• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

精子 YOLOv8E-TrackEVD：一种新型的精子检测与跟踪方法。

Sperm YOLOv8E-TrackEVD: A Novel Approach for Sperm Detection and Tracking.

机构信息

College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 200234, China.

出版信息

Sensors (Basel). 2024 May 28;24(11):3493. doi: 10.3390/s24113493.

DOI:10.3390/s24113493
PMID:38894284
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11175353/
Abstract

Male infertility is a global health issue, with 40-50% attributed to sperm abnormalities. The subjectivity and irreproducibility of existing detection methods pose challenges to sperm assessment, making the design of automated semen analysis algorithms crucial for enhancing the reliability of sperm evaluations. This paper proposes a comprehensive sperm tracking algorithm (Sperm YOLOv8E-TrackEVD) that combines an enhanced YOLOv8 small object detection algorithm (SpermYOLOv8-E) with an improved DeepOCSORT tracking algorithm (SpermTrack-EVD) to detect human sperm in a microscopic field of view and track healthy sperm in a sample in a short period effectively. Firstly, we trained the improved YOLOv8 model on the VISEM-Tracking dataset for accurate sperm detection. To enhance the detection of small sperm objects, we introduced an attention mechanism, added a small object detection layer, and integrated the SPDConv and Detect_DyHead modules. Furthermore, we used a new distance metric method and chose IoU loss calculation. Ultimately, we achieved a 1.3% increase in precision, a 1.4% increase in recall rate, and a 2.0% improvement in mAP@0.5:0.95. We applied SpermYOLOv8-E combined with SpermTrack-EVD for sperm tracking. On the VISEM-Tracking dataset, we achieved 74.303% HOTA and 71.167% MOTA. These results show the effectiveness of the designed Sperm YOLOv8E-TrackEVD approach in sperm tracking scenarios.

摘要

男性不育是一个全球性的健康问题，其中 40-50%归因于精子异常。现有的检测方法存在主观性和不可重复性，这给精子评估带来了挑战，因此设计自动化精液分析算法对于提高精子评估的可靠性至关重要。本文提出了一种全面的精子跟踪算法（Sperm YOLOv8E-TrackEVD），该算法将增强的 YOLOv8 小目标检测算法（SpermYOLOv8-E）与改进的 DeepOCSORT 跟踪算法（SpermTrack-EVD）相结合，用于在显微镜视野中检测人类精子，并在短时间内有效地跟踪样本中的健康精子。首先，我们在 VISEM-Tracking 数据集上训练改进的 YOLOv8 模型，以实现对精子的准确检测。为了增强对小精子目标的检测能力，我们引入了注意力机制，添加了小目标检测层，并集成了 SPDConv 和 Detect_DyHead 模块。此外，我们使用了新的距离度量方法，并选择了 IoU 损失计算。最终，我们的方法在精度上提高了 1.3%，召回率提高了 1.4%，mAP@0.5:0.95 提高了 2.0%。我们将 SpermYOLOv8-E 与 SpermTrack-EVD 结合应用于精子跟踪。在 VISEM-Tracking 数据集上，我们实现了 74.303%的 HOTA 和 71.167%的 MOTA。这些结果表明，所设计的 Sperm YOLOv8E-TrackEVD 方法在精子跟踪场景中是有效的。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/5eaf857d11c9/sensors-24-03493-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/8a206cfa8e4c/sensors-24-03493-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/f7c266edf9a0/sensors-24-03493-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/0ad6559497a7/sensors-24-03493-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/6b0c983509f5/sensors-24-03493-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/34df5e53c4c9/sensors-24-03493-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/a17d52045cf6/sensors-24-03493-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/48c3cd5f6ead/sensors-24-03493-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/4a36e2366802/sensors-24-03493-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/d3c9db026f80/sensors-24-03493-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/c579f5fa53c4/sensors-24-03493-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/0ad4e55d1ad6/sensors-24-03493-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/651c8e478f41/sensors-24-03493-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/6ac972c7f00c/sensors-24-03493-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/cdf43e54e9e5/sensors-24-03493-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/5eaf857d11c9/sensors-24-03493-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/8a206cfa8e4c/sensors-24-03493-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/f7c266edf9a0/sensors-24-03493-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/0ad6559497a7/sensors-24-03493-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/6b0c983509f5/sensors-24-03493-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/34df5e53c4c9/sensors-24-03493-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/a17d52045cf6/sensors-24-03493-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/48c3cd5f6ead/sensors-24-03493-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/4a36e2366802/sensors-24-03493-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/d3c9db026f80/sensors-24-03493-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/c579f5fa53c4/sensors-24-03493-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/0ad4e55d1ad6/sensors-24-03493-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/651c8e478f41/sensors-24-03493-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/6ac972c7f00c/sensors-24-03493-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/cdf43e54e9e5/sensors-24-03493-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/bf66/11175353/5eaf857d11c9/sensors-24-03493-g015.jpg

