YOLOV11 MODELLERİ İLE MİKROSKOBİK GÖRÜNTÜLERDEN İDRAR SEDİMENT PARÇACIKLARININ TESPİTİ

Yusuf Furkan Yalçın; Feyza Altunbey Özbay

doi:10.17780/ksujes.1690992

TR EN

YOLOV11 MODELLERİ İLE MİKROSKOBİK GÖRÜNTÜLERDEN İDRAR SEDİMENT PARÇACIKLARININ TESPİTİ

Öz

Böbrek hastalıkları, böbrek taşları ve diğer idrar yolu enfeksiyonlarının teşhisi ve tedavisi için idrardaki sediment parçacıkların analiz edilmesi ve incelenmesi kritik önem taşımaktadır. İdrar tahlili örneklerinin toplanması ve analizinin manuel yapılması insan gücü ve zaman gerektirdiği için beşerî hata riski taşımaktadır. Bu sorunların üstesinden gelmek için Evrişimli Sinir Ağlarına (CNNs) dayanan gelişmiş bir derin öğrenme modeli olan YOLOv11’in beş farklı varyantı ilk kez bu çalışmada idrar sediment parçacıklarının tespiti kullanılmıştır. YOLOv11’nin beş varyantı (YOLOv11n, YOLOv11s, YOLOv11m, YOLOv11l, YOLOv11x) ile mikroskobik idrar parçacıkların görüntüleri yedi kategoride (erythrocyte, leukocyte, epithelial cell, crystals, cast, mycete, epithelial nuclei) incelenmiştir. 5376 idrar sediment görüntüsünden oluşan veri kümesi için yapılan değerlendirmelerde, tüm varyantlar arasında YOLOv11s 0.5 IoU eşiğinde %89.6’lık mAP ile en yüksek değeri elde ederken, YOLOv11l ise %89.4 mAP değeri ile ikinci yüksek performansı göstermiştir. Görüntü başına tespit hızı ise YOLOv11s ve YOLOv11l için sırasıyla 6.7 ms ve 23.3 ms olarak elde edilmiştir. Bu çalışmada elde edilen sonuçlar, mikroskobik görüntülerden idrar sediment parçacıklarının tespit edilmesi için YOLOv11 modelinin uygulanabilme potansiyelini vurgulamaktadır.

Anahtar Kelimeler

DETECTION OF URINARY SEDIMENT PARTICLES FROM MICROSCOPIC IMAGES USING YOLOV11 MODELS

Abstract

The analysis and examination of urinary sediment particles are critically important for the diagnosis and treatment of kidney diseases, kidney stones, and other urinary tract infections. Since the collection and manual analysis of urine samples are labor-intensive and time-consuming processes, they carry a significant risk of human error. To address these challenges, this study proposes for the first time the use of five different variants of YOLOv11, a deep learning model based on Convolutional Neural Networks (CNNs), for the detection of urinary sediment particles. The five YOLOv11 variants (YOLOv11n, YOLOv11s, YOLOv11m, YOLOv11l, and YOLOv11x) were applied to microscopic images of urine samples across seven particle categories: erythrocyte, leukocyte, epithelial cell, crystals, cast, mycete, and epithelial nuclei. Based on evaluations conducted on a dataset comprising 5376 urinary sediment images, YOLOv11s achieved the highest performance with a mAP of 89.6% at a 0.5 IoU threshold, followed by YOLOv11l with a mAP of 89.4%. In terms of detection speed per image, YOLOv11s and YOLOv11l reached 6.7 ms and 23.3 ms, respectively. The results of this study highlight the potential applicability of the YOLOv11 model for the accurate and efficient detection of urinary sediment particles in microscopic images.

