Review
BibTex RIS Cite

Periferik arter hastalığının teşhisinde yapay zeka tabanlı karar destek sistemi yaklaşımı

Year 2026, Volume: 2 Issue: 1, 11 - 18, 29.01.2026

Abstract

Periferik arter hastalıkları (PAD), bacak damarlarında plak birikimi veya daralma nedeniyle kan akışının azalması sonucu ortaya çıkan dolaşım sistemi hastalıklarıdır. Bu çalışmada, PAH tanısı için Ayak Bileği-Kol İndeksi (ABI) kullanılmıştır ve yöntem, yapay zeka tabanlı bir sınıflandırma modeli ile desteklenmiştir. İki kol ve iki bacaktan hem sistolik hem de diyastolik kan basıncını kaydedebilen özel olarak tasarlanmış bir cihaz geliştirilmiştir. 948 kişiyi içeren veri setinden, ortalama arter basıncı, nabız basıncı ve vücut kitle indeksi gibi çeşitli türetilmiş özellikler elde edilmiş; yaş da analizde kullanılan değişkenler arasına dahil edilmiştir. Sınıf dengesizliğini ortadan kaldırmak için, veriler Gauss Gürültü Tabanlı Artırma tekniği ile %50 oranında çoğaltılmıştır. Çok Katmanlı Yapay Sinir Ağı modeli, ABI < 1,0 olan bireyleri “risk altında” olarak sınıflandırılmıştır. Model, doğrulama verilerinde %97,3 ve test verilerinde %98,7 doğrulukla riskli bireyleri başarıyla sınıflandırılmıştır.

References

  • Altunkaya, S., & Bayrak, M. (2008). Korotkoff sound signals recording and analyzing system. In The International Conference on Electrical Engineering (Vol. 6, No. 6th International Conference on Electrical Engineering ICEENG 2008, pp. 1-8. Military Technical College.
  • Forghani, N., Maghooli, K., Dabanloo, N. J., Farahani, A. V., & Forouzanfar, M. (2021). Intelligent oscillometric system for automatic detection of peripheral arterial disease. IEEE Journal of Biomedical and Health Informatics, 25(8), 3209-3218.
  • Fuentes, R., & Bañuelos, M. A. (2004). Digital blood pressure monitor. Journal of Applied Research and Technology, 2(3), 224-229.
  • Gao, J. M., Ren, Z. H., Pan, X., Chen, Y. X., Zhu, W., Li, W., … Fu, G. X. (2022). Identifying peripheral arterial disease in elderly patients using machine learning algorithms. Aging Clinical and Experimental Research, 1-7.
  • Gong, Y., Cao, K. W., Xu, J. S., Li, J. X., Hong, K., Cheng, X. S., & Su, H. (2015). Valuation of normal range of ankle systolic blood pressure in subjects with normal arm systolic blood pressure. PLOS ONE, 10(6), e0122248.
  • Hirsch, A. T., Haskal, Z. J., Hertzer, N. R., et al. (2006). ACC/AHA 2006 guidelines for the management of patients with peripheral arterial disease (lower extremity, renal, mesenteric, and abdominal aortic): Executive summary. Journal of the American College of Cardiology, 47(6), 1239-1312.
  • Karabay, Ö., Karaçelik, M., Yılık, L., Tekin, N., İriz, A. B., Kumdereli, S., & Çalkavur, T. (2012). İskemik periferik arter hastalığı: Bir tarama çalışması. Türk Göğüs Kalp Damar Cerrahisi Dergisi, 20(3), 450-457.
  • Lopez, S. (2012). Blood pressure monitor fundamentals and design (Application Note AN4328). Freescale Semiconductor.
  • Marius, R. A., Iliuta, L., Guberna, S. M., & Sinescu, C. (2014). The role of ankle brachial index for predicting peripheral arterial disease. Maedica, 9(3), 295-302.
  • Vlachopoulos, C., Georgakopoulos, C., Koutagiar, I., & Tousoulis, D. (2018). Diagnostic modalities in peripheral artery disease. Current Opinion in Pharmacology, 39, 68–76.

