Research Article
BibTex RIS Cite

Enhancing Giardia intestinalis Image Detection through YOLOv8-Based Deep Learning Techniques

Year 2024, , 64 - 85, 14.07.2024
https://doi.org/10.47933/ijeir.1403833

Abstract

Giardia intestinalis (G. intestinalis), a parasitic organism that causes gastrointestinal infections, represents a huge challenge in precisely identifying species from microscopic images. The complexities of accurate diagnosis and treatment require a shift towards automated solutions that enhance diagnostic efficiency and accuracy. In this study, we take advantage of the YOLOv8 deep learning model, comparing its performance with traditional methods, to enhance Giardia intestinalis detection.Our dataset, which has been carefully obtained by Burdur Mehmet Akif Ersoy University Faculty of Veterinary Medicine, Department of Pathology adds a unique dimension to our research. The dataset consists of 264 images of G. İntestinalis and is subjected to preprocessing with RGB/grayscale filters and contrast-limited adaptive histogram equalization for optimal model input.Deep learning architectures tested, including YOLOv8, show an accuracy rate of 95%. Notably, the YOLOv8 model shows promising results, indicating its potential to transform the diagnosis of G. intestinalis. Beyond immediate application, our research paves the way for the integration of YOLOv8 into broader healthcare contexts, promising effective tools for managing G. İntestinalis infections.Furthermore, our study allows the transfer of G. İntestinalis diagnostic expertise from expert veterinarians to the AI model. Veterinarians working in this field can now obtain preliminary diagnostic information through a mobile application. This innovative approach enhances the competence of veterinarians and expands their experience in this field.This research significantly pushes the boundaries in G. İntestinalis image analysis but also puts the foundation for the broader use of advanced deep learning techniques in medical applications. The implications of our findings extend beyond G. İntestinalis diagnosis, providing insight into the transformative impact of YOLOv8 in medical and biological image analysis. Our study opens the way for future developments, shaping the path of intelligent computer vision methods in real-world medical applications.

References

  • 1. Zhang, Z., Wang, L., Yang, Y., Pan, Y., & Yan, H. (2023). A review of the application of deep learning in medical image detection and segmentation. BioMed Research International, 2023.
  • 2. Redmon, J., & Farhadi, A. (2017). YOLOv3: Unifying object detection and instance segmentation. arXiv preprint arXiv:1704.06825.
  • 3. Shahriar, N. (2023, October 4). What is Convolutional Neural Network - CNN (Deep Learning). Medium. https://nafizshahriar.medium.com/what-is-convolutional-neural-network-cnn-deep-learning-b3921bdd82d5
  • 4. Roboflow. (2022). Data Augmentation in Computer Vision - Roboflow Blog. https://blog.roboflow.com/data-augmentation/: https://blog.roboflow.com/data-augmentation/Khan, S., Ahmad, M., & Lee, S. (2022). A novel YOLOv5-based deep learning framework for bone tumor detection in X-ray images. Computational and Applied Mathematics, 3, 127.
  • 5. Abbas, A. M., & Al-Jubouri, M. N. (2018). A framework for image detection using convolutional neural networks. In Proceedings of the 10th International Conference on Intelligent Systems and Applications (INTELSIS 2018) (pp. 1-4). IEEE.
  • 6. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
  • 7. Redmon, J., & Farhadi, A. (2018). YOLOv9: You Only Look Once, version 9. arXiv preprint arXiv:1804.02767.
  • 8. Zhang, H., Wang, G., & Li, Z. (2021). Generating diverse image datasets with limited labeling. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(1), 131-146.
  • 9. Jayakumaran, A. (2022). Illuminating the shadows: Bone tumor detection with YOLO and computer vision. Medium.
  • 10. Akyıldız, B. (2022). What is YOLOv8? How to use it? Medium.
  • 11. Photon, W. (2022). Revolutionizing the ability of YOLOv8: A depth insight of object detection, detection, and instance segmentation. Medium.
  • 12. Roboflow (2022). How to Train a YOLOv8 Detection Model. Roboflow Blog.
  • 13. Wang, S., Zhang, X., Yang, Y., & Zhu, W. (2021). A novel YOLOv5-based framework for bone tumor detection in X-ray images. Sensors, 21(5), 1592.
  • 14. Zhang, Y., Ma, L., & Zhang, J. (2021). A lightweight YOLOv5 model for real-time object detection on the edge device. IEEE Access, 9, 156960-156971.
  • 15. Song, Y., Liu, Y., & Zhang, S. (2019). Research on the application of deep learning in image detection based on YOLOv3. Journal of Physics: Conference Series, 1237(1), 012031.
  • 16. Zhang, S., Wang, Y., & Luo, X. (2022). Research on YOLOv5 model for apple leaf disease recognition. Journal of Physics: Conference Series, 2147(1), 012016.
  • 17. A0922 (2022). Confusion matrix of YOLOv8. Medium.
  • 18. Ultralytics (2021). Ultralytics YOLOv5. https://github.com/ultralytics: https://github.com/ultralytics