Keywords

References

Akhtar, S., Hanif, M., Rashid, A., Aurangzeb, K., Khan, E. A., Saraoglu, H. M., & Javed, K. (2024). An optimized data and model centric approach for multi-class automated urine sediment classification. IEEE Access. https://doi.org/10.1109/ACCESS.2024.3385864
Avci, D., Leblebicioglu, M. K., Poyraz, M., & Dogantekin, E. (2014). A new method based on adaptive discrete wavelet entropy energy and neural network classifier (ADWEENN) for recognition of urine cells from microscopic images independent of rotation and scaling. Journal of Medical Systems, 38, 1–9. https://doi.org/10.1007/s10916-014-0007-3
Diwan, T., Anirudh, G., & Tembhurne, J. V. (2023). Object detection using YOLO: challenges, architectural successors, datasets and applications. Multimedia Tools and Applications, 82(6), 9243–9275. https://doi.org/10.1007/s11042-022-13644-y
Franti, P., & Mariescu-Istodor, R. (2023). Soft precision and recall. Pattern Recognition Letters, 167, 115–121. https://doi.org/10.1016/j.patrec.2023.02.005
Hao, F., Li, X., Li, M., Wu, Y., & Zheng, W. (2022). An accurate urine red blood cell detection method based on multi-focus video fusion and deep learning with application to diabetic nephropathy diagnosis. Electronics, 11(24), 4176. https://doi.org/10.3390/electronics11244176
Ji, Q., Jiang, Y., Wu, Z., Liu, Q., & Qu, L. (2023). An image recognition method for urine sediment based on semi-supervised learning. IRBM, 44(2), 100739. https://doi.org/10.1016/j.irbm.2022.09.006
Ji, Q., Li, X., Qu, Z., & Dai, C. (2019). Research on urine sediment images recognition based on deep learning. IEEE Access, 7, 166711–166720. https://doi.org/ 10.1109/ACCESS.2019.2953775
Khalid, Z. M., Hawezi, R. S., & Amin, S. R. M. (2022, February). Urine sediment analysis by using convolution neural network. In 2022 8th International Engineering Conference on Sustainable Technology and Development (IEC) (pp. 173–178). IEEE. https://doi.org/10.1109/IEC54822.2022.9807482

Khanam, R., & Hussain, M. (2024). Yolov11: An overview of the key architectural enhancements. arXiv preprint arXiv:2410.17725. https://doi.org/ 10.48550/arXiv.2410.17725
Liang, Y., Tang, Z., Yan, M., & Liu, J. (2018a). Object detection based on deep learning for urine sediment examination. Biocybernetics and Biomedical Engineering, 38(3), 661–670. https://doi.org/10.1016/j.bbe.2018.05.004
Liang, Y., Kang, R., Lian, C., & Mao, Y. (2018b). An end-to-end system for automatic urinary particle recognition with convolutional neural network. Journal of Medical Systems, 42(9), 165. https://doi.org/10.1007/s10916-018-1014-6
Lyu, H., Xu, F., Jin, T., Zheng, S., Zhou, C., Cao, Y., ... & Li, D. (2023). Automated detection of multi-class urinary sediment particles: An accurate deep learning approach. Biocybernetics and Biomedical Engineering, 43(4), 672–683. https://doi.org/10.1016/j.bbe.2023.09.003
Nagai, T., Onodera, O., & Okuda, S. (2022). Deep learning classification of urinary sediment crystals with optimal parameter tuning. Scientific Reports, 12(1), 21178. https://doi.org/10.1038/s41598-022-25385-x
Suhail, K., & Brindha, D. (2024). Microscopic urinary particle detection by different YOLOv5 models with evolutionary genetic algorithm based hyperparameter optimization. Computers in Biology and Medicine, 169, 107895. https://doi.org/10.1016/j.compbiomed.2023.107895
Suhail, K., & Brindha, D. (2021). A review on various methods for recognition of urine particles using digital microscopic images of urine sediments. Biomedical Signal Processing and Control, 68, 102806. https://doi.org/10.1016/j.bspc.2021.102806
Sun, Q., Yang, S., Sun, C., & Yang, W. (2019). Exploiting aggregate channel features for urine sediment detection. Multimedia Tools and Applications, 78, 23883–23895. https://doi.org/10.1007/s11042-018-6241-9
Yan, M., Liu, Q., Yin, Z., Wang, D., & Liang, Y. (2020). A bidirectional context propagation network for urine sediment particle detection in microscopic images. In ICASSP 2020 – 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 981–985). IEEE. https://doi.org/10.1109/ICASSP40776.2020.9054367

Details

Primary Language

Turkish

Subjects

Artificial Intelligence (Other)

Journal Section

Research Article

Authors

Yusuf Furkan Yalçın
0009-0008-8683-9929
Türkiye

Feyza Altunbey Özbay ^*
0000-0003-0629-6888
Türkiye

Publication Date

December 3, 2025

Submission Date

May 3, 2025

Acceptance Date

October 10, 2025

Published in Issue

Year 2025 Volume: 28 Number: 4

DOI

https://doi.org/10.17780/ksujes.1690992

IZ

https://izlik.org/JA38ES42CT

Cite

RIS / Bibtex

APA

Yalçın, Y. F., & Altunbey Özbay, F. (2025). YOLOV11 MODELLERİ İLE MİKROSKOBİK GÖRÜNTÜLERDEN İDRAR SEDİMENT PARÇACIKLARININ TESPİTİ. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 28(4), 1748-1758. https://doi.org/10.17780/ksujes.1690992