| 35 | 0 | 158 |
| 下载次数 | 被引频次 | 阅读次数 |
精子形态学是男性生育力评估的关键指标。目前临床普遍采用的人工染色镜检法主要基于二维静态图像进行精子形态分析，无法全面反映精子真实三维结构，且制片与染色过程易造成形态改变；加之人工判读主观性较强，导致检测结果准确性与可靠性受限。而人工智能辅助未染色精子形态学评估（AIAUSMA）通过计算机视觉与深度学习算法，实现了对未染色精子形态的实时、无创、多维度分析。该技术突破了传统精子染色方法为基础评估手段的局限性。本专家建议结合临床需求与技术发展趋势，综合了国内外相关研究，在AIAUSMA的数据采集与标注、算法设计与性能评估、标准检测流程、质量控制标准、临床应用场景等方面给出了详细的建议，旨在促进AIAUSMA技术的标准化和临床实践应用，进一步提高精子形态学评估的准确性和可比性。
Abstract：Sperm morphology is a key indicator for evaluating male fertility.Currently,the conventional manual staining and microscopic analysis is widely used in clinical practice to analyze sperm morphology based mainly on two-dimensional static images,which cannot fully characterize the genuine three-dimensional structure of sperm.Moreover,smear preparation,fixation and staining procedures are prone to morphological alterations.In addition,high subjectivity in manual interpretation limits the accuracy and reliability of test results.Artificial intelligence-assisted unstained sperm morphological assessment(AIAUSMA)enables realtime,noninvasive and multidimensional analysis of unstained sperm morphology via computer vision and deep learning algorithms.This technology overcomes the limitations of traditional sperm assessment based on staining methods.Integrating clinical needs,technological advances and global evidence,this expert recommendation provides detailed proposals on data acquisition and annotation,algorithm development and validation,standardized workflows,quality control,and clinical indications for AIAUSMA.It intends to facilitate the standardized development and clinical implementation of AIAUSMA,and ultimately enhance the accuracy and comparability of sperm morphology evaluation.
[1]Lingappa HA,Govindashetty AM,Krishnamurthy A,et al.Quest for an ideal,simple and cost-effective stain for morphological assessment of sperms[J].J Clin Diagn Res,2015,9:EC01-EC04.
[2]World Health Organization.WHO laboratory manual for the examination and processing of human semen[M].6th ed.Geneva:World Health Organization,2021.
[3]Menkveld R.Clinical significance of the low normal sperm morphology value as proposed in the fifth edition of the WHO laboratory manual for the examination and processing of human semen[J].Asian J Androl,2010,12:47-58.
[4]Marnet B,Vieitez G,Milhet P,et al. Computer-assisted assessment of sperm morphology:comparison with conventional techniques[J].Int J Androl,2000,23:22-28.
[5]Ilardo C,Defort N,Gala A,et al. Retrospective study investigating the performance of the SQA-vision analyser compared with manual semen analysis[J].Scand J Clin Lab Invest,2024,84:373-378.
[6]Chaiya J,Vinayanuvattikhun N,Tanprasertkul C,et al.Effect of staining methods on human sperm morphometrics using HT CASA II[J].J Gynecol Obstet Hum Reprod,2022,51:102322.
[7]Yüzkat M,Ilhan HO,Aydin N.Multi-model CNN fusion for sperm morphology analysis[J].Comput Biol Med,2021,137:104790.
[8]Mortimer ST,van der Horst G,Mortimer D.The future of computer-aided sperm analysis[J].Asian J Androl,2015,17:545-553.
[9]Lu JC,Huang YF,LüNQ.Computer-aided sperm analysis:past,present and future[J].Andrologia,2014,46:329-338.
[10]Esteves SC,Zini A,Aziz N,et al.Critical appraisal of World Health Organization’s new reference values for human semen characteristics and effect on diagnosis and treatment of subfertile men[J].J Androl,2012,33:16-23.
