Samer K. AL-SALIHI, Sezgin AYDIN, Nebras H. GHAEB

SWFT: Subbands wavelet for local features transform descriptor for corneal diseases diagnosis

Human cornea is the front see-through shield of the eye. It refracts light onto the retina to induce vision. Therefore, any defect in the cornea may lead to vision disturbance. This deficiency is estimated by sets of topographical images measured, and assessed by an ophthalmologist. Consequently, an important priority is the early and accurate diagnosis of diseases that may affect corneal integrity through the use of machine learning algorithms. Images produced by a Pentacam device can be subjected to rotation or some distortion during acquisition; therefore, accurate diagnosis requires the use of local features in the image. Accordingly, a new algorithm called subbands wavelet for local features transform (SWFT) which is mainly based on the algorithm of a scale-invariant feature transform (SIFT) has been developed. This algorithm uses wavelets as a multiresolution analysis to produce images with different scales instead of using the difference of Gaussians as in the SIFT algorithm. The experimental results on the corneal topography dataset indicate that the proposed SWFT outperforms the baseline SIFT algorithm

PDF

___

1] Doi K. Computer-aided diagnosis in medical imaging: historical review, current status and future potential. Computerized Medical Imaging and Graphics 2007; 31 (4-5): 198-211.
[2] Kononenko I. Machine learning for medical diagnosis: history, state of the art and perspective. Artificial Intelligence in Medicine 2001; 23 (1): 89-109.
[3] Yang AY, Chow J, Liu J. Focus: sensory biology and pain: corneal innervation and sensation: the eye and beyond. The Yale Journal of Biology and Medicine 2018; 91 (1): 13.
[4] Nemeth G, Vajas A, Kolozsvari B, Berta A, Modis Jr L. Anterior chamber depth measurements in phakic and pseudophakic eyes: Pentacam versus ultrasound device. Journal of Cataract and Refractive Surgery 2006; 32 (8): 1331-1335.
[5] Fontes BM, Ambrósio Jr R, Jardim D, Velarde GC, Nosé W. Corneal biomechanical metrics and anterior segment parameters in mild keratoconus. Ophthalmology 2010; 117 (4): 673-679.
[6] Cortes C, Vapnik V. Support-vector networks. Machine Learning 1995; 20, 273-297. doi: 10.1007/BF00994018
[7] Sinjab MM. Corneal Tomography in Clinical Practice (Pentacam System): Basics and Clinical Interpretation. Jaypee Brothers, Medical Publishers Pvt. Limited; 2018.
[8] Verma M, Raman B. Local neighborhood difference pattern: A new feature descriptor for natural and texture image retrieval. Multimedia Tools and Applications 2018; 77 (10): 11843-11866.
[9] Arya KV, Rajput SS, Upadhyay S. Noise-robust low-resolution face recognition using sift features. In: Computa- tional Intelligence: Theories, Applications and Future Directions-Volume II 2019; 1: 645-655.
[10] Kamdi VD, Raut AD, Pardakhe V, Jagtap VA. A study of SURF features for” face recognition. International Journal of Electronics, Communication and Soft Computing Science and Engineering 2018: 90-94.
[11] Dai-Hong J, Lei D, Dan L, San-You Z. Moving-object tracking algorithm based on PCA-SIFT and optimization for underground coal mines. IEEE Access 2019; 7: 35556-35563.
[12] Cheng Q, Shao K, Li C, Li S, Li J et al. Distributed system architecture for high-resolution remote sensing image retrieval by combining deep and traditional features. Image and Signal Processing for Remote Sensing 2018; 1: 1078918-1078919. doi: 10.1117/12.2323310
[13] Xie B, Qin J, Xiang X, Li H, Pan L. An image retrieval algorithm based on gist and SIFT Features. IJ Network Security 2018; 20 (4): 609-616.
[14] Lowe DG. Object recognition from local scale-invariant features. In: Proceedings of the Seventh IEEE International Conference on Computer Vision 1999; Kerkyra, Greece; 1999. pp. 1150-1157.
