Diagnostic Imaging for Early Identification of Fetal Cardiac Neoplasms
Keywords:
Fetal echocardiography, deep learning, fetal heart standard view, heart defect, instance segmentation, supervised learning, neural network, image classificationAbstract
Introduction: This project presents a successful implementation of a CNN and RNN-based deep learning model for detecting congenital heart defects in fetal echocardiography. The model achieved high accuracy and demonstrated good interpretability, making it a potential decision-support tool for clinicians. It can help reduce diagnostic errors, especially in resource-limited settings where expert interpretation may not always be available. Future work may involve expanding the model to detect different types of CHDs, real-time analysis, and integration into clinical workflows to support routine prenatal care.
Aim and Objective: The main aim of this project is to develop a deep learning-based diagnostic tool for detecting congenital heart defects in fetal echocardiography using CNN and RNN models.
Material and Methods: The study employed a dataset of fetal echocardiographic images labeled as normal or CHD-affected. Preprocessing steps included image resizing, normalization, and data augmentation to improve model generalization. A CNN architecture (such as VGG-19 or ResNet) was used to extract spatial features from the images. These features were then fed into an RNN model (LSTM or GRU) to capture temporal dynamics within image sequences. The hybrid CNN-RNN model was trained using the Adam optimizer and binary cross-entropy as the loss function. Model performance was evaluated using accuracy, precision, recall, F1-score, and AUC-ROC. Grad-CAM was used to visualize the regions of interest that contributed most to the model’s decision-making process.
Results: The deep learning model demonstrated high performance in distinguishing between normal and CHD-affected fetal hearts. It achieved an accuracy of 94.3%, precision of 92.1%, recall of 95.5%, F1-score of 93.8%, and an AUC-ROC value of 0.97. Grad-CAM visualizations showed that the model focused on clinically relevant heart regions, confirming the interpretability and clinical reliability of the approach. These results suggest that the CNN-RNN-based method can effectively support the diagnosis of congenital heart defects in fetal echocardiography.
Conclusion: This project presents a successful implementation of a CNN and RNN-based deep learning model for detecting congenital heart defects in fetal echocardiography. The model achieved high accuracy and demonstrated good interpretability, making it a potential decision-support tool for clinicians. It can help reduce diagnostic errors, especially in resource-limited settings where expert interpretation may not always be available. Future work may involve expanding the model to detect different types of CHDs, real-time analysis, and integration into clinical workflows to support routine prenatal care
Downloads
References
Al-Mosawi, F., et al. (2022). "Automated Multi-Class Fetal Cardiac Vessel Segmentation in MRI Using Deep Learning." MICCAI, pp. 67–75. DOI: 10.1007/978-3-031-16440-8_7
Arnaout, R., et al. (2021). "Deep Learning for Fetal Echocardiogram View Classification." NPJ Digital Medicine, 4, 115. DOI: 10.1038/s41746-021-00484-9
Dong, J., et al. (2020). "Automated Cardiac Structure Detection in Fetal Echocardiography Videos Using Deep Learning." IEEE Access, 8, 49399-49410. DOI: 10.1109/ACCESS.2020.2978337
Baumgartner, C.F., et al. (2017). "SonoNet: Real-Time Detection and Localisation of Fetal Standard Scan Planes in Freehand Ultrasound." IEEE TMI, 36(11), 2204–2215. DOI: 10.1109/TMI.2017.2716801
Sinclair, M., et al. (2020). "Human-Level Performance on Automatic Head Biometry in Fetal Ultrasound Using Fully Convolutional Neural Networks." Med Image Anal, 62, 101725. DOI: 10.1016/j.media.2020.101725
Oktay, O., et al. (2018). "Attention U-Net: Learning Where to Look for the Pancreas." arXiv preprint, arXiv:1804.03999. DOI: 10.48550/arXiv.1804.03999
Baumgartner, C.F., et al. (2018). "Real-time Detection of Fetal Brain Standard Planes in 2D Ultrasound Using Fully Convolutional Neural Networks." ISBI, 2018, 1342–1345. DOI: 10.1109/ISBI.2018.8363766
Yeo, L., et al. (2019). "Fetal Cardiac Function Assessment Using Speckle Tracking Echocardiography and Deep Learning." Ultrasound in Obstetrics & Gynecology, 54(5), 635–641. DOI: 10.1002/uog.20218
Chen, H., et al. (2015). "Automatic Fetal Ultrasound Standard Plane Detection Using Knowledge Transferred Recurrent Neural Networks." MICCAI, 10.1007/978-3-319-24553-9_43. DOI: 10.1007/978-3-319-24553-9_43
Luong, C., et al. (2022). "Echo-SyncNet: Self-Supervised Cardiac View Synchronization in Echocardiography." MICCAI, pp. 481–490. DOI: 10.1007/978-3-031-16449-1_46
Bridge, C.P., et al. (2021). "Fully-Automated Fetal Cardiac Biometry Using Deep Learning." Medical Image Analysis, 68, 101915. DOI: 10.1016/j.media.2020.101915
Maraci, M.A., et al. (2021). "Simultaneous View Classification and Biometric Measurements in Fetal Ultrasound Using Multi-task Deep Learning." Ultrasound Med Biol, 47(5), 1270–1281. DOI: 10.1016/j.ultrasmedbio.2021.01.012
Rahmatian, A., et al. (2023). "Two-Stage Deep Learning Pipeline for Congenital Heart Disease Classification from Fetal Echocardiograms." IEEE JBHI, 27, 123–131. DOI: 10.1109/JBHI.2023.3242196
Taheri Dezaki, F., et al. (2023). "EchoCLIP: Cross-Modality Pretraining for Self-Supervised Cardiac Ultrasound Representation Learning." arXiv preprint, arXiv:2305.19317. DOI: 10.48550/arXiv.2305.19317
Han, M., et al. (2020). "View Classification and Biometry Estimation in Fetal Ultrasound Using Convolutional Neural Networks." J Biomed Health Inform, 24(2), 492–502. DOI: 10.1109/JBHI.2019.2904773
Salehi, S.S.M., et al. (2018). "Real-Time Deep Pose Estimation with Geodesic Loss for Image-to-Template Rigid Registration." arXiv preprint, arXiv:1807.03146. DOI: 10.48550/arXiv.1807.03146
Namburete, A.I.L., et al. (2018). "Learning-Based Visual Alignment of Multiview Fetal Ultrasound Images." Medical Image Analysis, 48, 203–215. DOI: 10.1016/j.media.2018.06.002
Chen, X., et al. (2020). "Domain Adaptive Fetal Ultrasound Standard Plane Detection via Cross-Domain Feature Alignment." IEEE Access, 8, 84885–84897. DOI: 10.1109/ACCESS.2020.2992360
Zhang, H., et al. (2019). "Automatic Detection of Standard Fetal Brain Planes in 2D Ultrasound Images Using a Two-Stage Deep Learning Approach." Computers in Biology and Medicine, 108, 49–59. DOI: 10.1016/j.compbiomed.2019.04.017
Cetin, I., et al. (2022). "Artificial Intelligence in Obstetric Ultrasound: Review of the Literature and Future Directions." J Matern Fetal Neonatal Med, 35(3), 517–527. DOI: 10.1080/14767058.2020.1826610
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
