classification; decision tree; image processing; k-nearest neighbors; Keratoconus; multi-layer perceptron; radial basis function; support vector machine; Base function; Descriptors; Disease detection; Images processing; Multilayers perceptrons; Orbscan II; Radial base function; Radial basis; Support vectors machine; Biomedical Engineering
Abstract :
[en] Keratoconus is an eye disease causing progressive corneal thinning. At an early stage (fruste keratoconus), the symptoms can overlap with those of other eye disorders, making the diagnosis difficult. This led us to propose a new image processing pipeline to automatically calculate the main descriptors used by Ophthalmologists to detect keratoconus, and then classify these data in order to propose an intelligent system able to help specialists in the early recognition of this pathology. To accomplish this, we elaborated a new benchmark database from a corneal topographic Orbscan II device. For keratoconus classification, five different machine learning methods are tested on our new locally collected database, which are: K-Nearest Neighbors (K-NN), Support Vector Machines (SVM), Decision Trees (DT), and two neural networks classifiers (the Radial Basis Function (RBF) and the Multi-Layer Perceptron (MLP)). Experimental results indicate that all classifiers achieved good precision when using all descriptors (the numerical parameters given by the ORBSCAN II topographer and the descriptors obtained after an image processing of the topography map). Furthermore, the SVM outperformed the other classifiers with an accuracy rate of 95.04%. These results were confirmed and validated by a group of experts in ophthalmology and prove the efficiency of our system and the coherence of our new database.
F114 - Informatique, Logiciel et Intelligence artificielle
Research institute :
R300 - Institut de Recherche en Technologies de l'Information et Sciences de l'Informatique R450 - Institut NUMEDIART pour les Technologies des Arts Numériques
Maeda N, Klyce SD, Smolek MK, Thompson HW, Automated keratoconus screening with corneal topography analysis, Invest Ophthalmol Vis Sci 35:2749-2757, 1994.
Razmjoo H, Peyman A, Rahimi A, Modrek H, Cornea collagen cross-linking for keratoconus: A comparison between accelerated and conventional methods, Adv Biomed Res 6:10, 2017.
Souza MB, Medeiros FW, Souza DB, Garcia R, Alves MR, Evaluation of machine learning classifiers in keratoconus detection from orbscan II examinations, Clinics 65(12):1223-1228, 2010.
Hidalgo R, Rodriguez P, Rozema JJ, NíDhubhghaill S, Zakaria N, Tassignon MJ, Koppen C, Evaluation of a machine-learning classifier for keratoconus detection based on Scheimpflug tomography, Cornea 35(6):827-832, 2016.
Hidalgo IR, Rozema JJ, Saad A, Koppen C, Validation of an objective keratoconus detection system implemented in a scheimpflug tomographer and comparison with other methods, Cornea 36(6):689-695, 2017.
Gatinel D, Screening for subclinical keratoconus and prevention of corneal ectasia with SCORE analyzer software, in Surgical Correction of Astigmatism, Springer International Publishing, New York, pp. 103-123, 2018.
Ali AH, Ghaeb NH, Musa ZM, Support vector machine for keratoconus detection by using topographic maps with the help of image processing techniques, IOSR J Pharm Biol Sci 12(6):50-58, 2017.
Castro-Luna1 G, Jimenez-Rodríguez D, Castaño-Fernandez AB, Perez-Rueda A, Diagnosis of subclinical keratoconus based on machine learning techniques, J Clin Med 10:4281, 2021. https://doi.org/10.3390/jcm10184281.
Lavric A, Anchidin L, Popa V, Keratoconus severity detection from elevation, topography and pachymetry raw data using a machine learning approach, IEEE Access 9, 2021.
Zaki WMDW, Daud MM, Saad AH, Hussain A, Mutalib HA, Towards automated keratoconus screening approach using lateral segment photographed images, 2020 IEEEEMBS Conf Biomedical Engineering and Sciences (IECBES), Malaysia, pp. 466-471, 2021.
Xu C, Zhao J, Iselin KC, Borroni D, Romano D, Gokul A, McGhee CNJ, Zhao Y, Sedaghat M-R, Hamed Momeni-Moghaddam H, Mohammed Ziaei M, Kaye S, Romano V, Zheng Y, Keratoconus detection of changes using deep learning of colour-coded maps, BMJ Open Ophthalmol 2021. doi: 10.1136/bmjophth-2021-000824.
Feng R, Xu Z, Zheng X, KerNet: A novel deep learning approach for keratoconus and subclinical keratoconus detection based on raw data of the Pentacam HR system, IEEE J Biomed Health Inform 25(10):3898-3910, 2021.
Mahmoud HAH, Mengash HA, Automated keratoconus detection by 3D corneal images reconstruction, Sensors 21(7):2326, 2021.
Subramanian P, Ramesh GP, Keratoconus classification with convolutional neural networks using segmentation and index quantification of eye topography images by particle swarm optimisation, BioMed Res Int 9: 2022, 2022, https://doi.org/10.1155/2022/ 8119685.
Emre S, Doganay S, Yologlu S, Evaluation of anterior segment parameters in keratoconic eyes measured with the Pentacam system, J Cataract Refract Surg 33(10):1708-1712, 2007.
Alio JL, Piñero DP, Aleson A, Teus MA, Barraquer RI, Murta J, Keratoconus-integrated characterization considering anterior corneal aberrations, internal astigmatism, and corneal biomechanics, J Cataract Refract Surg 37(3):552-568, 2011
Cavas-Martínez F, De la Cruz Sanchez E, Nieto Martínez J, Fernandez Cañavate FJ, Fernandez-Pacheco DG, Corneal topography in keratoconus: State of the art, Eye Vis 3:5, 2016.
Huang D, Tang M, Shekhar R, Mathematical model of corneal surface smoothing after laser refractive surgery, Am J Ophthalmol 135(3):267-278, 2003.
Albertazzi R, Tratamiento del queratocono con segmentos intra corneales, in Albertazzi R (ed.), Queratocono: Pautas Para su Diagnostico y Tratamiento, Edicione Scientificas Argentina Para la Keratoconus Society, Buenos Aires, pp. 205-268, 2010.
Bessaid A, Feroui A, Messadi M, Detection of blood vessels from retinal images using watershed transformation, J Mech Med Biol 9(4):1-10, 2009.
Saka M, Hlavac V, Boyle R, Image Processing, Analysis, and Machine Vision, 3rd edn., Thomson, 2008.
EL Habib Daho M, Chikh A, Combining bootstrapping samples, random subspaces and random forests to build classifiers, J Med Imaging Health Infor 5(3):539-544 (6), 2015, https://doi.org/10.1166/jmihi.2015.1423