Plasma detection model based on improved You Only Look Once version 5_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.

1.	徐忠, 邱颖婕, 杨剑豪. 血浆外观目测比较色板的制作和应用. 中国输血杂志, 2016, 29(6): 641-643.
2.	林思妮, 孙丽华. 标本溶血对检验结果产生的影响以及预防策略分析. 智慧健康, 2023, 9(3): 18-21.
3.	Larrán B, Miranda M, Herrero-Latorre C, et al. Influence of haemolysis on the mineral profile of cattle serum. Animals, 2021, 11(12): 3336.
4.	Lægreid I J, Wilson T, Næss K H, et al. Whole blood transfusion and paroxysmal nocturnal haemoglobinuria meet again: minor incompatibility, major trouble. Vox Sanguinis, 2022, 117(11): 1323-1326.
5.	Winter K M, Webb R G, Marks D C. Red cells manufactured from lipaemic whole blood donations: do they have higher haemolysis. Vox Sanguinis, 2022, 117(12): 1351-1359.
6.	庄冶. 血清标本发生溶血和脂血对生化检验结果的影响观察. 中国医药指南, 2020, 18(2): 112-113.
7.	张叶华. 血标本溶血的原因及急诊采血的护理研究进展. 中外医学研究, 2020, 18(35): 183-186.
8.	Hawkins R. Discrepancy between visual and spectrophotometric assessment of sample haemolysis. Annals of Clinical Biochemistry, 2002, 39(5): 521-522.
9.	Simundic A M, Nikolac N, Ivankovic V, et al. Comparison of visual vs. automated detection of lipemic, icteric and hemolyzed specimens: can we rely on a human eye?. Clinical Chemistry and Laboratory Medicine, 2009, 47(11): 1361-1365.
10.	Luksic A H, Nikolac Gabaj N, Miler M, et al. Visual assessment of hemolysis affects patient safety. Clinical Chemistry and Laboratory Medicine, 2018, 56(4): 574-581.
11.	栾庆玲, 隋馨, 高雅松, 等. 溶血试验方法在医疗器械领域的应用. 国际感染病学(电子版), 2020, 9(2): 20.
12.	Selcuk S Y, Yang X, Bai B, et al. Automated HER2 scoring in breast cancer images using deep learning and pyramid sampling. BME Front, 2024, 5: 0048.
13.	殷悦, 杨琦, 张省委, 等. 全自动生化分析仪检测溶血指数的临床应用. 标记免疫分析与临床, 2021, 28(8): 1415-1420.
14.	赵心宇, 顾益玲, 张丽红. 溶血标本生化项目替代性检验的可行性分析. 中国卫生检验杂志, 2022, 32(18): 2257-2260.
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.	Wang L, Ji W, Wang G, et al. Intelligent design and optimization of exercise equipment based on fusion algorithm of YOLOv5-ResNet 50. Alexandria Engineering Journal, 2024, 104(22): 710-722.
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.	Li C, Zhao G, Gu D, et al. Improved lightweight YOLOv5 using attention mechanism for satellite components recognition. IEEE Sensors Journal, 2023, 23(1): 514-526.
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.	Lau K W, Po L, Rehman Y A U. Large separable kernel attention: rethinking the large kernel attention design in CNN. Expert Systems with Applications, 2024, 236: 121352.
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.	Yang X, del Rey Castillo E, Zou Y, et al. UAV-deployed deep learning network for real-time multi-class damage detection using model quantization techniques. Automation in Construction, 2024, 159: 105254.

1. 徐忠, 邱颖婕, 杨剑豪. 血浆外观目测比较色板的制作和应用. 中国输血杂志, 2016, 29(6): 641-643.
2. 林思妮, 孙丽华. 标本溶血对检验结果产生的影响以及预防策略分析. 智慧健康, 2023, 9(3): 18-21.
3. Larrán B, Miranda M, Herrero-Latorre C, et al. Influence of haemolysis on the mineral profile of cattle serum. Animals, 2021, 11(12): 3336.
4. Lægreid I J, Wilson T, Næss K H, et al. Whole blood transfusion and paroxysmal nocturnal haemoglobinuria meet again: minor incompatibility, major trouble. Vox Sanguinis, 2022, 117(11): 1323-1326.
5. Winter K M, Webb R G, Marks D C. Red cells manufactured from lipaemic whole blood donations: do they have higher haemolysis. Vox Sanguinis, 2022, 117(12): 1351-1359.
6. 庄冶. 血清标本发生溶血和脂血对生化检验结果的影响观察. 中国医药指南, 2020, 18(2): 112-113.
7. 张叶华. 血标本溶血的原因及急诊采血的护理研究进展. 中外医学研究, 2020, 18(35): 183-186.
8. Hawkins R. Discrepancy between visual and spectrophotometric assessment of sample haemolysis. Annals of Clinical Biochemistry, 2002, 39(5): 521-522.
9. Simundic A M, Nikolac N, Ivankovic V, et al. Comparison of visual vs. automated detection of lipemic, icteric and hemolyzed specimens: can we rely on a human eye?. Clinical Chemistry and Laboratory Medicine, 2009, 47(11): 1361-1365.
10. Luksic A H, Nikolac Gabaj N, Miler M, et al. Visual assessment of hemolysis affects patient safety. Clinical Chemistry and Laboratory Medicine, 2018, 56(4): 574-581.
11. 栾庆玲, 隋馨, 高雅松, 等. 溶血试验方法在医疗器械领域的应用. 国际感染病学(电子版), 2020, 9(2): 20.
12. Selcuk S Y, Yang X, Bai B, et al. Automated HER2 scoring in breast cancer images using deep learning and pyramid sampling. BME Front, 2024, 5: 0048.
13. 殷悦, 杨琦, 张省委, 等. 全自动生化分析仪检测溶血指数的临床应用. 标记免疫分析与临床, 2021, 28(8): 1415-1420.
14. 赵心宇, 顾益玲, 张丽红. 溶血标本生化项目替代性检验的可行性分析. 中国卫生检验杂志, 2022, 32(18): 2257-2260.
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. Wang L, Ji W, Wang G, et al. Intelligent design and optimization of exercise equipment based on fusion algorithm of YOLOv5-ResNet 50. Alexandria Engineering Journal, 2024, 104(22): 710-722.
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. Li C, Zhao G, Gu D, et al. Improved lightweight YOLOv5 using attention mechanism for satellite components recognition. IEEE Sensors Journal, 2023, 23(1): 514-526.
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. Lau K W, Po L, Rehman Y A U. Large separable kernel attention: rethinking the large kernel attention design in CNN. Expert Systems with Applications, 2024, 236: 121352.
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. Yang X, del Rey Castillo E, Zou Y, et al. UAV-deployed deep learning network for real-time multi-class damage detection using model quantization techniques. Automation in Construction, 2024, 159: 105254.

Journal of Biomedical Engineering

Latest ArticlesPlasma detection model based on improved You Only Look Once version 5

Abstract Full text Figures/Tables Video References Cited by

Format

Content