Classification of normal and abnormal fetal heart ultrasound images and identification of ventricular septal defects based on deep learning
-
Yiru Yang
,
Peizhong Liu
Abstract
Objectives
Congenital heart defects (CHDs) are the most common birth defects. Recently, artificial intelligence (AI) was used to assist in CHD diagnosis. No comparison has been made among the various types of algorithms that can assist in the prenatal diagnosis.
Methods
Normal and abnormal fetal ultrasound heart images, including five standard views, were collected according to the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) Practice guidelines. You Only Look Once version 5 (YOLOv5) models were trained and tested. An excellent model was screened out after comparing YOLOv5 with other classic detection methods.
Results
On the training set, YOLOv5n performed slightly better than the others. On the validation set, YOLOv5n attained the highest overall accuracy (90.67 %). On the CHD test set, YOLOv5n, which only needed 0.007 s to recognize each image, had the highest overall accuracy (82.93 %), and YOLOv5l achieved the best accuracy on the abnormal dataset (71.93 %). On the VSD test set, YOLOv5l had the best performance, with a 92.79 % overall accuracy rate and 92.59 % accuracy on the abnormal dataset. The YOLOv5 models achieved better performance than the Fast region-based convolutional neural network (RCNN) & ResNet50 model and the Fast RCNN & MobileNetv2 model on the CHD test set (p<0.05) and VSD test set (p<0.01).
Conclusions
YOLOv5 models are able to accurately distinguish normal and abnormal fetal heart ultrasound images, especially with respect to the identification of VSD, which have the potential to assist ultrasound in prenatal diagnosis.
-
Research funding: None declared.
-
Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Competing interests: Authors state no conflict of interest.
-
Informed consent: Informed consent was obtained from all individuals included in this study.
-
Ethical approval: This study was approved by the Medical Ethics Committee of the Second Affiliated Hospital of Fujian Medical University.
References
1. Hu, P, Qiao, F, Wang, Y, Meng, L, Ji, X, Luo, C, et al.. Clinical application of targeted next-generation sequencing in fetuses with congenital heart defect. Ultrasound Obstet Gynecol 2018;52:205–11. https://doi.org/10.1002/uog.19042.Suche in Google Scholar PubMed
2. Shabana, NA, Shahid, SU, Irfan, U. Genetic contribution to congenital heart disease (CHD). Pediatr Cardiol 2020;41:12–23. https://doi.org/10.1007/s00246-019-02271-4.Suche in Google Scholar PubMed
3. Wang, J, Liu, X, Wang, F, Zheng, L, Gao, F, Zhang, H, et al.. Automated interpretation of congenital heart disease from multi-view echocardiograms. Med Image Anal 2021;69:101942. https://doi.org/10.1016/j.media.2020.101942.Suche in Google Scholar PubMed
4. Zhao, QM, Liu, F, Wu, L, Ma, XJ, Niu, C, Huang, GY. Prevalence of congenital heart disease at live birth in China. J Pediatr 2019;204:53–8. https://doi.org/10.1016/j.jpeds.2018.08.040.Suche in Google Scholar PubMed
5. International Society of Ultrasound in Obstetrics and Gynecology, Carvalho, JS, Allan, LD, Chaoui, R, Copel, JA, DeVore, GR, et al.. ISUOG Practice Guidelines (updated): sonographic screening examination of the fetal heart. Ultrasound Obstet Gynecol 2013;41:348–59. https://doi.org/10.1002/uog.12403.Suche in Google Scholar PubMed
6. Mwikirize, C, Nosher, JL, Hacihaliloglu, I. Convolution neural networks for real-time needle detection and localization in 2D ultrasound. Int J Comput Assist Radiol Surg 2018;13:647–57. https://doi.org/10.1007/s11548-018-1721-y.Suche in Google Scholar PubMed
7. Elpeltagy, M, Sallam, H. Automatic prediction of COVID-19 from chest images using modified ResNet50. Multimed Tool Appl 2021;80:1–13. https://doi.org/10.1007/s11042-021-10783-6.Suche in Google Scholar PubMed PubMed Central
8. Xie, B, Lei, T, Wang, N, Cai, H, Xian, J, He, M, et al.. Computer-aided diagnosis for fetal brain ultrasound images using deep convolutional neural networks. Int J Comput Assist Radiol Surg 2020;15:1303–12. https://doi.org/10.1007/s11548-020-02182-3.Suche in Google Scholar PubMed
9. Chen, P, Chen, Y, Deng, Y, Wang, Y, He, P, Lv, X, et al.. A preliminary study to quantitatively evaluate the development of maturation degree for fetal lung based on transfer learning deep model from ultrasound images. Int J Comput Assist Radiol Surg 2020;15:1407–15. https://doi.org/10.1007/s11548-020-02211-1.Suche in Google Scholar PubMed
10. Cho, HC, Sun, S, Min Hyun, C, Kwon, JY, Kim, B, Park, Y, et al.. Automated ultrasound assessment of amniotic fluid index using deep learning. Med Image Anal 2021;69:101951. https://doi.org/10.1016/j.media.2020.101951.Suche in Google Scholar PubMed
