Ran Yang, Fritzie T Celino-Brady, Jessica E M Dunleavy, Katinka A Vigh-Conrad, Georgia R Atkins, Rachel L Hvasta, Christopher R X Pombar, Alexander N Yatsenko, Kyle E Orwig, Moira K O'Bryan, Ana C Lima, Donald F Conrad
{"title":"SATINN v2:多实验室数据集成的小鼠睾丸组织学自动图像分析。","authors":"Ran Yang, Fritzie T Celino-Brady, Jessica E M Dunleavy, Katinka A Vigh-Conrad, Georgia R Atkins, Rachel L Hvasta, Christopher R X Pombar, Alexander N Yatsenko, Kyle E Orwig, Moira K O'Bryan, Ana C Lima, Donald F Conrad","doi":"10.1093/biolre/ioaf033","DOIUrl":null,"url":null,"abstract":"<p><p>Analysis of testis histology is fundamental to the study of male fertility, but it is a slow task with a high skill threshold. Here, we describe new neural network models for the automated classification of cell types and tubule stages from whole-slide brightfield images of mouse testis. The cell type classifier recognizes 14 cell types, including multiple steps of meiosis I prophase, with an external validation accuracy of 96%. The tubule stage classifier distinguishes all 12 canonical tubule stages with external validation accuracy of 63%, which increases to 96% when allowing for ±1 stage tolerance. We addressed generalizability of SATINN, through extensive training diversification and testing on external (non-training population) wildtype and mutant datasets. This allowed us to use SATINN to successfully process data generated in multiple laboratories. We used SATINN to analyze testis images from eight different mutant lines, generated from three different labs with a range of tissue processing protocols. Finally, we show that it is possible to use SATINN output to cluster histology images in latent space, which, when applied to the eight mutant lines, reveals known relationships in their pathology. This work represents significant progress towards a tool for robust, automated testis histopathology that can be used by multiple labs.</p>","PeriodicalId":8965,"journal":{"name":"Biology of Reproduction","volume":" ","pages":"996-1014"},"PeriodicalIF":3.1000,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"SATINN v2: automated image analysis for mouse testis histology with multi-laboratory data integration†.\",\"authors\":\"Ran Yang, Fritzie T Celino-Brady, Jessica E M Dunleavy, Katinka A Vigh-Conrad, Georgia R Atkins, Rachel L Hvasta, Christopher R X Pombar, Alexander N Yatsenko, Kyle E Orwig, Moira K O'Bryan, Ana C Lima, Donald F Conrad\",\"doi\":\"10.1093/biolre/ioaf033\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Analysis of testis histology is fundamental to the study of male fertility, but it is a slow task with a high skill threshold. Here, we describe new neural network models for the automated classification of cell types and tubule stages from whole-slide brightfield images of mouse testis. The cell type classifier recognizes 14 cell types, including multiple steps of meiosis I prophase, with an external validation accuracy of 96%. The tubule stage classifier distinguishes all 12 canonical tubule stages with external validation accuracy of 63%, which increases to 96% when allowing for ±1 stage tolerance. We addressed generalizability of SATINN, through extensive training diversification and testing on external (non-training population) wildtype and mutant datasets. This allowed us to use SATINN to successfully process data generated in multiple laboratories. We used SATINN to analyze testis images from eight different mutant lines, generated from three different labs with a range of tissue processing protocols. Finally, we show that it is possible to use SATINN output to cluster histology images in latent space, which, when applied to the eight mutant lines, reveals known relationships in their pathology. This work represents significant progress towards a tool for robust, automated testis histopathology that can be used by multiple labs.</p>\",\"PeriodicalId\":8965,\"journal\":{\"name\":\"Biology of Reproduction\",\"volume\":\" \",\"pages\":\"996-1014\"},\"PeriodicalIF\":3.1000,\"publicationDate\":\"2025-05-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Biology of Reproduction\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://doi.org/10.1093/biolre/ioaf033\",\"RegionNum\":2,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"REPRODUCTIVE BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Biology of Reproduction","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1093/biolre/ioaf033","RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"REPRODUCTIVE BIOLOGY","Score":null,"Total":0}
SATINN v2: automated image analysis for mouse testis histology with multi-laboratory data integration†.
Analysis of testis histology is fundamental to the study of male fertility, but it is a slow task with a high skill threshold. Here, we describe new neural network models for the automated classification of cell types and tubule stages from whole-slide brightfield images of mouse testis. The cell type classifier recognizes 14 cell types, including multiple steps of meiosis I prophase, with an external validation accuracy of 96%. The tubule stage classifier distinguishes all 12 canonical tubule stages with external validation accuracy of 63%, which increases to 96% when allowing for ±1 stage tolerance. We addressed generalizability of SATINN, through extensive training diversification and testing on external (non-training population) wildtype and mutant datasets. This allowed us to use SATINN to successfully process data generated in multiple laboratories. We used SATINN to analyze testis images from eight different mutant lines, generated from three different labs with a range of tissue processing protocols. Finally, we show that it is possible to use SATINN output to cluster histology images in latent space, which, when applied to the eight mutant lines, reveals known relationships in their pathology. This work represents significant progress towards a tool for robust, automated testis histopathology that can be used by multiple labs.
期刊介绍:
Biology of Reproduction (BOR) is the official journal of the Society for the Study of Reproduction and publishes original research on a broad range of topics in the field of reproductive biology, as well as reviews on topics of current importance or controversy. BOR is consistently one of the most highly cited journals publishing original research in the field of reproductive biology.