[15] Morel JM, Yu G. On the consistency of the SIFT method. 2011. [16] Ke Y, Sukthankar R. PCA-SIFT: a more distinctive representation for local image descriptors. InProceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2004; 2: II-II.
17] Mikolajczyk K, Tuytelaars T, Schmid C, Zisserman A, Matas J et al. A comparison of aﬀine region detectors. International Journal of Computer Vision 2005; 65 (1-2): 43-72.
[18] Bay H, Tuytelaars T, Van Gool L. Surf: speeded up robust features. In: European Conference on Computer Vision 2006; New York, USA; 2006. pp. 404-417.
[19] Morel JM, Yu G. ASIFT: A new framework for fully aﬀine invariant image comparison. SIAM Journal on Imaging Sciences 2009; 2 (2): 438-69.
[20] Tang H, Tang F. AH-SIFT: Augmented histogram based SIFT descriptor. In: 2012 19th IEEE International Conference on Image Processing 2012; New York, USA; 2012. pp. 2357-2360.
[21] Bauer J, Sünderhauf N, Protzel P. Comparing several implementations of two recently published feature detectors. IFAC Proceedings Volumes 2007; 40 (15): 143-148.
[22] Potthast C, Breitenmoser A, Sha F, Sukhatme GS. Active multi-view object recognition and online feature selection. InRobotics Research 2018; 1: 471-488.
[23] Dong J, Liu W, Zhu Y. Application of SIFT feature vector matching algorithm in binocular stereo vision system. In: Fifth Symposium on Novel Optoelectronic Detection Technology and Application; Xi’an, China; 2019. pp. 110231U-110241U.
[24] Alhwarin F, Ristić Durrant D, Gräser A. VF-SIFT: very fast SIFT feature matching. In: Joint Pattern Recognition Symposium; Heidelberg, Germany; 2010. pp. 222-231.
[25] Abdel-Hakim AE, Farag AA. CSIFT: A SIFT descriptor with color invariant characteristics. 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’06) 2006; 2: 1978-1983.
[26] Brown M, Lowe DG. Invariant features from interest point groups. In: BMVC 2002; 4.
[27] Nowak E, Jurie F, Triggs B. Sampling strategies for bag-of-features image classification. InEuropean Conference on Computer Vision 2006, pp. 490-503.
[28] Daubechies I. The wavelet transform, time-frequency localization and signal analysis. IEEE Transactions on Infor- mation Theory 1990; 36 (5): 961-1005.
[29] Subasi A, Yaman E. EMG Signal classification using discrete wavelet transform and rotation forest. In: International Conference on Medical and Biological Engineering 2019.pp. 29-35.
[30] Rhif M, Ben Abbes A, Farah IR, Martínez B, Sang Y. Wavelet transform application for/in non-stationary time- series analysis: A review. Applied Sciences 2019; 9 (7): 1345-1346.
[31] Meng Y, Ye C, Yu S, Qin J, Zhang J et al. Sugarcane node recognition technology based on wavelet analysis. Computers and Electronics in Agriculture 2019; 158: 68-78.
[32] Maity R, Suman M. Predictability of hydrological systems using the wavelet transformation: Application to drought prediction. In: Hydrology in a Changing World 2019, pp. 109-137.
[33] Bharath I, Devendiran S, Mathew AT. Bearing condition monitoring using tunable q-factor wavelet transform, spectral features and classification algorithm. Materials Today: Proceedings 2018; 5 (5): 11476-11490. doi: 10.1016/j.matpr.2018.02.115
[34] Koh JE, Acharya UR, Hagiwara Y, Raghavendra U, Tan JH et al. Diagnosis of retinal health in digital fundus images using continuous wavelet transform (CWT) and entropies. Computers in Biology and Medicine 2017; 84: 89-97. doi: 10.1016/j.compbiomed.2017.03.008
[35] Amiri M, Lina JM, Pizzo F, Gotman J. High frequency oscillations and spikes: separating real HFOs from false oscillations. Clinical Neurophysiology 2016; 127 (1): 187-196. doi: 10.1016/j.clinph.2015.04.290
[36] Starck JL, Murtagh F, Fadili J. Sparse Image and Signal Pprocessing: Wavelets and Related Geometric Multiscale Analysis. USA: Cambridge University Press, 2015.