11. Xie, HN, Wang, N, He, M, Zhang, LH, Cai, HM, Xian, JB, et al.. Using deep-learning algorithms to classify fetal brain ultrasound images as normal or abnormal. Ultrasound Obstet Gynecol 2020;56:579–87. https://doi.org/10.1002/uog.21967.Suche in Google Scholar PubMed
12. Zeng, Y, Tsui, PH, Wu, W, Zhou, Z, Wu, S. Fetal ultrasound image segmentation for automatic head circumference biometry using deeply supervised attention-gated V-net. J Digit Imag 2021;34:134–48. https://doi.org/10.1007/s10278-020-00410-5.Suche in Google Scholar PubMed PubMed Central
13. Chen, X, He, M, Dan, T, Wang, N, Lin, M, Zhang, L, et al.. Automatic measurements of fetal lateral ventricles in 2D ultrasound images using deep learning. Front Neurol 2020;11:526. https://doi.org/10.3389/fneur.2020.00526.Suche in Google Scholar PubMed PubMed Central
14. Waern, M, Mellander, M, Berg, A, Carlsson, Y. Prenatal detection of congenital heart disease – results of a Swedish screening program 2013–2017. BMC Pregnancy Childbirth 2021;21:579. https://doi.org/10.1186/s12884-021-04028-5.Suche in Google Scholar PubMed PubMed Central
15. Gómez, O, Martínez, JM, Olivella, A, Bennasar, M, Crispi, F, Masoller, N, et al.. Isolated ventricular septal defects in the era of advanced fetal echocardiography: risk of chromosomal anomalies and spontaneous closure rate from diagnosis to age of one year. Ultrasound Obstet Gynecol 2014;43:65–71.10.1002/uog.12527Suche in Google Scholar PubMed
16. Bruneau, BG. The developmental genetics of congenital heart disease. Nature 2008;451:943–8. https://doi.org/10.1038/nature06801.Suche in Google Scholar PubMed
17. Hernandez-Andrade, E, Patwardhan, M, Cruz-Lemini, M, Luewan, S. Early evaluation of the fetal heart. Fetal Diagn Ther 2017;42:161–73. https://doi.org/10.1159/000477564.Suche in Google Scholar PubMed
18. Chassagnon, G, Vakalopoulou, M, Paragios, N, Revel, MP. Artificial intelligence applications for thoracic imaging. Eur J Radiol 2020;123:108774. https://doi.org/10.1016/j.ejrad.2019.108774.Suche in Google Scholar PubMed
19. Rizzo, G, Pietrolucci, ME, Capponi, A, Mappa, I. Exploring the role of artificial intelligence in the study of fetal heart. Int J Cardiovasc Imag 2022;38:1017–9. https://doi.org/10.1007/s10554-022-02588-x.Suche in Google Scholar PubMed
20. Dozen, A, Komatsu, M, Sakai, A, Komatsu, R, Shozu, K, Machino, H, et al.. Image segmentation of the ventricular septum in fetal cardiac ultrasound videos based on deep learning using time-series information. Biomolecules 2020;10:1526. https://doi.org/10.3390/biom10111526.Suche in Google Scholar PubMed PubMed Central
21. Dong, J, Liu, S, Liao, Y, Wen, H, Lei, B, Li, S, et al.. A generic quality control framework for fetal ultrasound cardiac four-chamber planes. IEEE J Biomed Health Inf 2020;24:931–42. https://doi.org/10.1109/jbhi.2019.2948316.Suche in Google Scholar PubMed
22. Van Nisselrooij, AEL, Teunissen, AKK, Clur, SA, Rozendaal, L, Pajkrt, E, Linskens, IH, et al.. Why are congenital heart defects being missed? Ultrasound Obstet Gynecol 2020;55:747–57. https://doi.org/10.1002/uog.20358.Suche in Google Scholar PubMed PubMed Central
23. Wu, W, Liu, H, Li, L, Long, Y, Wang, X, Wang, Z, et al.. Application of local fully convolutional neural network combined with YOLO v5 algorithm in small target detection of remote sensing image. PLoS One 2021;16:e0259283. https://doi.org/10.1371/journal.pone.0259283.Suche in Google Scholar PubMed PubMed Central
24. Luo, Y, Zhang, Y, Sun, X, Dai, H, Chen, X. Intelligent solutions in chest abnormality detection based on YOLOv5 and ResNet50. J Healthc Eng 2021;2021:2267635. https://doi.org/10.1155/2021/2267635.Suche in Google Scholar PubMed PubMed Central
25. Lyu, S, Li, R, Zhao, Y, Li, Z, Fan, R, Liu, S. Green citrus detection and counting in orchards based on YOLOv5-CS and AI edge system. Sensors 2022;22:576. https://doi.org/10.3390/s22020576.Suche in Google Scholar PubMed PubMed Central
26. Tazin, T, Sarker, S, Gupta, P, Ayaz, F, Islam, S, Monirujjaman Khan, M, et al.. A robust and novel approach for brain tumor classification using convolutional neural network. Comput Intell Neurosci 2021;2021:2392395. https://doi.org/10.1155/2021/2392395.Suche in Google Scholar PubMed PubMed Central
27. Lang, X, Ren, Z, Wan, D, Zhang, Y, Shu, S. MR-YOLO: an improved YOLOv5 network for detecting magnetic ring surface defects. Sensors 2022;22:9897. https://doi.org/10.3390/s22249897.Suche in Google Scholar PubMed PubMed Central
Supplementary Material
This article contains supplementary material (https://doi.org/10.1515/jpm-2023-0041).