[11]Boitrelle F,Albert M,Petit JM,et al.Small human sperm vacuoles observed under high magnification are pocket-like nuclear concavities linked to chromatin condensation failure[J].Reprod Biomed Online,2013,27:201-211.
[12]Cherouveim P,Velmahos C,Bormann CL.Artificial intelligence for sperm selection-a systematic review[J].Fertil Steril,2023,120:24-31.
[13]Yang H,Ma M,Chen X,et al.Multidimensional morphological analysis of live sperm based on multiple-target tracking[J].Comput Struct Biotechnol J,2024,24:176-184.
[14]Jaruenpunyasak J,Maneelert P,Nawae M,et al.Artificial intelligence model for the assessment of unstained live sperm morphology[J].Reprod Fertil,2025,6:e250014.
[15]Dai C,Zhang Z,Jahangiri S,et al.Automated motility and morphology measurement of live spermatozoa[J].Andrology,2021,9:1205-1213.
[16]Dai C,Shan G,Jahangiri S,et al.Staining-free,automated sperm analysis for in vitro fertilization lab use[J].J Urol,2022,208:1303-1312.
[17]Wang CY,Bochkovskiy A,Liao HYM.YOLOv7:trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[A].in:Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition[C],2023.
[18]Chen H.Blendmask:top-down meets bottom-up for instance segmentation[A].in:Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition[C],2020.
[19]Agarwal A,Sharma R,Gupta S,et al.Standardized laboratory procedures,quality control and quality assurance are key requirements for accurate semen analysis in the evaluation of infertile male[J].World J Mens Health,2022,40:52-65.
[20]Brito LFC.Quality assurance applied to semen analysis[J].Anim Reprod Sci,2024,270:107615.
[21]Panner Selvam MK,Moharana AK,Baskaran S,et al.Current updates on involvement of artificial intelligence and machine learning in semen analysis[J].Medicina(Kaunas),2024,60:279.
[22]Ghayda RA,Cannarella R,Calogero AE,et al. Artificial intelligence in andrology:from semen analysis to image diagnostics[J].World J Mens Health,2024,42:39-61.
[23]Qaderi K,Sharifipour F,Dabir M,et al.Artificial intelligence(AI)approaches to male infertility in IVF:a mapping review[J].Eur J Med Res,2025,30:246.
[24]Esteva A,Robicquet A,Ramsundar B,et al.A guide to deep learning in healthcare[J].Nat Med,2019,25:24-29.
[25]Li S,Yue R,Lu S,et al.Artificial intelligence and machine learning in acute respiratory distress syndrome management:recent advances[J].Front Med(Lausanne),2025,12:1597556.
[26]Kandel ME,Rubessa M, He YR,et al. Reproductive outcomes predicted by phase imaging with computational specificity of spermatozoon ultrastructure[J].Proc Natl Acad Sci U S A,2020,117:18302-18309.
[27]Tjoa E,Guan C.A survey on explainable artificial intelligence(XAI):toward medical XAI[J].IEEE Trans Neural Netw Learn Syst,2021,32:4793-4813.
[28]Kelly CJ,Karthikesalingam A,Suleyman M,et al. Key challenges for delivering clinical impact with artificial intelligence[J].BMC Med,2019,17:195.
[29]Topol EJ.High-performance medicine:the convergence of human and artificial intelligence[J].Nat Med,2019,25:44-56.
[30]Pelzman DL,Sandlow JI.Sperm morphology:Evaluating its clinical relevance in contemporary fertility practice[J].Reprod Med Biol,2024,23:e12594.
基本信息:
中图分类号:R698.2
引用信息:
[1]郭毅,马萌萌,卢文红,等.人工智能辅助未染色精子形态学评估技术与专家建议[J].生殖医学杂志,2026,35(07):837-845.
基金信息:
上海申康医院发展中心临床科技创新项目（SHDC12021115）; 同济大学附属妇产科医院院级临床研究培育项目成果转化专项（2025B07）
2026-02-11
2026
2026-04-25
2026-06-10
2026
1
2026-07-15
2026-07-15