Study on lightweight plasma recognition algorithm based on depth image perception_Journal of Biomedical Engineering

Authors：

ZHANG Hanwen ¹ , SUN Yu ^1,2 , JIANG Hao ¹ , HU Jintian ¹ ,  LUO Gangyin ¹ ,  LI Dong ¹ , CAO Weijuan ³ , QIU Xiang ³

1. Engineering Laboratory of Advanced In Vitro Diagnostic Technology Chinese Academy of Sciences, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Science, Suzhou, Jiangsu 215163, P. R. China;
2. College of Electrical and Automation Engineering, Nanjing Normal University, Nanjing 210023, P. R. China;
3. Suzhou Blood Center, Suzhou, Jiangsu 215006, P. R. China;

Corresponding author：

LUO Gangyin, Email: luogy@sibet.ac.cn

Keywords：

Omni-dimensional dynamic convolution; Pooling separable kernel attention; Residual bifusion feature pyramid network; Re-parameterization convolution

DOI：

10.7507/1001-5515.202404064

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

In the clinical stage, suspected hemolytic plasma may cause hemolysis illness, manifesting as symptoms such as heart failure, severe anemia, etc. Applying a deep learning method to plasma images significantly improves recognition accuracy, so that this paper proposes a plasma quality detection model based on improved “You Only Look Once” 5th version (YOLOv5). Then the model presented in this paper and the evaluation system ‌were introduced‌ into the plasma datasets, and ‌the average accuracy of the final classification reached 98.7%‌. The results of this paper's experiment were obtained through the combination of several key algorithm modules including‌ omni-dimensional dynamic convolution, pooling with separable kernel attention, residual bi-fusion feature pyramid network, ‌and‌ re-parameterization convolution. The method of this paper‌ obtains the feature information of spatial mapping efficiently, and enhances the average recognition accuracy of plasma quality detection. This paper presents a high-efficiency detection method for plasma images, aiming to provide a practical approach to prevent hemolysis illnesses caused by external factors.

Citation： ZHANG Hanwen, SUN Yu, JIANG Hao, HU Jintian, LUO Gangyin, LI Dong, CAO Weijuan, QIU Xiang. Study on lightweight plasma recognition algorithm based on depth image perception. Journal of Biomedical Engineering, 2025, 42(1): 123-131. doi: 10.7507/1001-5515.202404064 Copy

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.	Redmon J, Divvala S, Girshick R, et al. You only look once: unified, real-time object detection//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Dallas: IEEE, 2016: 779-788.
16.
17.	Al Ameri M, Memon Q. Real-time object tracking with YOLOv5 and recurrent network//2024 7th International Conference on Electronics, Communications, and Control Engineering (ICECC), Kuala Lumpur, Malaysia: IEEE, 2024: 28-32.
18.
19.	Yang B, Bender G, Le Q V, et al. Condconv: conditionally parameterized convolutions for efficient inference//The 33rd International Conference on Neural Information Processing Systems, Vancouver, Canada: NeurIPS, 2019: 1307-1318.
20.	Chen Y, Dai X, Liu M, et al. Dynamic convolution: attention over convolution kernels//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle: IEEE, 2020: 11027-11036.
21.	Li C, Zhou A, Yao A. Omni-dimensional dynamic convolution//The Tenth International Conference on Learning Representations (ICLR 2022), 2022, arXiv: 2209.07947.
22.
23.	Chen P Y, Hsieh J W, Wang C Y, et al. Residual bi-fusion feature pyramid network for accurate single-shot object detection. arXiv preprint, 2019, arXiv: 1911.12051.
24.	Ding X, Zhang X, Ma N, et al. RepVgg: making VGG-style ConvNets great again//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville: IEEE, 2021: 13728-13737.
25.

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15. Redmon J, Divvala S, Girshick R, et al. You only look once: unified, real-time object detection//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Dallas: IEEE, 2016: 779-788.
16.
17. Al Ameri M, Memon Q. Real-time object tracking with YOLOv5 and recurrent network//2024 7th International Conference on Electronics, Communications, and Control Engineering (ICECC), Kuala Lumpur, Malaysia: IEEE, 2024: 28-32.
18.
19. Yang B, Bender G, Le Q V, et al. Condconv: conditionally parameterized convolutions for efficient inference//The 33rd International Conference on Neural Information Processing Systems, Vancouver, Canada: NeurIPS, 2019: 1307-1318.
20. Chen Y, Dai X, Liu M, et al. Dynamic convolution: attention over convolution kernels//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle: IEEE, 2020: 11027-11036.
21. Li C, Zhou A, Yao A. Omni-dimensional dynamic convolution//The Tenth International Conference on Learning Representations (ICLR 2022), 2022, arXiv: 2209.07947.
22.
23. Chen P Y, Hsieh J W, Wang C Y, et al. Residual bi-fusion feature pyramid network for accurate single-shot object detection. arXiv preprint, 2019, arXiv: 1911.12051.
24. Ding X, Zhang X, Ma N, et al. RepVgg: making VGG-style ConvNets great again//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville: IEEE, 2021: 13728-13737.
25.

Journal of Biomedical Engineering

Study on lightweight plasma recognition algorithm based on depth image perception

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content