Multi-Region Detection of eye Conjunctiva Images Using DNCNN and YOLOv8 Algorithms

Year 2024, , 1181 - 1193, 31.12.2024

Emine Cengil

https://doi.org/10.17798/bitlisfen.1539250

Abstract

Artificial intelligence is encountered in many areas today. It makes our lives easier with its use in our daily lives. With the advancement of medical big data and artificial intelligence, eye images have begun to be used in the detection of endocrine, cardiovascular, neurological, renal, hematological and many other diseases. It is possible to find more connections between systemic disorders and eye disorders and apply them to increase the effectiveness of artificial intelligence. The eye is an anatomically complex organ. Detection of the conjunctiva regions of the eye generally plays an important role in the diagnosis of eye diseases and applications related to eye health. The conjunctiva is a thin membrane tissue that covers the inner surface of the eyelids and the white part of the eye. Detection and analysis of this region is used in the examination of inflammation, redness, dryness and other disorders in the eye. The relevant regions were found using conjunctiva images in the study. Conjunctiva region detection Images were taken from a public database and enhanced with the image enhancement method DNCNN. The YOLO algorithm is applied to raw images and DNCNN enhanced images separately using the same parameters. As a result, the effect of the deep learning based method on finding the truth in images is presented with F1-confidence curve, precision-confidence curve, recall-confidence curve, precision-recall curve and confusion matrix metrics. In the proposed method, the mAP value is given as 0.984 in all classes.

Keywords

Convolutional Neural Network (CNN), Artificial Intelligence, YOLOv8, Image Enhancement, Object Detection

Ethical Statement

The study is complied with research and publication ethics.

References

• C. L. Shumway, M. Motlagh, and M. Wade, "Anatomy, head and neck, eye conjunctiva," 2018.
• R. Ram and M. B. Goren, "Palpebral Conjunctiva," in Encyclopedia of Ophthalmology, U. Schmidt-Erfurth and T. Kohnen, Eds., Berlin, Heidelberg: Springer, 2014. doi: 10.1007/978-3-642-35951-4_845-1.
• B. Çuvadar and H. Yılmaz, "Non-invasive hemoglobin estimation from conjunctival images using deep learning," Medical Engineering & Physics, vol. 120, p. 104038, 2023.
• A. K. Bitto and I. Mahmud, "Multi categorical of common eye disease detect using convolutional neural network: A transfer learning approach," Bulletin of Electrical Engineering and Informatics, vol. 11, no. 4, pp. 2378–2387, 2022.
• S. Dhalla et al., "Semantic segmentation of palpebral conjunctiva using predefined deep neural architectures for anemia detection," Procedia Computer Science, vol. 218, pp. 328–337, 2023.
• E. Purwanti, H. Amelia, M. A. Bustomi, M. A. Yatijan, and R. N. Putri, "Anemia Detection Using Convolutional Neural Network Based on Palpebral Conjunctiva Images," in 2023 14th International Conference on Information & Communication Technology and System (ICTS), pp. 117–122, IEEE, 2023.
• X. Li et al., "Identifying diabetes from conjunctival images using a novel hierarchical multi-task network," Scientific Reports, vol. 12, no. 1, p. 264, 2022.
• E. R. Ghugare, R. Patil, and Z. B. Khan, "CNN based Non-Invasive Technique for Diabetic Detection from Conjunctival Image," in 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), pp. 1–5, IEEE, 2023.
• B. Pallavi et al., "A Deep Learning-based System for Detecting Anemia from Eye Conjunctiva Images Taken from a Smartphone," IETE Technical Review, pp. 1–13, 2023.
• G. Dimauro et al., "Detecting clinical signs of anaemia from digital images of the palpebral conjunctiva," IEEE Access, vol. 7, pp. 113488–113498, 2019.
• S. Wei et al., "Developing a Deep Learning Model to Evaluate Bulbar Conjunctival Injection with Color Anterior Segment Photographs," Journal of Clinical Medicine, vol. 12, no. 2, p. 715, 2023.
• S. H. Elgohary et al., "A Machine Learning Method to Screen Anemia From Conjunctiva Images Taken by Smartphone," in 2022 10th International Japan-Africa Conference on Electronics, Communications, and Computations (JAC-ECC), pp. 125–128, IEEE, 2022.
• P. Appiahene et al., "Detection of anemia using conjunctiva images: A smartphone application approach," Medicine in Novel Technology and Devices, vol. 18, p. 100237, 2023.
• Eye conjunctiva detector, [Online]. Available: https://universe.roboflow.com/eyeconjunctivadetector/eye-conjunctiva-detector. [Accessed: Aug. 2, 2024].
• M. A. Günen and E. BEŞDOK, "Effect of denoising methods for hyperspectral images classification: DnCNN, NGM, CSF, BM3D and Wiener," Mersin Photogrammetry Journal, vol. 5, no. 1, pp. 1–9, 2022.
• K. Zhang, W. Zuo, Y. Chen, D. Meng, and L. Zhang, "Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising," IEEE Transactions on Image Processing, vol. 26, no. 7, pp. 3142–3155, Jul. 2017.
• M. Ma and H. Pang, "SP-YOLOv8s: An improved YOLOv8s model for remote sensing image tiny object detection," Applied Sciences, vol. 13, no. 14, p. 8161, 2023.
• C. Wang et al., "Strawberry Detection and Ripeness Classification Using YOLOv8+ Model and Image Processing Method," Agriculture, vol. 14, no. 5, p. 751, 2024.
• F. Solimani et al., "Optimizing tomato plant phenotyping detection: Boosting YOLOv8 architecture to tackle data complexity," Computers and Electronics in Agriculture, vol. 218, p. 108728, 2024.
• A. Yapıcı and M. A. Akcayol, "Derin Öğrenme ile Görüntülerde Gürültü Azaltma Üzerine Kapsamlı Bir İnceleme," International Journal of Advances in Engineering and Pure Sciences, vol. 34, no. 1, pp. 65–90, 2022.
• R. Padilla, S. L. Netto, and E. A. Da Silva, "A survey on performance metrics for object-detection algorithms," in 2020 International Conference on Systems, Signals and Image Processing (IWSSIP), pp. 237–242, IEEE, 2020.