Automated detection of multi-class urinary sediment particles: An accurate deep learning approach

• Autorzy: Lyu He , Xu Fanxin , Jin Tao , Zheng Siyi , Zhou Chenchen , Cao Yang , Luo Bin , Huang Qinzhen , Xiang Wei , Li Dong

Identyfikatory

DOI

10.1016/j.bbe.2023.09.003

• Języki publikacji: EN

Abstrakty

EN

Urine microscopy is an essential diagnostic tool for kidney and urinary tract diseases, with automated analysis of urinary sediment particles improving diagnostic efficiency. However, some urinary sediment particles remain challenging to identify due to individual variations, blurred boundaries, and unbalanced samples. This research aims to mitigate the adverse effects of urine sediment particles while improving multi-class detection performance. We proposed an innovative model based on improved YOLOX for detecting urine sediment particles (YUS-Net). The combination of urine sediment data augmentation and overall pre-trained weights enhances model optimization potential. Furthermore, we incorporate the attention module into the critical feature transfer path and employ a novel loss function, Varifocal loss, to facilitate the extraction of discriminative features, which assists in the identification of densely distributed small objects. Based on the USE dataset, YUS-Net achieves the mean Average Precision (mAP) of 96.07%, 99.35% average precision, and 96.77% average recall, with a latency of 26.13 ms per image. The specific metrics for each category are as follows: cast: 99.66% AP; cryst: 100% AP; epith: 92.31% AP; epithn: 100% AP; eryth: 92.31% AP; leuko: 99.90% AP; mycete: 99.96% AP. With a practical network structure, YUS-Net achieved efficient, accurate, end-to-end urinary sediment particle detection. The model takes native high-resolution images as input without additional steps. Finally, a data augmentation strategy appropriate for the urinary microscopic image domain is established, which provides a novel approach for applying other methods in urine microscopic images.

Słowa kluczowe

EN

urinary sediment; cell detection; YOLOX; data augmention; attention mechanism; varifocal loss

PL

osad moczu; wykrywanie komórek; YOLOX; mechanizm uwagi

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2023
• Tom: Vol. 43, no. 4
• Strony: 672--683
• Opis fizyczny: Bibliogr. 48 poz., rys., tab.

Twórcy

autor

Lyu He

• Key Laboratory of Electronic and Information Engineering, State Ethnic Affairs Commission (Southwest Minzu University), Chengdu, China

autor

Xu Fanxin

• Key Laboratory of Electronic and Information Engineering, State Ethnic Affairs Commission (Southwest Minzu University), Chengdu, China
• Chongqing Key Laboratory of Computational Intelligence, Chongqing University of Posts and Telecommunications, Chongqing, China

autor

Jin Tao

• Key Laboratory of Electronic and Information Engineering, State Ethnic Affairs Commission (Southwest Minzu University), Chengdu, China

autor

Zheng Siyi

• Key Laboratory of Electronic and Information Engineering, State Ethnic Affairs Commission (Southwest Minzu University), Chengdu, China

autor

Zhou Chenchen

• Key Laboratory of Electronic and Information Engineering, State Ethnic Affairs Commission (Southwest Minzu University), Chengdu, China

autor

Cao Yang

• Key Laboratory of Electronic and Information Engineering, State Ethnic Affairs Commission (Southwest Minzu University), Chengdu, China

autor

Luo Bin

• Sichuan Huhui Software CO., LTD, Sichuan, China

autor

Huang Qinzhen

• Key Laboratory of Electronic and Information Engineering, State Ethnic Affairs Commission (Southwest Minzu University), Chengdu, China

autor

Xiang Wei

• Key Laboratory of Electronic and Information Engineering, State Ethnic Affairs Commission (Southwest Minzu University), Chengdu, China

autor

Li Dong

• Harbor-UCLA Medical Center, David Geffen School of Medicine at UCLA, Los Angeles, USA

