{"title":"利用胚胎图像预测短期授精后受精的机器学习模型。","authors":"Masato Saito, Hirofumi Haraguchi, Ikumi Nakajima, Shinya Fukuda, Chenghua Zhu, Norio Masuya, Kazunori Matsumoto, Yuya Yoshikawa, Tomoki Tanaka, Satoshi Kishigami, Leona Matsumoto","doi":"10.1002/rmb2.12649","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>This study established a machine learning model (MLM) trained on embryo images to predict fertilization following short-term insemination for early rescue ICSI and compared its predictive performance with the embryologist's manual classification.</p><p><strong>Methods: </strong>Embryo images at 4.5 and 8 h post-insemination were preprocessed into vectors using ResNet50. The Light Gradient Boosting Machine (Light GBM) was employed for training vectors. Fertilization in the test dataset was assessed by MLM, with seven senior and 11 junior embryologists. Predictive metrics were analyzed using repeated measures ANOVA and paired <i>t</i>-tests.</p><p><strong>Results: </strong>Comparing MLM, senior embryologists, and junior embryologists, significant differences were observed in accuracy (0.71 ± 0.01, 0.75 ± 0.05, 0.61 ± 0.05), recall (0.84 ± 0.02, 0.84 ± 0.10, 0.61 ± 0.07), F1-score (0.78 ± 0.01, 0.81 ± 0.04, 0.66 ± 0.04), and area under the curve (0.73 ± 0.0 3, 0.73 ± 0.06, 0.61 ± 0.07), the MLM outperforming junior embryologists with <1 year of experience. No significant differences were observed between the MLM and senior embryologists with over 5 years of experience.</p><p><strong>Conclusions: </strong>MLM can effectively predict fertilization following short-term insemination by analyzing cytoplasmic changes in images. These results underscore the potential to enhance clinical decision-making and improve patient outcomes.</p>","PeriodicalId":21116,"journal":{"name":"Reproductive Medicine and Biology","volume":"24 1","pages":"e12649"},"PeriodicalIF":2.7000,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12000234/pdf/","citationCount":"0","resultStr":"{\"title\":\"A machine learning model for predicting fertilization following short-term insemination using embryo images.\",\"authors\":\"Masato Saito, Hirofumi Haraguchi, Ikumi Nakajima, Shinya Fukuda, Chenghua Zhu, Norio Masuya, Kazunori Matsumoto, Yuya Yoshikawa, Tomoki Tanaka, Satoshi Kishigami, Leona Matsumoto\",\"doi\":\"10.1002/rmb2.12649\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Purpose: </strong>This study established a machine learning model (MLM) trained on embryo images to predict fertilization following short-term insemination for early rescue ICSI and compared its predictive performance with the embryologist's manual classification.</p><p><strong>Methods: </strong>Embryo images at 4.5 and 8 h post-insemination were preprocessed into vectors using ResNet50. The Light Gradient Boosting Machine (Light GBM) was employed for training vectors. Fertilization in the test dataset was assessed by MLM, with seven senior and 11 junior embryologists. Predictive metrics were analyzed using repeated measures ANOVA and paired <i>t</i>-tests.</p><p><strong>Results: </strong>Comparing MLM, senior embryologists, and junior embryologists, significant differences were observed in accuracy (0.71 ± 0.01, 0.75 ± 0.05, 0.61 ± 0.05), recall (0.84 ± 0.02, 0.84 ± 0.10, 0.61 ± 0.07), F1-score (0.78 ± 0.01, 0.81 ± 0.04, 0.66 ± 0.04), and area under the curve (0.73 ± 0.0 3, 0.73 ± 0.06, 0.61 ± 0.07), the MLM outperforming junior embryologists with <1 year of experience. No significant differences were observed between the MLM and senior embryologists with over 5 years of experience.</p><p><strong>Conclusions: </strong>MLM can effectively predict fertilization following short-term insemination by analyzing cytoplasmic changes in images. These results underscore the potential to enhance clinical decision-making and improve patient outcomes.</p>\",\"PeriodicalId\":21116,\"journal\":{\"name\":\"Reproductive Medicine and Biology\",\"volume\":\"24 1\",\"pages\":\"e12649\"},\"PeriodicalIF\":2.7000,\"publicationDate\":\"2025-04-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12000234/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Reproductive Medicine and Biology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1002/rmb2.12649\",\"RegionNum\":3,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/1/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q2\",\"JCRName\":\"OBSTETRICS & GYNECOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Reproductive Medicine and Biology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1002/rmb2.12649","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/1/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"OBSTETRICS & GYNECOLOGY","Score":null,"Total":0}
A machine learning model for predicting fertilization following short-term insemination using embryo images.
Purpose: This study established a machine learning model (MLM) trained on embryo images to predict fertilization following short-term insemination for early rescue ICSI and compared its predictive performance with the embryologist's manual classification.
Methods: Embryo images at 4.5 and 8 h post-insemination were preprocessed into vectors using ResNet50. The Light Gradient Boosting Machine (Light GBM) was employed for training vectors. Fertilization in the test dataset was assessed by MLM, with seven senior and 11 junior embryologists. Predictive metrics were analyzed using repeated measures ANOVA and paired t-tests.
Results: Comparing MLM, senior embryologists, and junior embryologists, significant differences were observed in accuracy (0.71 ± 0.01, 0.75 ± 0.05, 0.61 ± 0.05), recall (0.84 ± 0.02, 0.84 ± 0.10, 0.61 ± 0.07), F1-score (0.78 ± 0.01, 0.81 ± 0.04, 0.66 ± 0.04), and area under the curve (0.73 ± 0.0 3, 0.73 ± 0.06, 0.61 ± 0.07), the MLM outperforming junior embryologists with <1 year of experience. No significant differences were observed between the MLM and senior embryologists with over 5 years of experience.
Conclusions: MLM can effectively predict fertilization following short-term insemination by analyzing cytoplasmic changes in images. These results underscore the potential to enhance clinical decision-making and improve patient outcomes.
期刊介绍:
Reproductive Medicine and Biology (RMB) is the official English journal of the Japan Society for Reproductive Medicine, the Japan Society of Fertilization and Implantation, the Japan Society of Andrology, and publishes original research articles that report new findings or concepts in all aspects of reproductive phenomena in all kinds of mammals. Papers in any of the following fields will be considered: andrology, endocrinology, oncology, immunology, genetics, function of gonads and genital tracts, erectile dysfunction, gametogenesis, function of accessory sex organs, fertilization, embryogenesis, embryo manipulation, pregnancy, implantation, ontogenesis, infectious disease, contraception, etc.