Using deep learning to predict the outcome of live birth from more than 10,000 embryo data
Abstract Background Recently, the combination of deep learning and time-lapse imaging provides an objective, standard and scientific solution for embryo selection. However, the reported studies were based on blastocyst formation or clinical pregnancy as the end point. To the best of our knowledge, t...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2022-01-01
|
| Series: | BMC Pregnancy and Childbirth |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12884-021-04373-5 |
| _version_ | 1798026541635993600 |
|---|---|
| author | Bo Huang Shunyuan Zheng Bingxin Ma Yongle Yang Shengping Zhang Lei Jin |
| author_facet | Bo Huang Shunyuan Zheng Bingxin Ma Yongle Yang Shengping Zhang Lei Jin |
| author_sort | Bo Huang |
| collection | DOAJ |
| description | Abstract Background Recently, the combination of deep learning and time-lapse imaging provides an objective, standard and scientific solution for embryo selection. However, the reported studies were based on blastocyst formation or clinical pregnancy as the end point. To the best of our knowledge, there is no predictive model that uses the outcome of live birth as the predictive end point. Can a deep learning model predict the probability of live birth from time-lapse system? Methods This study retrospectively analyzed the time-lapse data and live birth outcomes of embryos samples from January 2018 to November 2019. We used the SGD optimizer with an initial learning rate of 0.025 and cosine learning rate reduction strategy. The network is randomly initialized and trained for 200 epochs from scratch. The model is quantitively evaluated over a hold-out test and a 5-fold cross-validation by the average area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Results The deep learning model was able to predict live birth outcomes from time-lapse images with an AUC of 0.968 in 5-fold stratified cross-validation. Conclusions This research reported a deep learning model that predicts the live birth outcome of a single blastocyst transfer. This efficient model for predicting the outcome of live births can automatically analyze the time-lapse images of the patient’s embryos without the need for manual embryo annotation and evaluation, and then give a live birth prediction score for each embryo, and sort the embryos by the predicted value. |
| first_indexed | 2024-04-11T18:36:58Z |
| format | Article |
| id | doaj.art-e51934955a7c427f9426aae9799a3b87 |
| institution | Directory Open Access Journal |
| issn | 1471-2393 |
| language | English |
| last_indexed | 2024-04-11T18:36:58Z |
| publishDate | 2022-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Pregnancy and Childbirth |
| spelling | doaj.art-e51934955a7c427f9426aae9799a3b872022-12-22T04:09:14ZengBMCBMC Pregnancy and Childbirth1471-23932022-01-012211710.1186/s12884-021-04373-5Using deep learning to predict the outcome of live birth from more than 10,000 embryo dataBo Huang0Shunyuan Zheng1Bingxin Ma2Yongle Yang3Shengping Zhang4Lei Jin5Reproductive Medicine Center, Tongji Hospital, Tongji Medicine College, Huazhong University of Science and TechnologySchool of Computer Science and Technology, Harbin Institute of TechnologyReproductive Medicine Center, Tongji Hospital, Tongji Medicine College, Huazhong University of Science and TechnologyReproductive Medicine Center, Tongji Hospital, Tongji Medicine College, Huazhong University of Science and TechnologySchool of Computer Science and Technology, Harbin Institute of TechnologyReproductive Medicine Center, Tongji Hospital, Tongji Medicine College, Huazhong University of Science and TechnologyAbstract Background Recently, the combination of deep learning and time-lapse imaging provides an objective, standard and scientific solution for embryo selection. However, the reported studies were based on blastocyst formation or clinical pregnancy as the end point. To the best of our knowledge, there is no predictive model that uses the outcome of live birth as the predictive end point. Can a deep learning model predict the probability of live birth from time-lapse system? Methods This study retrospectively analyzed the time-lapse data and live birth outcomes of embryos samples from January 2018 to November 2019. We used the SGD optimizer with an initial learning rate of 0.025 and cosine learning rate reduction strategy. The network is randomly initialized and trained for 200 epochs from scratch. The model is quantitively evaluated over a hold-out test and a 5-fold cross-validation by the average area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Results The deep learning model was able to predict live birth outcomes from time-lapse images with an AUC of 0.968 in 5-fold stratified cross-validation. Conclusions This research reported a deep learning model that predicts the live birth outcome of a single blastocyst transfer. This efficient model for predicting the outcome of live births can automatically analyze the time-lapse images of the patient’s embryos without the need for manual embryo annotation and evaluation, and then give a live birth prediction score for each embryo, and sort the embryos by the predicted value.https://doi.org/10.1186/s12884-021-04373-5Time-lapse microscopyEmbryo developmentEmbryo qualityPregnancy |
| spellingShingle | Bo Huang Shunyuan Zheng Bingxin Ma Yongle Yang Shengping Zhang Lei Jin Using deep learning to predict the outcome of live birth from more than 10,000 embryo data BMC Pregnancy and Childbirth Time-lapse microscopy Embryo development Embryo quality Pregnancy |
| title | Using deep learning to predict the outcome of live birth from more than 10,000 embryo data |
| title_full | Using deep learning to predict the outcome of live birth from more than 10,000 embryo data |
| title_fullStr | Using deep learning to predict the outcome of live birth from more than 10,000 embryo data |
| title_full_unstemmed | Using deep learning to predict the outcome of live birth from more than 10,000 embryo data |
| title_short | Using deep learning to predict the outcome of live birth from more than 10,000 embryo data |
| title_sort | using deep learning to predict the outcome of live birth from more than 10 000 embryo data |
| topic | Time-lapse microscopy Embryo development Embryo quality Pregnancy |
| url | https://doi.org/10.1186/s12884-021-04373-5 |
| work_keys_str_mv | AT bohuang usingdeeplearningtopredicttheoutcomeoflivebirthfrommorethan10000embryodata AT shunyuanzheng usingdeeplearningtopredicttheoutcomeoflivebirthfrommorethan10000embryodata AT bingxinma usingdeeplearningtopredicttheoutcomeoflivebirthfrommorethan10000embryodata AT yongleyang usingdeeplearningtopredicttheoutcomeoflivebirthfrommorethan10000embryodata AT shengpingzhang usingdeeplearningtopredicttheoutcomeoflivebirthfrommorethan10000embryodata AT leijin usingdeeplearningtopredicttheoutcomeoflivebirthfrommorethan10000embryodata |