Related Experiment Video
Updated: May 16, 2025
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Prediction of IUGR condition at birth by means of CTG recordings and a ResNet model
Edoardo Spairani1, Giulio Steyde2, Federica Spuri Forotti2
1Department of Electrical, Computer and Biomedical Engineering, Università di Pavia, 27100, Pavia, Italy.
Insights
Deep learning models predict intrauterine growth restriction (IUGR) using cardiotocography (CTG) signals, achieving 80% accuracy. This approach offers a promising tool for early IUGR detection, complementing existing methods.
Area of Science:
- Perinatal Medicine
- Artificial Intelligence in Healthcare
- Signal Processing
Background:
- Sub-optimal uterine-placental perfusion and fetal nutrition cause intrauterine growth restriction (IUGR), a condition challenging to diagnose antenatally.
- Cardiotocography (CTG) is used for fetal monitoring, but reliably diagnosing IUGR before birth remains difficult.
- Deep learning (DL) presents a potential solution for improving IUGR diagnosis using CTG data.
Purpose of the Study:
- To develop and evaluate a deep learning (DL) model for predicting intrauterine growth restriction (IUGR) at birth.
- To utilize antenatal cardiotocography (CTG) signals as input for the DL model.
- To enhance the early detection capabilities for IUGR.
Main Methods:
- A ResNet architecture was employed, utilizing a two-step training process.
- The model was trained in two phases: initial training on "presumed" data and fine-tuning on "confirmed" data.
- This approach focused on minimizing data loss and refining performance based on confirmed outcomes.
Main Results:
- The DL model achieved a balanced accuracy of 80% on a hold-out test set of confirmed IUGR cases.
- This performance surpasses that achieved using standard clinical guidelines.
- The study utilized a significantly larger dataset compared to similar research in the field.
Conclusions:
- The developed DL model demonstrates strong potential for predicting IUGR from CTG signals.
- Integrating DL with CTG analysis can complement imaging technologies for improved early IUGR detection.
- This AI-driven approach offers a valuable tool for antenatal monitoring and diagnosis.
Objective:
Sub-optimal uterine-placental perfusion and fetal nutrition can lead to intrauterine growth restriction (IUGR), also called fetal growth restriction (FGR). Antenatal cardiotocography (CTG) can aid in the early detection of IUGR. Reliably diagnosing IUGR before delivery remains challenging, and deep learning (DL) techniques offer potential solutions. This paper describes the development of a DL approach to predict an IUGR condition at birth by using CTG signals collected during antenatal monitoring.
Materials And Methods:
Our method is encapsulated in the concept of a two-step training process of a ResNet architecture. The primary focus is on the minimization of data loss, which motivates the division into "presumed" and "confirmed" datasets, which is employed to distinguish based on the presence of information at birth. The method involves fine-tuning: the initial training utilizes "presumed" data to train the network, and the subsequent training employs data representing certain knowledge to refine its performance.
Results:
The DL model reaches a balanced accuracy of 80% on a hold-out test set of confirmed cases, which is better than what obtained by using standard clinical guidelines.
Discussion:
The results of our work are compared to the results of similar papers dealing with the prediction of IUGR condition at birth and in general with the prediction of fetal pathological conditions. Our final results are obtained using a very large dataset compared to other papers reported in the literature.
Conclusion:
The inclusion of DL methods on CTG signals may complement imaging technologies and improve the early detection of IUGR.
