Segementasi Optik Disc dan Cup untuk Membantu Pendeteksian Glaukoma Menggunakan Segmentation Transformer
Abstract
Abstrak-Glaukoma kondisi di mana saraf optik yang menghubungkan mata ke otak menjadi rusak. Glaukoma dapat menyebabkan kehilangan kemampuan penglihatan jika tidak didiagnosis dan ditangani secepat mungkin. Salah satu metode yang dilibatkan dalam mendiagnosis glaukoma menghitung rasio antara optik disc dan cup citra fundus mata. Untuk menghitung rasio antara disc dan cup citra fundus mata, diperlukan sebuah proses segmentasi citra fundus mata untuk dapat mensegmentasikan bagian disc dan cup nya. Saat ini tugas segmentasi dapat dilakukan menggunakan algoritma visi komputer modern. Transformer sendiri telah menjadi salah satu state art of model yang sering diterapkan studi kasus yang menggunakan deep learning karena performanya yang mampu menandingi Convolutinal Neural Networks (CNN). Tugas akhir ini akan membahas implementasi Transformer studi kasus segmentasi disc dan cup citra fundus mata menggunakan metode Segmentation Transformer (SETR) dengan dataset REFUGE dan DRISHTI-GS1. Hasil dice coefficients score dengan menggunakan Cross Dataset Evaluation berhasil mendapatkan skor 86 persen untuk bagian disc dan 78 persen untuk bagian cup.
Kata kunci - glaukoma, disc, cup, segmentasi, segmentation transformers, transformers.
References
S. Zheng et al., “Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers,” 2020, doi: 10.1109/cvpr46437.2021.00681.
A. Sevastopolsky, “Optic disc and cup segmentation methods for glaucoma detection with modification of U-Net convolutional neural network,” Pattern Recognit. Image Anal., vol. 27, no. 3, pp. 618–624, 2017, doi: 10.1134/S1054661817030269.
S. Sreng, N. Maneerat, K. Hamamoto, and K. Y. Win, “Deep learning for optic disc segmentation and glaucoma diagnosis on retinal images,” Appl. Sci., vol. 10, no. 14, 2020, doi: 10.3390/app10144916.
S. Li, X. Sui, X. Luo, X. Xu, Y. Liu, and R. Goh, “Medical Image Segmentation using Squeeze-and-Expansion Transformers,” pp. 807–815, 2021, doi: 10.24963/ijcai.2021/112.
J. I. Orlando et al., “REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs,” Med. Image Anal., vol. 59, 2020, doi: 10.1016/j.media.2019.101570.
J. Sivaswamy, S. R. Krishnadas, and A. Chakravarty, “Dataset for the Assessment of Glaucoma from the Optic Nerve Head Analysis,” JSM Biomed Imaging Data Pap 2(1) 1004, vol. 2, pp. 1–7, 2015.
A. Vaswani et al., “Attention is all you need,” Adv. Neural Inf. Process. Syst., vol. 2017-Decem, no. Nips, pp. 5999–6009, 2017.
H. Fu, J. Cheng, Y. Xu, D. W. K. Wong, J. Liu, and X. Cao, “Joint Optic Disc and Cup Segmentation Based on Multi-Label Deep Network and Polar Transformation,” IEEE Trans. Med. Imaging, vol. 37, no. 7, pp. 1597–1605, 2018, doi: 10.1109/TMI.2018.2791488.
N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov, and S. Zagoruyko, “End-to-End Object Detection with Transformers,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2020, vol. 12346 LNCS, no. 7, pp. 213–229, doi: 10.1007/978-3-030-58452-8_13.
A. Dosovitskiy et al., “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” 2020, [Online]. Available: http://arxiv.org/abs/2010.11929.
S. Bakas et al., “Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge,” 2018, [Online]. Available: http://arxiv.org/abs/1811.02629.
T. Ridnik, E. Ben-Baruch, A. Noy, and L. Zelnik-Manor, “ImageNet-21K Pretraining for the Masses,” pp. 1–20, 2021, [Online]. Available: http://arxiv.org/abs/2104.10972.
