{"title":"Artificial intelligence interpretation of touch print smear cytology of testicular specimen from patients with azoospermia.","authors":"Chen-Hao Hsu, Chun-Fu Yeh, I-Shen Huang, Wei-Jen Chen, Yu-Ching Peng, Cheng-Han Tsai, Mong-Chi Ko, Chun-Ping Su, Hann-Chyun Chen, Wei-Lin Wu, Tyng-Luh Liu, Kuang-Min Lee, Chiao-Hsuan Li, Ethan Tu, William J Huang","doi":"10.1007/s10815-024-03215-5","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>Identification of mature sperm at microdissection testicular sperm extraction (mTESE) is a crucial step of sperm retrieval to help patients with non-obstructive azoospermia (NOA) proceed to intracytoplasmic sperm injection. Touch print smear (TPS) cytology allows immediate interpretation and prompt sperm identification intraoperatively. In this study, we leverage machine learning (ML) to facilitate TPS reading and conquer the learning curve for new operators.</p><p><strong>Materials and methods: </strong>One hundred seventy-six microscopic TPS images from the testicular specimen of patients with azoospermia at Taipei Veterans General Hospital were retrospectively collected, including categories of Sertoli cell, primary spermatocytes, round spermatids, elongated spermatids, immature sperm, and mature sperm. Among them, 118 images were assigned as the training set and 29 images as the validation set. RetinaNet (Lin et al. in IEEE Trans Pattern Anal Mach Intell. 42:318-327, 2020), a one-stage detection framework, was adopted for cell detection. The performance was evaluated at the cell level with average precision (AP) and recall, and the precision-recall (PR) curve was displayed among an independent testing set that contains 29 images that aim to assess the model.</p><p><strong>Results: </strong>The training set consisted of 4772 annotated cells, including 1782 Sertoli cells, 314 primary spermatocytes, 443 round spermatids, 279 elongated spermatids, 504 immature sperm, and 1450 mature sperm. This study demonstrated the performance of each category and the overall AP and recall on the validation set, which were 80.47% and 96.69%. The overall AP and recall were 79.48% and 93.63% on the testing set, while increased to 85.29% and 93.80% once the post-meiotic cells were merged into one category.</p><p><strong>Conclusions: </strong>This study proposed an innovative approach that leveraged ML methods to facilitate the diagnosis of spermatogenesis at mTESE for patients with NOA. With the assistance of ML techniques, surgeons could determine the stages of spermatogenesis and provide timely histopathological diagnosis for infertile males.</p>","PeriodicalId":15246,"journal":{"name":"Journal of Assisted Reproduction and Genetics","volume":null,"pages":null},"PeriodicalIF":3.2000,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Assisted Reproduction and Genetics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s10815-024-03215-5","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"GENETICS & HEREDITY","Score":null,"Total":0}
引用次数: 0
Abstract
Purpose: Identification of mature sperm at microdissection testicular sperm extraction (mTESE) is a crucial step of sperm retrieval to help patients with non-obstructive azoospermia (NOA) proceed to intracytoplasmic sperm injection. Touch print smear (TPS) cytology allows immediate interpretation and prompt sperm identification intraoperatively. In this study, we leverage machine learning (ML) to facilitate TPS reading and conquer the learning curve for new operators.
Materials and methods: One hundred seventy-six microscopic TPS images from the testicular specimen of patients with azoospermia at Taipei Veterans General Hospital were retrospectively collected, including categories of Sertoli cell, primary spermatocytes, round spermatids, elongated spermatids, immature sperm, and mature sperm. Among them, 118 images were assigned as the training set and 29 images as the validation set. RetinaNet (Lin et al. in IEEE Trans Pattern Anal Mach Intell. 42:318-327, 2020), a one-stage detection framework, was adopted for cell detection. The performance was evaluated at the cell level with average precision (AP) and recall, and the precision-recall (PR) curve was displayed among an independent testing set that contains 29 images that aim to assess the model.
Results: The training set consisted of 4772 annotated cells, including 1782 Sertoli cells, 314 primary spermatocytes, 443 round spermatids, 279 elongated spermatids, 504 immature sperm, and 1450 mature sperm. This study demonstrated the performance of each category and the overall AP and recall on the validation set, which were 80.47% and 96.69%. The overall AP and recall were 79.48% and 93.63% on the testing set, while increased to 85.29% and 93.80% once the post-meiotic cells were merged into one category.
Conclusions: This study proposed an innovative approach that leveraged ML methods to facilitate the diagnosis of spermatogenesis at mTESE for patients with NOA. With the assistance of ML techniques, surgeons could determine the stages of spermatogenesis and provide timely histopathological diagnosis for infertile males.
期刊介绍:
The Journal of Assisted Reproduction and Genetics publishes cellular, molecular, genetic, and epigenetic discoveries advancing our understanding of the biology and underlying mechanisms from gametogenesis to offspring health. Special emphasis is placed on the practice and evolution of assisted reproduction technologies (ARTs) with reference to the diagnosis and management of diseases affecting fertility. Our goal is to educate our readership in the translation of basic and clinical discoveries made from human or relevant animal models to the safe and efficacious practice of human ARTs. The scientific rigor and ethical standards embraced by the JARG editorial team ensures a broad international base of expertise guiding the marriage of contemporary clinical research paradigms with basic science discovery. JARG publishes original papers, minireviews, case reports, and opinion pieces often combined into special topic issues that will educate clinicians and scientists with interests in the mechanisms of human development that bear on the treatment of infertility and emerging innovations in human ARTs. The guiding principles of male and female reproductive health impacting pre- and post-conceptional viability and developmental potential are emphasized within the purview of human reproductive health in current and future generations of our species.
The journal is published in cooperation with the American Society for Reproductive Medicine, an organization of more than 8,000 physicians, researchers, nurses, technicians and other professionals dedicated to advancing knowledge and expertise in reproductive biology.