Identification of Potential Diabetes Mellitus Through Iris Images
Keywords:
Iridology, K-NN, Matlab, Image Processing, SVMAbstract
Diagnosis of a disease is usually carried out in a laboratory test, where this test requires a lot of money and time. As time goes by in the world of health, there is one method that can be used to detect the level of condition in the body by recognizing the patterns that form on the iris of the eye or better known as Iridology. By using digital image processing, the disease diagnosis process using the iridology method can be carried out. The aim of this program is to identify diabetes milletus by processing digital images in the form of iris images using Matlab . The research data used comes from the open source website, namely Github , which consists of diabetic iris image data and normal iris image data. Data is taken in the form of digital images. In this research there are several stages, including pre-processing, feature extraction, and classification. The feature extraction process is based on statistical characteristics (contrast, correlation, energy, homogeneity) from the iris image, then classified using the K-Nearest Neighbors (KNN) algorithm and Support Vector Machine (SVM) as a comparison algorithm. The results obtained from this research show that K-NN has better performance with accuracy values of 84%, sensitivity 72%, and specificity 96%.
References
Bouaziz, Draa, A., & Chikhi, S. (2015). Artificial bees for multilevel thresholding of iris images. Swarm Evol. Comput, 21 , 32–40. https://doi.org/10.1016/j.swevo.2014.12.002
Cahyanti, D., Rahmayani, A., & Ainy, S. (2020). Analysis of the performance of the Knn method on a dataset of patients with breast cancer. Indonesia. J. Data Sci., 1 , 39–43.
Cholil, RS, Handayani, T., Prathivi, R., & Ardianita, T. (2021). Implementation of the K-Nearest Neighbor (KNN) Classification Algorithm for Scholarship Acceptance Selection Classification. IJCIT Journal, 6 (2), 118-127. https://ejournal.bsi.ac.id/ejurnal/index.php/ijcit/article/view/10438
Diatri, IGAA (2018). Iris Identification with Snake Model-PSO and Gabor 2-D. JST (Journal of Science and Technology, 7 (1), 25. https://doi.org/10.23887/jst-undiksha.v7i1.13010
Divya, D.C., Gururaj, L.H., Rohan, R., Bhagyalakshmi, V., Rashmi, A.H., Domnick, A., Flammini, F. (2021). An efficient machine learning approach to nephrology through iris recognition. ResearchGate Journal, 1 (1), 10. https://www.researchgate.net/publication/355200132_An_efficient_machine_learning_approach_to_nephrology_through_iris_recognition
MM, R., Wijayanto, I., & Aulia, S. (2019). Biometric Iris Recognition Using Lbp With Classifiers Knn Biometrick Iris Recognition Using Dwt With Classifiers K-Nearest. 6 (1), 817–825.
Nugraha, ARD, Auliasari, K., & Agus Pranoto, Y. (2020). IMPLEMENTATION OF THE K-NEAREST NEIGHBOR (KNN) METHOD FOR SELECTION OF PROSPECTIVE NEW EMPLOYEES (Case Study: BFI Finance Surabaya). JATI (Informatics Engineering Student Journal), 4 (2), 14–20. https://doi.org/10.36040/jati.v4i2.2656
Pavaloi, I., Nita, C.D., & Lazar, L.C. (2019). Novel matching method for automatic iris recognition using SIFT features. ISSCS 2019 - Int. Symp. Signals, Circuits Syst , 1–4. https://doi.org/10.1109/ISSCS.2019.8801797
Rahardja, A., Juardi, T., & Agung, H. (2019). Implementation of the K-Nearest Neighbor Algorithm on Laptop Recommendation Websites. J. Buana Inform, 10 (1), 75. https://doi.org/10.24002/jbi.v10i1.1847
Rahmadya. (2019). Training, Validating, Testing and Corpus. Accessed from https://rahmadya.com/
Rahmawati & Amiruddin. (2017). Glycohemoglobin, Hypertension, IMT Against Vision Impairment in Elderly Diabetes Mellitus Patients. Public Health Media Journal, 13 (1), 1528. https://journal.unhas.ac.id/index.php/mkmi/article/view/1582
Rosiva, AS, Zarlis, M., & Wanayumini. (2022). Leaf Image Classification with GLCM (Gray Level Co-Occurrence) and K-NN (K-Nearest Neighbor). Journal of Matrics, 21 (2), 477-486. https://journal.universitasbumigora.ac.id/index.php/matrik/article/view/1572
Shidiq, F., Hidayat, WE, & Kurniati, IN (2021). Application of the K-Nearest Neighbor (KNN) Method to Determine Betta Fish by Extracting Shape and Canny Features. Innovatics Journal, 3 (2), 39-46. https://jurnal.unsil.ac.id/index.php/innovatics/article/view/3093
Sucipto, DB, & Riana, D. (2013). Potential Glaucoma Diagnosis Application Through Iris Images Using Artificial Neural Networks Using Back Propagation Method. 1 (3), 19–27.
Suhendri & Rahayu, P. (2019). Gray Level Co-Occurrence (GLCM) Method for Classifying Water Guava Leaf Types Using the Neural Network Algorithm. Joint Journal, 01 (01), 15-22. https://jurnal.stmik-amikbandung.ac.id/joint/article/view/4
Wahid, GSF, Purnamasari, R., & Saidah, S. (2019). Personal Identification Through the Iris of the Eye Using the Compound Local Binary Pattern Method and Support Vector Machine Classification Personal Identification Based on Compound Local Binary. 6 (2), 3959–3966.
Yahya, Y., & Puspita, WH (2020). Application of the K-Nearest Neighbor Algorithm to Classify the Effectiveness of Vape (Electric Cigarette) Sales in 'Lombok Vape On. Infotek J. Inform. and Technol., 3 (2), 104–114. https://doi.org/10.29408/jit.v3i2.2279
Yani, S., Jumeilah, FS, & Kadafi, M. (2020). K-Nearest Neighbor Algorithm to Determine the Eligibility of Families Receiving Non-Cash Food Assistance (Case Study: Karya Jaya Village). Journal of Information Technology Ampera, 1 (2), 75–87. https://doi.org/10.51519/journalita.volume1.isssue2.year2020.page75-87
Yustanti, W. (2012). K-Nearest Neighbor Algorithm to Predict Land Selling Prices. JMSK: Journal of Mathematics, Statistics & Computing, 9 (1), 57-68. http://journal.unhas.ac.id/index.php/jmsk