YOLOv8 Tabanlı Derin Öğrenme Teknikleri Yoluyla Giardia İntestinalis Görüntü Algılamanın Geliştirilmesi

Year 2024, , 64 - 85, 14.07.2024
https://doi.org/10.47933/ijeir.1403833

Abstract

Mide-bağırsak enfeksiyonlarına neden olan parazit bir organizma olan Giardia bağırsakis (G. bağırsakis), mikroskobik görüntülerden türlerin kesin olarak tanımlanmasında büyük bir zorluk teşkil etmektedir. Doğru teşhis ve tedavinin karmaşıklığı, teşhis verimliliğini ve doğruluğunu artıran otomatik çözümlere doğru bir geçişi gerektirir. Bu çalışmada, Giardia bağırsak tespitini geliştirmek için performansını geleneksel yöntemlerle karşılaştıran YOLOv8 derin öğrenme modelinden yararlanıyoruz. Burdur Mehmet Akif Ersoy Üniversitesi Veteriner Fakültesi Patoloji Anabilim Dalı tarafından özenle elde edilen veri setimiz, araştırmamıza benzersiz bir boyut kazandırıyor. Veri seti, G. İntestinalis'in 264 görüntüsünden oluşur ve optimum model girişi için RGB/gri tonlamalı filtreler ve kontrast sınırlı uyarlanabilir histogram eşitleme ile ön işleme tabi tutulur. YOLOv8 de dahil olmak üzere test edilen derin öğrenme mimarileri %95'lik bir doğruluk oranı gösterir. YOLOv8 modelinin umut verici sonuçlar vermesi dikkat çekicidir; bu da onun G.intestinalis tanısını dönüştürme potansiyeline işaret etmektedir. Araştırmamız, anında uygulamanın ötesinde, YOLOv8'in daha geniş sağlık hizmetleri bağlamlarına entegrasyonunun önünü açarak G. İntestinalis enfeksiyonlarının yönetilmesi için etkili araçlar vaat ediyor. Ayrıca çalışmamız, G. İntestinalis teşhis uzmanlığının uzman veteriner hekimlerden AI modeline aktarılmasına olanak tanıyor. Bu alanda çalışan veteriner hekimler artık mobil uygulama üzerinden ön teşhis bilgilerini alabiliyor. Bu yenilikçi yaklaşım, veteriner hekimlerin yetkinliğini artırıyor ve bu alandaki deneyimlerini genişletiyor. Bu araştırma, G. İntestinalis görüntü analizinde sınırları önemli ölçüde zorluyor, aynı zamanda gelişmiş derin öğrenme tekniklerinin tıbbi uygulamalarda daha geniş kullanımının temelini atıyor. Bulgularımızın sonuçları G. İntestinalis tanısının ötesine geçerek YOLOv8'in tıbbi ve biyolojik görüntü analizindeki dönüştürücü etkisine dair içgörü sağlıyor. Çalışmamız, gerçek dünyadaki tıbbi uygulamalarda akıllı bilgisayarlı görme yöntemlerinin yolunu şekillendirerek gelecekteki gelişmelerin yolunu açıyor.