相似文献

1
Sperm YOLOv8E-TrackEVD: A Novel Approach for Sperm Detection and Tracking.精子 YOLOv8E-TrackEVD：一种新型的精子检测与跟踪方法。
Sensors (Basel). 2024 May 28;24(11):3493. doi: 10.3390/s24113493.
2
An assessment tool for computer-assisted semen analysis (CASA) algorithms.计算机辅助精液分析（CASA）算法评估工具。
Sci Rep. 2022 Oct 7;12(1):16830. doi: 10.1038/s41598-022-20943-9.
3
Multi-Object Pedestrian Tracking Using Improved YOLOv8 and OC-SORT.基于改进的YOLOv8和OC-SORT的多目标行人跟踪
Sensors (Basel). 2023 Oct 13;23(20):8439. doi: 10.3390/s23208439.
4
Study on Sperm-Cell Detection Using YOLOv5 Architecture with Labaled Dataset.基于标注数据集的使用 YOLOv5 架构的精子细胞检测研究。
Genes (Basel). 2023 Feb 9;14(2):451. doi: 10.3390/genes14020451.
5
Automated motility and morphology measurement of live spermatozoa.活精子的自动运动和形态测量。
Andrology. 2021 Jul;9(4):1205-1213. doi: 10.1111/andr.13002. Epub 2021 Apr 2.
6
VISEM-Tracking, a human spermatozoa tracking dataset.VISEM-Tracking，一个人类精子追踪数据集。
Sci Data. 2023 May 9;10(1):260. doi: 10.1038/s41597-023-02173-4.
7
Faster region convolutional neural network and semen tracking algorithm for sperm analysis.用于精子分析的快速区域卷积神经网络和精液追踪算法
Comput Methods Programs Biomed. 2021 Mar;200:105918. doi: 10.1016/j.cmpb.2020.105918. Epub 2021 Jan 9.
8
Multi-Target Tracking of Human Spermatozoa in Phase-Contrast Microscopy Image Sequences using a Hybrid Dynamic Bayesian Network.基于混合动态贝叶斯网络的相差显微镜图像序列中人精子的多目标跟踪。
Sci Rep. 2018 Mar 22;8(1):5068. doi: 10.1038/s41598-018-23435-x.
9
Development of computer-assisted sperm analysis plugin for analyzing sperm motion in microfluidic environments using Image-J.开发用于在微流控环境中使用Image-J分析精子运动的计算机辅助精子分析插件。
Theriogenology. 2015 Nov;84(8):1367-77. doi: 10.1016/j.theriogenology.2015.07.021. Epub 2015 Jul 26.
10
Validation of a novel computer-assisted sperm analysis (CASA) system using multitarget-tracking algorithms.使用多目标跟踪算法验证新型计算机辅助精液分析（CASA）系统。
Fertil Steril. 2010 Apr;93(6):1911-20. doi: 10.1016/j.fertnstert.2008.12.064. Epub 2009 Feb 6.

引用本文的文献

1
Deep feature engineering for accurate sperm morphology classification using CBAM-enhanced ResNet50.使用CBAM增强的ResNet50进行精确精子形态分类的深度特征工程
PLoS One. 2025 Sep 10;20(9):e0330914. doi: 10.1371/journal.pone.0330914. eCollection 2025.
2
Deep Learning Models for Multi-Part Morphological Segmentation and Evaluation of Live Unstained Human Sperm.用于活体未染色人类精子多部分形态学分割与评估的深度学习模型
Sensors (Basel). 2025 May 14;25(10):3093. doi: 10.3390/s25103093.

本文引用的文献

1
A Hand-Held Platform for Boar Sperm Viability Diagnosis Based on Smartphone.基于智能手机的猪精子活力诊断手持平台。
Biosensors (Basel). 2023 Nov 8;13(11):978. doi: 10.3390/bios13110978.
2
Sperm motility assessed by deep convolutional neural networks into WHO categories.基于深度卷积神经网络的精子活动力评估与世界卫生组织分类。
Sci Rep. 2023 Sep 7;13(1):14777. doi: 10.1038/s41598-023-41871-2.
3
Trends in sperm quality by computer-assisted sperm analysis of 49,189 men during 2015-2021 in a fertility center from China.2015 年至 2021 年期间在中国一家生育中心通过计算机辅助精液分析对 49189 名男性的精子质量趋势分析。
Front Endocrinol (Lausanne). 2023 Jul 12;14:1194455. doi: 10.3389/fendo.2023.1194455. eCollection 2023.
4
A Dual Architecture Fusion and AutoEncoder for Automatic Morphological Classification of Human Sperm.一种融合双架构和自动编码器的人类精子自动形态分类方法。
Sensors (Basel). 2023 Jul 22;23(14):6613. doi: 10.3390/s23146613.
5
Artificial Intelligence in Andrology: From Semen Analysis to Image Diagnostics.男科领域的人工智能：从精液分析到图像诊断
World J Mens Health. 2024 Jan;42(1):39-61. doi: 10.5534/wjmh.230050. Epub 2023 Jun 15.
6
VISEM-Tracking, a human spermatozoa tracking dataset.VISEM-Tracking，一个人类精子追踪数据集。
Sci Data. 2023 May 9;10(1):260. doi: 10.1038/s41597-023-02173-4.
7
Understanding the genetics of human infertility.理解人类不孕不育的遗传学。
Science. 2023 Apr 14;380(6641):158-163. doi: 10.1126/science.adf7760. Epub 2023 Apr 13.
8
Male factor infertility and assisted reproductive technologies: indications, minimum access criteria and outcomes.男性因素不孕与辅助生殖技术：适应证、微创准入标准与结局。
J Endocrinol Invest. 2023 Jun;46(6):1079-1085. doi: 10.1007/s40618-022-02000-4. Epub 2023 Jan 12.
9
Spatiotemporal trends in human semen quality.人类精液质量的时空变化趋势。
Nat Rev Urol. 2022 Oct;19(10):597-626. doi: 10.1038/s41585-022-00626-w. Epub 2022 Aug 17.
10
Automated identification of rare sperm becomes possible: Is sperm selection the next frontier in male infertility?罕见精子的自动识别成为可能：精子筛选会是男性不育领域的下一个前沿方向吗？
Fertil Steril. 2022 Jul;118(1):100. doi: 10.1016/j.fertnstert.2022.05.005.