, share, and reproduce the article for non-commercial purposes, provided that appropriate credit is given to the original authors and source, and a link to the license is included.