[37] Bhosale NV, Kanade SS. A comparative analysis on hardware system of 2D discrete wavelet transformation for the image denoising application. In: International Conference on Intelligent Data Communication Technologies and Internet of Things; New York, USA; 2018. pp. 1076-1087
[38] Acharya UR, Raghavendra U, Fujita H, Hagiwara Y, Koh JE et al. Automated characterization of fatty liver disease and cirrhosis using curvelet transform and entropy features extracted from ultrasound images. Computers in Biology and Medicine 2016; 79: 250-258. doi: 10.1016/j.compbiomed.2016.10.022
[39] Kashyap N, Sinha GR. Image watermarking using 3-level discrete wavelet transform (DWT). International Journal of Modern Education and Computer Science 2012; 4 (3): 50-51. doi: 10.5815/ijmecs.2012.03.07
[40] Cortes C, Vapnik V. Support-vector networks. Machine Learning 1995; 20(3): 273-297. doi: 10.1007/BF00994018 [41] Suykens JA, Vandewalle J. Least squares support vector machine classifiers. Neural Processing Letters 1999; 9(3): 293-300. doi: 10.1023/A:1018628609742
[42] Dadi HS, Pillutla GM. Improved face recognition rate using HOG features and SVM classifier. IOSR Journal of Electronics and Communication Engineering 2016; 11(4): 34-44. doi: 10.9790/2834-1104013444
[43] Mun S, Park S, Han DK, Ko H. Generative adversarial network based acoustic scene training set augmentation and selection using SVM hyper-plane. Proceeding of DCASE 2017;1: 93-97.
[44] Fung GM, Mangasarian OL. Multicategory proximal support vector machine classifiers. Machine Learning 2005; 59(1-2): 77-97. doi: 10.1007/s10994-005-0463-6
[45] Bergman S. The Kernel Function and Conformal Mapping. USA: American Mathematical Society Publication, 1970.
[46] Lane F, Azad RM, Ryan C. Principled evolutionary algorithm search operator design and the kernel trick. In: 2016 IEEE Symposium Series on Computational Intelligence (SSCI); Athens, Greece; 2016. pp. 1-9.
[47] Achirul Nanda M, Boro Seminar K, Nandika D, Maddu A. A comparison study of kernel functions in the support vector machine and its application for termite detection. Information 2018; 9(1): 5.
[48] Khalaf M, Hussain AJ, Alafandi O, Al-Jumeily D, Alloghani M et al. An application of using support vector machine based on classification technique for predicting medical data sets. International Conference on Intelligent Computing 2019; 3: 580-591.
[49] Schölkopf B, Smola AJ, Bach F. Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. USA: MIT Press, 2002.
[50] Liu Y, Lian J, Bartolacci MR, Zeng QA. Density-based penalty parameter optimization on C-SVM. The Scientific World Journal 2014; 1: 1-20. doi: 10.1155/2014/851814
[51] Kaestner CA. A. Support vector machines and kernel functions for text processing. Revista de Informática Teórica e Aplicada 2013; 20(3): 130-154. doi: 10.22456/2175-2745.39702
[52] Saito T, Rehmsmeier M. The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PloS One 2015; 10 (3): e0118432-.e0118432. doi: 10.1371/journal.pone.0118432
[53] Agarwal S, Roth D. Learning a sparse representation for object detection. In: European Conference on Computer Vision 2002; USA; 2012. pp. 113-127.
[54] Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters 2006; 27 (8): 861-874. doi: 10.1016/j.patrec.2005.10.010
[55] Hashemi H, Mehravaran S. Day to day clinically relevant corneal elevation, thickness, and curvature parameter using the orbscan II scanning slit topographer and the pentacam scheimpflug imaging device. Middle East African Journal of Ophthalmology 2010; 1: 44-45. doi: 10.4103/0974-9233.61216
[56] Tschandl, P, Rosendahl C, Kittler H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Scientific Data 2018; 5: 180161. doi: 10.1038/sdata.2018.161
57] Kermany D, Zhang K, Goldbaum M. Labeled optical coherence tomography (OCT) and Chest X-Ray images for classification. Mendeley Data 2018, 2.
[58] Suckling J, Astley S, Betal D, Cerneaz N, Dance DR et al. Mammographic Image Analysis Society MiniMammo- graphic Database, 2005, v1.21 [Dataset].