Bibliografia

• [1] Dirks J, Remuzzi G, Horton S, Schieppati A, Rizvi SAH. Diseases of the Kidney and the Urinary System. Disease control priorities in developing countries. The International Bank for Reconstruction and Development/The World Bank; 2006.
• [2] Flores-Mireles AL, Walker JN, Caparon M, Hultgren SJ. Urinary tract infections: Epidemiology, mechanisms of infection and treatment options. Nat Rev Microbiol 2015;13:269-84. https:// doi.org/10.1038/nrmicro3432.
• [3] Li PKT, Burdmann EA, Mehta RL, World Kidney Day Steering Committee. Acute kidney injury: global health alert. Arab J Nephrol Transplant 2013;2013(6):75-81. https://doi.org/ 10.5812/nephropathol.10449.
• [4] Bagshaw SM, Gibney RTN. Acute kidney injury: clinical value of urine microscopy in acute kidney injury. Nat Rev Nephrol 2009;5:185-6. https://doi.org/10.1038/nrneph.2009.23.
• [5] Ostermann M, Joannidis M. Acute kidney injury 2016: diagnosis and diagnostic workup. Crit Care 2016;20:1-13. https://doi.org/10.1186/s13054-016-1478-z.
• [6] Akben SB. Early stage of chronic kidney disease by using statistical evaluation of the previous measurement results. Biocybernet Biomed Eng 2016;36:626-31. https://doi.org/ 10.1016/J.BBE.2016.08.004.
• [7] Suhail K, Brindha D. A review on various methods for recognition of urine particles using digital microscopic images of urine sediments. Biomed Signal Process Control 2021;68. https://doi.org/10.1016/j.bspc.2021.102806 102806.
• [8] Shen ML, Zhang R. Urine sediment recognition method based on SVM and AdaBoost. In: Proceedings – 2009 international conference on computational intelligence and software engineering, CiSE 2009. https://doi.org/10.1109/ CISE.2009.5365881.
• [9] Zhou Y, Zhou H. Automatic classification and recognition of particles in urinary sediment images. Lect Notes Electr Eng 2012;107LNEE:1071-8. https://doi.org/10.1007/978-94-007- 1839-5_116.
• [10] Sun Q, Yang S, Sun C, Yang W. An automatic method for red blood cells detection in urine sediment micrograph. In: Proceedings – 2018 33rd youth academic annual conference of Chinese association of automation, YAC 2018. Institute of Electrical and Electronics Engineers Inc.; 2018. p. 241-5. https://doi.org/10.1109/YAC.2018.8406379.
• [11] Sun Q, Yang S, Sun C, Yang W. Exploiting aggregate channel features for urine sediment detection. Multimed Tools Appl 2019;78:23883-95. https://doi.org/10.1007/s11042-018-6241-9.
• [12] Dela Cruz JC, Garcia RG, Avilledo MID, Buera JCM, Chan RVS, Espana PGT. Microscopic image analysis and counting of red blood cells and white blood cells in a urine sample. ACM Int Conf Proc Ser 2019:113-8. https://doi.org/10.1145/ 3326172.3326185.
• [13] Tangsuksant W, Pintavirooj C, Taertulakarn S, Daochai S. Development algorithm to count blood cells in urine sediment using ANN and Hough Transform. In: BMEiCON 2013-6th biomedical engineering international conference. https://doi.org/10.1109/BMEICON.2013.6687725.
• [14] Zhang X, Chen G, Saruta K, Terata Y. Detection and classification of RBCs and WBCs in urine analysis with deep network. In: ACHI 2018, the eleventh international conference on advances in computer-human interactions. p. 194-8.
• [15] Zhang X, Jiang L, Yang D, Yan J, Lu X. Urine sediment recognition method based on multi-view deep residual learning in microscopic image. J Med Syst 2019;43:1-10. https://doi.org/10.1007/s10916-020-01558-x.
• [16] Liu W, Li W, Gong W. Ensemble of fine-tuned convolutional neural networks for urine sediment microscopic image classification. IET Comput Vis 2020;14:18-25. https://doi.org/ 10.1049/IET-CVI.2018.5829.