Artificial Intelligence Based Decision Support System Approach in Peripheral Artery Disease Detection

Year 2026, Volume: 2 Issue: 1, 11 - 18, 29.01.2026

Abstract

Peripheral arterial diseases (PAD) are circulatory system diseases that occur due to reduced blood flow due to plaque accumulation or narrowing in the veins of the legs. In this study, Ankle-Braid Index (ABI) was used for the diagnosis of PAH and the method was supported by an artificial intelligence-based classification model. A custom-designed device capable of recording both systolic and diastolic blood pressure from two arms and two legs was developed. From the dataset containing 948 individuals, several derived features were obtained, including mean arterial pressure, pulse pressure, and body mass index; age was also included among the variables used in the analysis. In order to eliminate class imbalance, the data was multiplied by 50% with the Gaussian Noise-Based Augmentation technique. The Multilayer Artificial Neural Network model classified individuals with ABI < 1.0 as “at risk”. The model successfully classified risky individuals with 97.3% accuracy in the validation data and 98.7% in the test data.

References

  • Altunkaya, S., & Bayrak, M. (2008). Korotkoff sound signals recording and analyzing system. In The International Conference on Electrical Engineering (Vol. 6, No. 6th International Conference on Electrical Engineering ICEENG 2008, pp. 1-8. Military Technical College.
  • Forghani, N., Maghooli, K., Dabanloo, N. J., Farahani, A. V., & Forouzanfar, M. (2021). Intelligent oscillometric system for automatic detection of peripheral arterial disease. IEEE Journal of Biomedical and Health Informatics, 25(8), 3209-3218.
  • Fuentes, R., & Bañuelos, M. A. (2004). Digital blood pressure monitor. Journal of Applied Research and Technology, 2(3), 224-229.
  • Gao, J. M., Ren, Z. H., Pan, X., Chen, Y. X., Zhu, W., Li, W., … Fu, G. X. (2022). Identifying peripheral arterial disease in elderly patients using machine learning algorithms. Aging Clinical and Experimental Research, 1-7.
  • Gong, Y., Cao, K. W., Xu, J. S., Li, J. X., Hong, K., Cheng, X. S., & Su, H. (2015). Valuation of normal range of ankle systolic blood pressure in subjects with normal arm systolic blood pressure. PLOS ONE, 10(6), e0122248.
  • Hirsch, A. T., Haskal, Z. J., Hertzer, N. R., et al. (2006). ACC/AHA 2006 guidelines for the management of patients with peripheral arterial disease (lower extremity, renal, mesenteric, and abdominal aortic): Executive summary. Journal of the American College of Cardiology, 47(6), 1239-1312.
  • Karabay, Ö., Karaçelik, M., Yılık, L., Tekin, N., İriz, A. B., Kumdereli, S., & Çalkavur, T. (2012). İskemik periferik arter hastalığı: Bir tarama çalışması. Türk Göğüs Kalp Damar Cerrahisi Dergisi, 20(3), 450-457.
  • Lopez, S. (2012). Blood pressure monitor fundamentals and design (Application Note AN4328). Freescale Semiconductor.
  • Marius, R. A., Iliuta, L., Guberna, S. M., & Sinescu, C. (2014). The role of ankle brachial index for predicting peripheral arterial disease. Maedica, 9(3), 295-302.
  • Vlachopoulos, C., Georgakopoulos, C., Koutagiar, I., & Tousoulis, D. (2018). Diagnostic modalities in peripheral artery disease. Current Opinion in Pharmacology, 39, 68–76.
There are 10 citations in total.

Details

Primary Language English
Subjects Cardiovascular Medicine and Haematology (Other)
Journal Section Review
Authors

Onur Koçak 0000-0002-8240-4046

Zelal Onay This is me

Cansel Fıçıcı

Ziya Telatar

Submission Date July 18, 2025
Acceptance Date November 19, 2025
Publication Date January 29, 2026
Published in Issue Year 2026 Volume: 2 Issue: 1

Cite

APA Koçak, O., Onay, Z., Fıçıcı, C., & Telatar, Z. (2026). Artificial Intelligence Based Decision Support System Approach in Peripheral Artery Disease Detection. Northern Journal of Health Sciences, 2(1), 11-18. https://izlik.org/JA85SK93BE