SMLS-YOLO: an extremely lightweight pathological myopia instance segmentation method

被引:2
作者
Xie, Hanfei [1 ,2 ]
Yuan, Baoxi [1 ,2 ]
Hu, Chengyu [1 ,2 ]
Gao, Yujie [1 ]
Wang, Feng [1 ]
Wang, Yuqian [3 ]
Wang, Chunlan [1 ]
Chu, Peng [3 ]
机构
[1] Xijing Univ, Sch Elect Informat, Xian, Peoples R China
[2] Xijing Univ, Xian Key Lab High Precis Ind Intelligent Vis Measu, Xian, Peoples R China
[3] Xijing Univ, Grad Off, Xian, Peoples R China
基金
中国国家社会科学基金;
关键词
pathological myopia; SMLS-YOLO; instance segmentation; lightweight; image feature extraction; OPTIC DISC; EXTRACTION; BOUNDARY;
D O I
10.3389/fnins.2024.1471089
中图分类号
Q189 [神经科学];
学科分类号
071006 ;
摘要
Pathological myopia is a major cause of blindness among people under 50 years old and can result in severe vision loss in extreme cases. Currently, its detection primarily relies on manual methods, which are slow and heavily dependent on the expertise of physicians, making them impractical for large-scale screening. To tackle these challenges, we propose SMLS-YOLO, an instance segmentation method based on YOLOv8n-seg. Designed for efficiency in large-scale screenings, SMLS-YOLO employs an extremely lightweight model. First, StarNet is introduced as the backbone of SMLS-YOLO to extract image features. Subsequently, the StarBlock from StarNet is utilized to enhance the C2f, resulting in the creation of the C2f-Star feature extraction module. Furthermore, shared convolution and scale reduction strategies are employed to optimize the segmentation head for a more lightweight design. Lastly, the model incorporates the Multi-Head Self-Attention (MHSA) mechanism following the backbone to further refine the feature extraction process. Experimental results on the pathological myopia dataset demonstrate that SMLS-YOLO outperforms the baseline YOLOv8n-seg by reducing model parameters by 46.9%, increasing Box mAP@0.5 by 2.4%, and enhancing Mask mAP@0.5 by 4%. Furthermore, when compared to other advanced instance segmentation and semantic segmentation algorithms, SMLS-YOLO also maintains a leading position, suggesting that SMLS-YOLO has promising applications in the segmentation of pathological myopia images.
引用
收藏
页数:15
相关论文
共 26 条
[1]   Two-Stage Mask-RCNN Approach for Detecting and Segmenting the Optic Nerve Head, Optic Disc, and Optic Cup in Fundus Images [J].
Almubarak, Haidar ;
Bazi, Yakoub ;
Alajlan, Naif .
APPLIED SCIENCES-BASEL, 2020, 10 (11)
[2]   Detecting the Optic Disc Boundary in Digital Fundus Images Using Morphological, Edge Detection, and Feature Extraction Techniques [J].
Aquino, Arturo ;
Emilio Gegundez-Arias, Manuel ;
Marin, Diego .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2010, 29 (11) :1860-1869
[3]   Myopia [J].
Baird, Paul N. ;
Saw, Seang-Mei ;
Lanca, Carla ;
Guggenheim, Jeremy A. ;
Smith III, Earl L. ;
Zhou, Xiangtian ;
Matsui, Kyoko-Ohno ;
Wu, Pei-Chang ;
Sankaridurg, Padmaja ;
Chia, Audrey ;
Rosman, Mohamad ;
Lamoureux, Ecosse L. ;
Man, Ryan ;
He, Mingguang .
NATURE REVIEWS DISEASE PRIMERS, 2020, 6 (01)
[4]   YOLACT Real-time Instance Segmentation [J].
Bolya, Daniel ;
Zhou, Chong ;
Xiao, Fanyi ;
Lee, Yong Jae .
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, :9156-9165
[5]   Joint optic disc and cup boundary extraction from monocular fundus images [J].
Chakravarty, Arunava ;
Sivaswamy, Jayanthi .
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2017, 147 :51-61
[6]   Application of Artificial Intelligence and Deep Learning for Choroid Segmentation in Myopia [J].
Chen, Hung-Ju ;
Huang, Yu-Len ;
Tse, Siu-Lun ;
Hsia, Wei-Ping ;
Hsiao, Chung-Hao ;
Wang, Yang ;
Chang, Chia-Jen .
TRANSLATIONAL VISION SCIENCE & TECHNOLOGY, 2022, 11 (02)
[7]  
Chen LC, 2018, Arxiv, DOI arXiv:1802.02611
[8]   THE MYOPIA BOOM [J].
Dolgin, Elie .
NATURE, 2015, 519 (7543) :276-278
[9]  
GeethaRamani R., 2014, 2014 IEEE International Conference on Computational Intelligence and Computing Research [Internet], P1
[10]   A Survey on Vision Transformer [J].
Han, Kai ;
Wang, Yunhe ;
Chen, Hanting ;
Chen, Xinghao ;
Guo, Jianyuan ;
Liu, Zhenhua ;
Tang, Yehui ;
Xiao, An ;
Xu, Chunjing ;
Xu, Yixing ;
Yang, Zhaohui ;
Zhang, Yiman ;
Tao, Dacheng .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (01) :87-110