Related Experiment Video
Updated: Jul 16, 2025
A Neural Network-Based Identification of Developmentally Competent or Incompetent Mouse Fully-Grown Oocytes
Published on: March 3, 2018
Similarity network fusion to identify phenotypes of small-for-gestational-age fetuses
Jezid Miranda1,2, Cristina Paules1,3, Guillaume Noell4
1BCNatal - Barcelona Center for Maternal-Fetal and Neonatal Medicine (Hospital Clínic and Hospital Sant Joan de Deu), IDIBAPS, University of Barcelona, and Centre for Biomedical Research on Rare Diseases (CIBER-ER), Barcelona, Spain.
Insights
Machine learning identified two distinct subtypes of fetal growth restriction (FGR) using multiomics data. These subtypes, with unique molecular and clinical features, improve prediction of adverse outcomes in pregnancies affected by FGR.
Area of Science:
- Perinatal Medicine
- Genomics
- Computational Biology
Background:
- Fetal growth restriction (FGR) impacts 5-10% of pregnancies and is a leading cause of fetal death.
- The diverse clinical presentation of FGR complicates standardized classification and risk assessment.
- Existing classification systems struggle with the multiphenotypic nature of FGR.
Purpose of the Study:
- To leverage machine learning and multiomics data to identify novel phenotypes within FGR.
- To develop an unbiased classification system for FGR based on molecular and clinical data.
- To improve clinical decision-making and prediction of adverse outcomes in FGR.
Main Methods:
- Unbiased cluster analysis of FGR cases using machine learning algorithms.
- Integration of multiomics data (genomic, transcriptomic, etc.) for comprehensive analysis.
- Comparison of machine learning-derived clusters against single data-type analyses and current clinical classifications.
Main Results:
- Confirmation of two distinct subtypes of human FGR with unique molecular and clinical signatures.
- Machine learning-driven FGR subtypes demonstrated superior predictive performance for adverse maternal and neonatal outcomes compared to single-data analyses.
- The identified subtypes provide biological support for refining current clinical FGR classification.
Conclusions:
- Machine learning and multiomics analysis can effectively delineate FGR subtypes.
- These novel FGR phenotypes offer improved stratification of perinatal risk.
- The proposed approach supports the refinement of clinical classification systems for FGR, enhancing patient care.
Abstract:
Fetal growth restriction (FGR) affects 5-10% of pregnancies, is the largest contributor to fetal death, and can have long-term consequences for the child. Implementation of a standard clinical classification system is hampered by the multiphenotypic spectrum of small fetuses with substantial differences in perinatal risks. Machine learning and multiomics data can potentially revolutionize clinical decision-making in FGR by identifying new phenotypes. Herein, we describe a cluster analysis of FGR based on an unbiased machine-learning method. Our results confirm the existence of two subtypes of human FGR with distinct molecular and clinical features based on multiomic analysis. In addition, we demonstrated that clusters generated by machine learning significantly outperform single data subtype analysis and biologically support the current clinical classification in predicting adverse maternal and neonatal outcomes. Our approach can aid in the refinement of clinical classification systems for FGR supported by molecular and clinical signatures.
