Yapay Afterimage Algoritması Kullanarak Medikal Verilerde Risk Tespiti

Abstract

Metaheuristic algoritmalar, insan sağlığını ilgilendiren; genetik, biyoinformatik, mikrobiyoloji vb. birçok alanda sıkça kullanılan bir metodoloji olmuştur. Metaheuristic yöntemler, klasik yaklaşımlar başarısız olduğunda veya hesaplama açısından pahalı olduğunda verimli çözümler sunar. Bu çalışmada Yapay Afterimage Algoritması, Kaggle’dan elde edilen 4 ayrı medikal veri seti üzerine uygulanmıştır. Literatürde daha önce Afterimage Algoritmasını sezgisel bir yöntem olarak modelleyen bir çalışma yoktur. Matematiksel alt yapısı diğer birçok yönteme göre daha basittir. Yapay Afterimage Algoritması kullanılarak, sağlıklı kişiler ve hastalardan alınan test örneklerinin kümeleri tespit edilmeye çalışılmıştır. Kümelere ait Accuracy, Precision, Recall ve F1 değerleri hesaplanmıştır. En yüksek Accuracy değeri 0,85, Precision değeri 0,9, Recall değeri 1 ve F1 skor değeri 0,86 olarak elde edilmiştir. Çalışma göstermektedir ki, yöntem medikal verilerde iyi bir oranda risk tespitini gerçekleştirebilmektedir.

Keywords

• Metasezgisel
• Yapay Afterimage Algoritması
• Kümeleme

Risk Detection in Medical Data Using Artificial Afterimage Algorithm

Abstract

Metaheuristic algorithms have become a frequently used methodology in many fields such as genetics, bioinformatics, microbiology, etc. related to human health. Metaheuristic methods provide efficient solutions when classical approaches fail or are computationally expensive. In this study, Artificial Afterimage Algorithm was applied to 4 different medical data sets obtained from Kaggle. There is no previous study in the literature that models the afterimage algorithm as a heuristic method. Its mathematical infrastructure is simpler than many other methods. Using the Artificial Afterimage Algorithm, clusters of test samples taken from healthy individuals and patients were tried to be detected. Accuracy, precision, recall and F1 values of the clusters were calculated. The highest Accuracy value was obtained as 0.85, Precision value as 0.9, Recall value as 1 and F1 score value as 0.86. The study shows that the method can perform a good rate of risk detection in medical data.

Keywords

• Metaheuristic
• Artificial Afterimage Algorithm
• Clustering

References

1. Ersöz B, Özmen M. Dijitalleşme ve Bilişim Teknolojilerinin Çalışanlar Üzerindeki Etkileri. AJIT-e: 2020; 11(42): 170-179.
2. Aydın A. Devlet erkinin yönetim paradigmasının yapay zeka bağlamında dönüşümü. G. Telli (Ed.), Yapay Zeka ve Gelecek içinde İstanbul: Doğu Kitapevi; 2019.
3. McCarthy J, Minsky ML, Rochester N, Shannon CE. “A proposal for the Dartmouth summer research project on artificial intelligence” https://www-formal.stanford.edu/jmc/history/dartmouth/dartmouth.html(accessed Nov 15, 2024).
4. McCarthy J. “What is artificial intelligence?” http://www-formal.stanford.edu/jmc/whatisai/(accessed Nov 18, 2024).
5. Şimşir İ, Mete B. Sağlık Hizmetlerinin Geleceği: Dijital Sağlık Teknolojileri. JOINIHP 2021; 2(1): 33-39.
6. Yorgancıoğlu Tarcan G, Yalçın Balçık, P, Sebik, NB. Türkiye ve Dünyada Sağlık Hizmetlerinde Yapay Zekâ. MEÜ Tip Fakültesi Tıp Tarihi ve Folklorik Tıp Dergisi 2024; 14(1):50-60.
7. Rajwar K, Deep K, Das S. An exhaustive review of the metaheuristic algorithms for search and optimization: taxonomy, applications, and open challenges. Artif Intell Rev 2023; 56: 13187–13257.
8. Faramarzi A, Heidarinejad M, Stephens B, Mirjalili S. Equilibrium optimizer: A novel optimization algorithm. Knowl Based Syst 2020; 191.

9. Demir M. Artificial Afterimage Algorithm: A New Bio-Inspired Metaheuristic Algorithm and Its Clustering Application. Appl Sci (Basel) 2025; 15(3): 1359.
10. Bender, MB, Feldman M, Sobin AJ. Palinopsia. Brain 1968; 91(2): 321–338.
11. Gersztenkorn D, Lee AG. Palinopsia revamped: A systematic review of the literatüre. Surv Ophthalmol 2014; 60: 1–35.
12. L˝orinc G, András H. Analyzing afterimage strength and duration test results with k-means clustering. In Szemelvények a BGE Kutatásaiból (II. Kötet) Budapest; 2023.
13. Lee HC. Introduction to Color Imaging Science Cambridge University Press: Cambridge; 2005.
14. Visual Angle-Wikipedia. Available online: https://en.wikipedia.org/wiki/Visual_angle(accessed Nov 20, 2024).
15. Taşkın Ç, Emel, GG. Clustering Approaches in Data Mining and an Application with Kohonen Networks in Retailing Sector. Süleyman Demirel University the Journal of Faculty of Economics and Administrative Sciences 2010; 15(3): 395-409.
16. Avşar İİ. Clustering of Türkiye and European Union Countries by Length of Railroad Lines. Journal of The Faculty of Applied Sciences of Tarsus University 2023; 3(1): 13-25.
17. Çatak FÖ. Development of data mining software framework by using map/reduce method in cloud computing systems. Phd. Thesis, İstanbul University Institute of Science, Department of Electronics, Informatics Program, İstanbul, 2014.
18. Chen RC, Dewi C, Huang SW, Caraka RE. Selecting critical features for data classification based on machine learning methods. J Big Data 2020; 7:52.
19. Yeşildal G. Diagnosing COVID-19 Disease through Medical Images. MSc, Ankara University Institute of Science, Department of Computer Engineering, Ankara, 2022.
20. Aslanyürek M, Mesut A. A New Method to Measure Clustering Performance and its Evaluation for Text Clustering. Eur J Sci Technol 2021; 27:, 53-65.
21. Colon Cancer, https://www.kaggle.com/datasets/lakshmi25npathi/colon-cancer(accessed Dec 10, 2024).
22. Prostate Cancer, https://www.kaggle.com/datasets/sajidsaifi/prostate-cancer(accessed Dec 10, 2024).
23. Indian Liver Patient Records, https://www.kaggle.com/datasets/uciml/indian-liver-patient-records(accessed Dec 10, 2024).
24. Coronary Heart Disease, https://www.kaggle.com/datasets/billbasener/coronary-heart-disease?resource=download (accessed Dec 10, 2024).