© 2023 Walter de Gruyter GmbH, Berlin/Boston
Artikel in diesem Heft
- Frontmatter
- Editorial
- ChatGPT and artificial intelligence in the Journal of Perinatal Medicine
- Reviews
- A systematic review and critical evaluation of quality of clinical practice guidelines on fetal growth restriction
- An exploration of barriers to access to trial of labor and vaginal birth after cesarean in the United States: a scoping review
- Opinion Paper
- A call for public funding of invasive and non-invasive prenatal testing
- Original Articles – Obstetrics
- The AccuFlow sensor: a novel digital health tool to assess intrapartum blood loss at cesarean delivery
- Risk factors associated with third- and fourth-degree perineal lacerations in singleton vaginal deliveries: a comprehensive United States population analysis 2016–2020
- Changes in use of 17-OHPC after the PROLONG trial: a physician survey
- Retrospective comparison of monochorionic diamniotic twin pregnancies stratified by spontaneous or artificial conception
- Associations of cesarean sections with comorbidities within the Pregnancy Risk Assessment Monitoring System
- The spatial expression of mTORC2-AKT-IP3R signal pathway in mitochondrial combination of endoplasmic reticulum of maternal fetal interface trophoblast in intrahepatic cholestasis of pregnancy
- Comprehensive analysis of macrosomia: exploring the association between first-trimester alanine aminotransferase and uric acid measurements in pregnant women
- Use, misuse, and overuse of antenatal corticosteroids. A retrospective cohort study
- Classification of normal and abnormal fetal heart ultrasound images and identification of ventricular septal defects based on deep learning
- Virtual touch IQ elastography in the evaluation of fetal liver and placenta in pregnancies with gestational diabetes mellitus
- Fetomaternal outcome of scarred uterine rupture compared with primary uterine rupture: a retrospective cohort study
- Original Articles – Fetus
- The assessment of fetal cardiac functions in pregnancies with autoimmune diseases: a prospective case-control study
- The relationship of maternal polymorphisms of genes related to meiosis and DNA damage repair with fetal chromosomal stability
- Original Articles – Neonates
- German obstetrician’s self-reported attitudes and handling in threatening preterm birth at the limits of viability
- Do parents get what they want during bad news delivery in NICU?
Artikel in diesem Heft
- Frontmatter
- Editorial
- ChatGPT and artificial intelligence in the Journal of Perinatal Medicine
- Reviews
- A systematic review and critical evaluation of quality of clinical practice guidelines on fetal growth restriction
- An exploration of barriers to access to trial of labor and vaginal birth after cesarean in the United States: a scoping review
- Opinion Paper
- A call for public funding of invasive and non-invasive prenatal testing
- Original Articles – Obstetrics
- The AccuFlow sensor: a novel digital health tool to assess intrapartum blood loss at cesarean delivery
- Risk factors associated with third- and fourth-degree perineal lacerations in singleton vaginal deliveries: a comprehensive United States population analysis 2016–2020
- Changes in use of 17-OHPC after the PROLONG trial: a physician survey
- Retrospective comparison of monochorionic diamniotic twin pregnancies stratified by spontaneous or artificial conception
- Associations of cesarean sections with comorbidities within the Pregnancy Risk Assessment Monitoring System
- The spatial expression of mTORC2-AKT-IP3R signal pathway in mitochondrial combination of endoplasmic reticulum of maternal fetal interface trophoblast in intrahepatic cholestasis of pregnancy
- Comprehensive analysis of macrosomia: exploring the association between first-trimester alanine aminotransferase and uric acid measurements in pregnant women
- Use, misuse, and overuse of antenatal corticosteroids. A retrospective cohort study
- Classification of normal and abnormal fetal heart ultrasound images and identification of ventricular septal defects based on deep learning
- Virtual touch IQ elastography in the evaluation of fetal liver and placenta in pregnancies with gestational diabetes mellitus
- Fetomaternal outcome of scarred uterine rupture compared with primary uterine rupture: a retrospective cohort study
- Original Articles – Fetus
- The assessment of fetal cardiac functions in pregnancies with autoimmune diseases: a prospective case-control study
- The relationship of maternal polymorphisms of genes related to meiosis and DNA damage repair with fetal chromosomal stability
- Original Articles – Neonates
- German obstetrician’s self-reported attitudes and handling in threatening preterm birth at the limits of viability
- Do parents get what they want during bad news delivery in NICU?
