Related Experiment Video
Updated: Oct 31, 2025
Retrospective Cardiac Gating with A Prototype Small-Animal X-ray Computed Tomograph
Published on: February 21, 2025
Deep neural network-based classification of cardiotocograms outperformed conventional algorithms
Jun Ogasawara1, Satoru Ikenoue2, Hiroko Yamamoto3
1Department of Pharmacology, School of Medicine, Keio University, Tokyo, 160-8582, Japan.
Insights
A new deep neural network, CTG-net, offers an objective method for evaluating fetal heart rate (FHR) and uterine contraction (UC) patterns. This automated system aids in detecting compromised fetal status, reducing potential hypoxic injuries.
Area of Science:
- Obstetrics and Gynecology
- Medical Imaging and Signal Processing
- Artificial Intelligence in Medicine
Background:
- Cardiotocography (CTG) is crucial for assessing fetal well-being, monitoring fetal heart rates (FHR) and uterine contractions (UC).
- Manual interpretation of CTG traces by obstetricians can be subjective, leading to diagnostic variability and potentially inappropriate interventions.
- There is a need for quantitative, unbiased algorithms to improve the accuracy of CTG evaluation.
Purpose of the Study:
- To develop and evaluate a novel deep neural network, CTG-net, for the automated detection of compromised fetal status using CTG data.
- To quantitatively assess fetal well-being by analyzing temporal patterns and interrelationships between FHR and UC signals.
- To compare the performance of CTG-net against conventional algorithms and other deep learning models.
Main Methods:
- A deep neural network (CTG-net) comprising three convolutional layers was designed to extract features from CTG signals.
- CTG data was used to classify fetuses into abnormal (umbilical artery pH < 7.20 or Apgar score < 7 at 1 min) and normal groups.
- Performance was evaluated using the F1 score and the area under the receiver operating characteristic curve (AUC), with comparisons to support vector machine (SVM), k-means clustering, and long short-term memory (LSTM) models.
Main Results:
- CTG-net achieved an AUC of 0.73 ± 0.04, demonstrating significantly superior performance compared to the LSTM model.
- The F1 score was utilized to assess classification accuracy, indicating the model's effectiveness in distinguishing between normal and abnormal fetal states.
- The proposed CTG-net system showed potential for accurate and objective CTG interpretation.
Conclusions:
- CTG-net provides a quantitative and automated approach to fetal status assessment, addressing the subjectivity of manual CTG interpretation.
- Early and accurate identification of compromised fetuses through CTG-net can facilitate timely interventions.
- The implementation of such AI-driven systems may lead to a reduction in adverse outcomes, such as hypoxic injury.
Abstract:
Cardiotocography records fetal heart rates and their temporal relationship to uterine contractions. To identify high risk fetuses, obstetricians inspect cardiotocograms (CTGs) by eye. Therefore, CTG traces are often interpreted differently among obstetricians, resulting in inappropriate interventions. However, few studies have focused on quantitative and nonbiased algorithms for CTG evaluation. In this study, we propose a newly constructed deep neural network model (CTG-net) to detect compromised fetal status. CTG-net consists of three convolutional layers that extract temporal patterns and interrelationships between fetal heart rate and uterine contraction signals. We aimed to classify the abnormal group (umbilical artery pH < 7.20 or Apgar score at 1 min < 7) and the normal group from CTG data. We evaluated the performance of the CTG-net with the F1 score and compared it with conventional algorithms, namely, support vector machine and k-means clustering, and another deep neural network model, long short-term memory. CTG-net showed the area under the receiver operating characteristic curve of 0.73 ± 0.04, which was significantly higher than that of long short-term memory. CTG-net, a quantitative and automated diagnostic aid system, enables early intervention for putatively abnormal fetuses, resulting in a reduction in the number of cases of hypoxic injury.
Related Concept Videos
01:30Classification of Signals
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
01:25Cardiomyopathy I: Introduction and Classification
01:28Dysrhythmias II: Classification of Tachyarrhythmias
01:26Classification of Systems-I
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:
01:25Correlation between ECG and Cardiac Cycle
A cardiac action potential originates in the SA node and spreads throughout the atria and the AV node in approximately 0.03 seconds. This results in the P wave in an ECG and triggers atrial contraction. The action potential is then briefly slowed at the AV node, allowing the atria to contract and fill the ventricles with blood before...