• [17] Ji Q, Li X, Qu Z, Dai C. Research on urine sediment images recognition based on deep learning. IEEE Access 2019;7:166711-20. https://doi.org/10.1109/ACCESS.2019.2953775.
• [18] Li T, Jin D, Du C, Cao X, Chen H, Yan J, et al. The image-based analysis and classification of urine sediments using a LeNet5 neural network. Comput Methods Biomech Biomed Eng: Imag Visualiz 2020;8:109-14. https://doi.org/10.1080/ 21681163.2019.1608307.
• [19] Khalid ZM, Hawezi RS, Amin SRM, Urine Sediment Analysis by Using Convolution Neural Network. 8th IEC 2022 – International engineering conference: towards engineering innovations and sustainability 2022;2022:173-8. https://doi. org/10.1109/IEC54822.2022.9807482.
• [20] Yildirim M, Bingol H, Cengil E, Aslan S, Baykara M. Automatic classification of particles in the urine sediment test with the developed artificial intelligence-based hybrid model. Diagnostics 2023;13:1-16. https://doi.org/ 10.3390/diagnostics13071299.
• [21] Li K, Li M, Wu Y, Li X, Zhou X. An accurate urine erythrocytes detection model coupled faster RCNN with VggNet. In: ACM international conference proceeding series, association for computing machinery. p. 224-8. https://doi.org/10.1145/ 3433996.3434037.
• [22] Dong S, Zhang S, Jiao L, Wang Q. Automatic urinary sediments visible component detection based on improved YOLO algorithm. In: Proceedings – 2020 international conference on computer vision, image and deep learning, CVIDL 2020. Institute of Electrical and Electronics Engineers Inc.; 2020. p. 485-90. https://doi.org/10.1109/ CVIDL51233.2020.00-45.
• [23] Liang Y, Kang R, Lian C, Mao Y. An end-to-end system for automatic urinary particle recognition with convolutional neural network. J Med Syst 2018;42. https://doi.org/10.1007/s10916-018-1014-6.
• [24] Liang Y, Tang Z, Yan M, Liu J. Object detection based on deep learning for urine sediment examination. Biocybernet Biomed Eng 2018;38:661-70. https://doi.org/10.1016/j. Bbe.2018.05.004.
• [25] Yan M, Liu Q, Yin Z, Wang D, Liang Y. A bidirectional context propagation network for urine sediment particle detection in microscopic images. In: ICASSP 2020-2020 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE; 2020. p. 981-5.
• [26] Avci D, Sert E, Dogantekin E, Yildirim O, Tadeusiewicz R, Plawiak P. A new super resolution Faster R-CNN model based detection and classification of urine sediments. Biocybernet Biomed Eng 2023;43:58-68. https://doi.org/10.1016/J. BBE.2022.12.001.
• [27] Li Q, Yu Z, Qi T, Zheng L, Qi S, He Z, et al. Inspection of visible components in urine based on deep learning. Med Phys 2020;47:2937-49. https://doi.org/10.1002/mp.14118.
• [28] Chen Z, Hu R, Chen F, Fan H, Ching FY, Li Z, et al. An efficient particle YOLO detector for urine sediment detection. Lect Notes Comput Sci (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2023;13657 LNCS:294-308. https://doi.org/10.1007/978-3-031- 20102-8_23.
• [29] Nagai T, Onodera O, Okuda S. Deep learning classification of urinary sediment crystals with optimal parameter tuning. Sci Rep 2022;12:1-9. https://doi.org/10.1038/S41598-022-25385-X/MEDIAOBJECTS/41598_2022_25385_MOESM1_ESM.XLSX.
• [30] Cavanaugh C, Perazella MA. Urine sediment examination in the diagnosis and management of kidney disease: Core curriculum 2019. Am J Kidney Dis 2019;73:258-72. https://doi. org/10.1053/j.ajkd.2018.07.012.