References

  • 1. Zhang, Z., Wang, L., Yang, Y., Pan, Y., & Yan, H. (2023). A review of the application of deep learning in medical image detection and segmentation. BioMed Research International, 2023.
  • 2. Redmon, J., & Farhadi, A. (2017). YOLOv3: Unifying object detection and instance segmentation. arXiv preprint arXiv:1704.06825.
  • 3. Shahriar, N. (2023, October 4). What is Convolutional Neural Network - CNN (Deep Learning). Medium. https://nafizshahriar.medium.com/what-is-convolutional-neural-network-cnn-deep-learning-b3921bdd82d5
  • 4. Roboflow. (2022). Data Augmentation in Computer Vision - Roboflow Blog. https://blog.roboflow.com/data-augmentation/: https://blog.roboflow.com/data-augmentation/Khan, S., Ahmad, M., & Lee, S. (2022). A novel YOLOv5-based deep learning framework for bone tumor detection in X-ray images. Computational and Applied Mathematics, 3, 127.
  • 5. Abbas, A. M., & Al-Jubouri, M. N. (2018). A framework for image detection using convolutional neural networks. In Proceedings of the 10th International Conference on Intelligent Systems and Applications (INTELSIS 2018) (pp. 1-4). IEEE.
  • 6. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
  • 7. Redmon, J., & Farhadi, A. (2018). YOLOv9: You Only Look Once, version 9. arXiv preprint arXiv:1804.02767.
  • 8. Zhang, H., Wang, G., & Li, Z. (2021). Generating diverse image datasets with limited labeling. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(1), 131-146.
  • 9. Jayakumaran, A. (2022). Illuminating the shadows: Bone tumor detection with YOLO and computer vision. Medium.
  • 10. Akyıldız, B. (2022). What is YOLOv8? How to use it? Medium.
  • 11. Photon, W. (2022). Revolutionizing the ability of YOLOv8: A depth insight of object detection, detection, and instance segmentation. Medium.
  • 12. Roboflow (2022). How to Train a YOLOv8 Detection Model. Roboflow Blog.
  • 13. Wang, S., Zhang, X., Yang, Y., & Zhu, W. (2021). A novel YOLOv5-based framework for bone tumor detection in X-ray images. Sensors, 21(5), 1592.
  • 14. Zhang, Y., Ma, L., & Zhang, J. (2021). A lightweight YOLOv5 model for real-time object detection on the edge device. IEEE Access, 9, 156960-156971.
  • 15. Song, Y., Liu, Y., & Zhang, S. (2019). Research on the application of deep learning in image detection based on YOLOv3. Journal of Physics: Conference Series, 1237(1), 012031.
  • 16. Zhang, S., Wang, Y., & Luo, X. (2022). Research on YOLOv5 model for apple leaf disease recognition. Journal of Physics: Conference Series, 2147(1), 012016.
  • 17. A0922 (2022). Confusion matrix of YOLOv8. Medium.
  • 18. Ultralytics (2021). Ultralytics YOLOv5. https://github.com/ultralytics: https://github.com/ultralytics
There are 18 citations in total.

Details

Primary Language English
Subjects Decision Support and Group Support Systems, Artificial Life and Complex Adaptive Systems, Artificial Intelligence (Other)
Journal Section Research Articles
Authors

Osama Burak Elhalid 0000-0002-8051-7813

Ali Hakan Isık 0000-0003-3561-9375

Özlem Özmen 0000-0002-1835-1082

Early Pub Date June 24, 2024
Publication Date July 14, 2024
Submission Date December 12, 2023
Acceptance Date January 9, 2024
Published in Issue Year 2024

Cite

APA Elhalid, O. B., Isık, A. H., & Özmen, Ö. (2024). Enhancing Giardia intestinalis Image Detection through YOLOv8-Based Deep Learning Techniques. International Journal of Engineering and Innovative Research, 6(2), 64-85. https://doi.org/10.47933/ijeir.1403833

88x31.png

This work is licensed under a Creative Commons Attribution 4.0 International License