Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Machine learning prediction of live birth after IVF using the morphological uterus sonographic assessment group features of adenomyosis

Alson, Sara LU orcid ; Björnsson, Ola LU orcid ; Henic, Emir LU ; Hansson, Stefan LU orcid and Sladkevicius, Povilas LU orcid (2026) In Scientific Reports 16.
Abstract
Predicting live birth after the first IVF/ICSI treatment is challenging, as many factors may interact to affect IVF/ICSI outcomes. Adenomyosis is one factor that impacts live birth rates. Machine learning algorithms have been shown valuable for detecting complex dependencies and predicting outcomes in different clinical settings. We aimed to develop a prediction model for live birth after IVF/ICSI treatment, using the Extreme Gradient Boosting (XGBoost) algoritm and incorporating the revised Morphological Uterus Sonographic Assessment (MUSA) group features of adenomyosis. We used a machine learning model based on data from 1037 women undergoing their first IVF/ICSI treatment between January 2019 and October 2022. The importance of each... (More)
Predicting live birth after the first IVF/ICSI treatment is challenging, as many factors may interact to affect IVF/ICSI outcomes. Adenomyosis is one factor that impacts live birth rates. Machine learning algorithms have been shown valuable for detecting complex dependencies and predicting outcomes in different clinical settings. We aimed to develop a prediction model for live birth after IVF/ICSI treatment, using the Extreme Gradient Boosting (XGBoost) algoritm and incorporating the revised Morphological Uterus Sonographic Assessment (MUSA) group features of adenomyosis. We used a machine learning model based on data from 1037 women undergoing their first IVF/ICSI treatment between January 2019 and October 2022. The importance of each variable on the model was illustrated with the Shapley additive explanations algorithm (SHAP) variable importance. The prediction model was presented with the area under receiver operating characteristics curve (ROC). The proposed XGBoost model had a test AUC of 0.66 and accuracy of 0.59. S-AMH was the best variable for predicting live birth with a mean SHAP of 0.21, followed by a regular junctional zone as the best ultrasonographic variable, mean SHAP 0.13. The predictive ability of MUSA features in relation to live birth was limited. Additional variables should be included in future prediction models. (Less)
Abstract (Swedish)
Predicting live birth after the first IVF/ICSI treatment is challenging, as many factors may interactto affect IVF/ICSI outcomes. Adenomyosis is one factor that impacts live birth rates. Machinelearning algorithms have been shown valuable for detecting complex dependencies and predictingoutcomes in different clinical settings. We aimed to develop a prediction model for live birth afterIVF/ICSI treatment, using the Extreme Gradient Boosting (XGBoost) algoritm and incorporatingthe revised Morphological Uterus Sonographic Assessment (MUSA) group features of adenomyosis.We used a machine learning model based on data from 1037 women undergoing their first IVF/ICSItreatment between January 2019 and October 2022. The importance of each variable... (More)
Predicting live birth after the first IVF/ICSI treatment is challenging, as many factors may interactto affect IVF/ICSI outcomes. Adenomyosis is one factor that impacts live birth rates. Machinelearning algorithms have been shown valuable for detecting complex dependencies and predictingoutcomes in different clinical settings. We aimed to develop a prediction model for live birth afterIVF/ICSI treatment, using the Extreme Gradient Boosting (XGBoost) algoritm and incorporatingthe revised Morphological Uterus Sonographic Assessment (MUSA) group features of adenomyosis.We used a machine learning model based on data from 1037 women undergoing their first IVF/ICSItreatment between January 2019 and October 2022. The importance of each variable on the modelwas illustrated with the Shapley additive explanations algorithm (SHAP) variable importance. Theprediction model was presented with the area under receiver operating characteristics curve (ROC). Theproposed XGBoost model had a test AUC of 0.66 and accuracy of 0.59. S-AMH was the best variablefor predicting live birth with a mean SHAP of 0.21, followed by a regular junctional zone as the bestultrasonographic variable, mean SHAP 0.13. The predictive ability of MUSA features in relation to livebirth was limited. Additional variables should be included in future prediction models. (Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Morpholigical Uterus Sonographic Assessment (MUSA) group, XGBoost, Live birth, IVF/ICSI, Adenomyosis, machine learning (ML), infertility
in
Scientific Reports
volume
16
article number
4324
publisher
Nature Publishing Group
external identifiers
  • pmid:41620505
  • scopus:105029069079
ISSN
2045-2322
DOI
10.1038/s41598-025-31013-1
project
Deep infiltrating endometriosis (DIE) diagnosed by ultrasound in subfertile women - effect on Anti-Müllerian Hormone (AMH), antral follicle count (AFC) and fertility treatment outcome.
language
English
LU publication?
yes
id
2e220541-a3b5-4c05-8504-896abe9b1d0b
date added to LUP
2026-02-02 11:36:02
date last changed
2026-02-13 04:00:30
@article{2e220541-a3b5-4c05-8504-896abe9b1d0b,
  abstract     = {{Predicting live birth after the first IVF/ICSI treatment is challenging, as many factors may interact to affect IVF/ICSI outcomes. Adenomyosis is one factor that impacts live birth rates. Machine learning algorithms have been shown valuable for detecting complex dependencies and predicting outcomes in different clinical settings. We aimed to develop a prediction model for live birth after IVF/ICSI treatment, using the Extreme Gradient Boosting (XGBoost) algoritm and incorporating the revised Morphological Uterus Sonographic Assessment (MUSA) group features of adenomyosis. We used a machine learning model based on data from 1037 women undergoing their first IVF/ICSI treatment between January 2019 and October 2022. The importance of each variable on the model was illustrated with the Shapley additive explanations algorithm (SHAP) variable importance. The prediction model was presented with the area under receiver operating characteristics curve (ROC). The proposed XGBoost model had a test AUC of 0.66 and accuracy of 0.59. S-AMH was the best variable for predicting live birth with a mean SHAP of 0.21, followed by a regular junctional zone as the best ultrasonographic variable, mean SHAP 0.13. The predictive ability of MUSA features in relation to live birth was limited. Additional variables should be included in future prediction models.}},
  author       = {{Alson, Sara and Björnsson, Ola and Henic, Emir and Hansson, Stefan and Sladkevicius, Povilas}},
  issn         = {{2045-2322}},
  keywords     = {{Morpholigical Uterus Sonographic Assessment (MUSA) group; XGBoost; Live birth; IVF/ICSI; Adenomyosis; machine learning (ML); infertility}},
  language     = {{eng}},
  month        = {{02}},
  publisher    = {{Nature Publishing Group}},
  series       = {{Scientific Reports}},
  title        = {{Machine learning prediction of live birth after IVF using the morphological uterus sonographic assessment group features of adenomyosis}},
  url          = {{http://dx.doi.org/10.1038/s41598-025-31013-1}},
  doi          = {{10.1038/s41598-025-31013-1}},
  volume       = {{16}},
  year         = {{2026}},
}