• [31] Xu F, Li X, Yang H, Wang Y, Xiang W. TE-YOLOF: Tiny and efficient YOLOF for blood cell detection. Biomed Signal Process Control 2022:73. https://doi.org/10.1016/j.bspc.2021.103416.
• [32] Ge Z, Liu S, Wang F, Li Z, Sun J. YOLOX: Exceeding YOLO Series in 2021; 2021. p. 1-7.
• [33] Wang CY, Mark Liao HY, Wu YH, Chen PY, Hsieh JW, Yeh IH. CSPNet: A new backbone that can enhance learning capability of CNN. IEEE computer society conference on computer vision and pattern recognition workshops 2020, 2020-June. p. 1571-80. https://doi.org/10.1109/CVPRW50498.2020.00203.
• [34] Liu S, Qi L, Qin H, Shi J, Jia J. Path aggregation network for instance segmentation. In: Proceedings of the IEEE computer society conference on computer vision and pattern recognition. IEEE Computer Society; 2018. p. 8759-68. https:// doi.org/10.1109/CVPR.2018.00913.
• [35] Bochkovskiy A, Wang C-Y, Liao H-YM. YOLOv4: Optimal speed and accuracy of object detection; 2020.
• [36] Zhang H, Cisse M, Dauphin YN, Lopez-Paz D. MixUp: Beyond empirical risk minimization. 6th International conference on learning representations, ICLR 2018 – Conference track proceedings, international conference on learning representations. ICLR; 2018.
• [37] Lin TY, Maire M, Belongie S, Hays J, Perona P, Ramanan D, et al. Microsoft COCO: Common objects in context. Lect Notes Comput Sci (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2014;8693 LNCS:740-55. https://doi.org/10.1007/978-3-319-10602-1_48/COVER.
• [38] Woo S, Park J, Lee JY, Kweon IS. CBAM: Convolutional block attention module. Lect Notes Comput Sci (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2018;11211 LNCS:3-19. https://doi. Org/10.1007/978-3-030-01234-2_1.
• [39] Zhang H, Wang Y, Dayoub F, Sünderhauf N. VarifocalNet: An IoU-aware dense object detector. In: Proceedings of the IEEE computer society conference on computer vision and pattern recognition. p. 8510-9. https://doi.org/10.1109/ CVPR46437.2021.00841.
• [40] Everingham M, Van Gool L, Williams CKI, Winn J, Zisserman A. The pascal visual object classes (VOC) challenge. Int J Comput Vis 2010;88:303-38. https://doi.org/10.1007/s11263- 009-0275-4.
• [41] Ren S, He K, Girshick R, Sun J. Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell 2017;39:1137-49. https://doi. org/10.1109/TPAMI.2016.2577031.
• [42] Singh B, Najibi M, Davis LS. Sniper: Efficient multi-scale training. Advances in neural information processing systems 2018;2018-December:9310-20.
• [43] Lin TY, Dollár P, Girshick R, He K, Hariharan B, Belongie S. Feature pyramid networks for object detection. Proceedings – 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, 2017. Janua: Institute of Electrical and Electronics Engineers Inc.. p. 936-44. https://doi.org/ 10.1109/CVPR.2017.106.
• [44] Sun P, Zhang R, Jiang Y, Kong T, Xu C, Zhan W, et al. Sparse RCNN: End-to-end object detection with learnable proposals. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. p. 14449-58. https://doi.org/10.1109/CVPR46437.2021.01422.
• [45] Zhu X, Su W, Lu L, Li B, Wang X, Dai J, et al. Deformable DETR: Deformable transformers for end-to-end object detection; 2020.
• [46] Lin TY, Goyal P, Girshick R, He K, Dollar P. Focal loss for dense object detection. IEEE Trans Pattern Anal Mach Intell 2020;42:318-27. https://doi.org/10.1109/TPAMI.2018.2858826.
• [47] ultralytics/yolov5: v6.2 – YOLOv5 Classification Models, Apple M1, Reproducibility, ClearML and Deci.ai integrations; 2022. https://doi.org/10.5281/zenodo.7002879.
• [48] Chen K, Wang J, Pang J, Cao Y, Xiong Y, Li X, et al. MMDetection: Open MMLab Detection Toolbox and Benchmark, 2019.