Artificial Intelligence Integrated Fetal Cardiotocography Education Module

Abstract

Introduction: Cardiotocography is an indispensable instrument for assessing fetal well-being at the antenatal period. The interpretation of cardiotocography requires substantial background information, knowledge and experience. In particular, inexperienced clinicians and midwives are easily prone to making mistakes during the evaluation of cardiotocography and making wrong decisions. The aim of this project is to develop a new artificial intelligence integrated interface to aid teaching the interpretation of fetal cardiotocography. Methods: The parameters of presence of uterine contractions, beat-to-beat variability, fetal heart rate and periodic changes (accelerations and decelerations) of 118 scanned fetal cardiotocographies acquired from Niğde Ömer Halisdemir University, Niğde Research and Training Hospital Obstetrics and Gynecology clinic were classified by 2 experienced obstetrics and gynecology specialists and were uploaded to the interface. Convolutional neural network (CNN) deep learning architecture was used and cardiotocography classification model was generated. Results: The developed interface consisted of 4 sections. First contained the basic information. Educational information for interpretation of cardiotocography was uploaded to the second section. The third section was designed for students’ self-training with randomly selected and previously classified cardiotocographies. The fourth section consisted of a deep learning based trained cardiotocography classification model, for uploading a test sample and generating classification result automatically. The performance of CNN module on the classification of cardiotocography dataset was 84%. Conclusion: Artificial intelligence integrated fetal cardiotocography educational interface was successfully developed. We believe applied utilization of the interface would provide great benefit to fetal cardiotocography education.

Keywords

• Cardiotocography
• artificial intelligence
• convolutional neural network
• interface

References

1. 1. Al-yousif S, Jaenul A, Al-dayyeni W, et al (2021) A systematic review of automated pre- processing , feature extraction and classification of cardiotocography. PeerJ Comput Sci. https://doi.org/10.7717/peerj-cs.452
2. 2. Visser GH, Ayres-de-campos D, Intrapartum F, et al (2015) International Journal of Gynecology and Obstetrics FIGO GUIDELINES FIGO consensus guidelines on intrapartum fetal monitoring : Adjunctive technologies ☆ , ★. 131:25–29
3. 3. Ayres-de-campos D, Arulkumaran S (2015) International Journal of Gynecology and Obstetrics FIGO GUIDELINES FIGO consensus guidelines on intrapartum fetal monitoring : Physiology of fetal oxygenation and the main goals of intrapartum fetal monitoring ☆ , ★. 131:5–8
4. 4. Castro L, Loureiro M, Henriques TS, Nunes I (2021) Systematic Review of Intrapartum Fetal Heart Rate Spectral Analysis and an Application in the Detection of Fetal Acidemia. 9:1–12. https://doi.org/10.3389/fped.2021.661400
5. 5. Anderson A (2013) Healthcare and Law Digest. AVMA Med Leg J 19:24–31. https://doi.org/10.1177/1356262213486434
6. 6. Royal College of Obstetricians & Gynaecologists (2021) 2020 final progress report
7. 7. Zhao Z, Deng Y, Zhang Y, et al (2019) DeepFHR : intelligent prediction of fetal Acidemia using fetal heart rate signals based on convolutional neural network. BMC Med Inform Decis Mak 5:1–15
8. 8. Beckley S, Stenhouse E, Greene K (2000) The development and evaluation of a computer-assisted teaching programme for intrapartum fetal monitoring. Br J Obstet Gynaecol 107:1138–1144

9. 9. Thellesen L, Bergholt T, Hedegaard M, et al (2017) Development of a written assessment for a national interprofessional cardiotocography education program. BMC Med Educ 17:1–9. https://doi.org/10.1186/s12909-017-0915-2
10. 10. Thellesen L, Hedegaard M, Bergholt T, et al (2015) Curriculum development for a national cardiotocography education program : a Delphi survey to obtain consensus on learning objectives. Acta Obstet Gynecol Scand 94:869–877. https://doi.org/10.1111/aogs.12662
11. 11. Bochkovskiy A, Wang C-Y, Liao H-YM (2020) YOLOv4: Optimal Speed and Accuracy of Object Detection. Arxiv 10934:. https://doi.org/10.48550/arxiv.2004.10934
12. 12. Yu J, Zhang W (2021) Face mask wearing detection algorithm based on improved YOLO-v4. Sensors 21:3263
13. 13. Kumar A, Kalia A, Sharma A, Kaushal M (2021) A hybrid tiny YOLO v4-SPP module based improved face mask detection vision system. J Ambient Intell Humaniz Comput 1–14. https://doi.org/10.1007/S12652-021-03541-X
14. 14. Wu D, Lv S, Jiang M, Song H (2020) Using channel pruning-based YOLO v4 deep learning algorithm for the real-time and accurate detection of apple flowers in natural environments. Comput Electron Agric 178:105742
15. 15. Dewi C, Chen R, Liu Y, et al (2021) Yolo v4 for advanced traffic sign recognition with synthetic training data generated by various gan. IEEE Access 9:97228–97242
16. 16. Mulyanto A, Borman R, Prasetyawan P, et al (2020) Indonesian Traffic Sign Recognition For Advanced Driver Assistent (ADAS) Using YOLOv4. In: 3rd International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)
17. 17. Albahli S, Nida N, Irtaza A, et al (2020) Melanoma lesion detection and segmentation using YOLOv4-DarkNet and active contour. IEEE Access 8:198403–198414
18. 18. Kolchev A, Pasynkov D, Egoshin I, et al (2022) YOLOv4-Based CNN Model versus Nested Contours Algorithm in the Suspicious Lesion Detection on the Mammography Image: A Direct Comparison in the Real. J Imaging 8:88
19. 19. Cömert Z (2016) Evaluation of Fetal Distress Diagnosis during Delivery Stages based on Linear and Nonlinear Features of Fetal Heart Rate for Neural Network Community. 156:26–31
20. 20. Sullivan MEO, Considine EC, Riordan MO, et al (2021) Challenges of Developing Robust AI for Intrapartum Fetal Heart Rate Monitoring. Front Artif Intell 4:1–8. https://doi.org/10.3389/frai.2021.765210
21. 21. Ponsiglione A, Cosentino C, Cesarelli G, et al (2021) A Comprehensive Review of Techniques for Processing and Analyzing Fetal Heart Rate Signals. Sensors 21:1–31
22. 22. Carbonne B, Sabri-Kaci I (2016) European Journal of Obstetrics & Gynecology and Reproductive Biology Assessment of an e-learning training program for cardiotocography analysis : a multicentre randomized study. Eur J Obstet Gynecol Reprod Biol 197:111–115. https://doi.org/10.1016/j.ejogrb.2015.12.001
23. 23. Gyllencreutz E, Varli I, Lindqvist PG, Holzmann M (2017) Reliability in cardiotocography interpretation – impact of extended on-site education in addition to web-based learning : an observational study. Acta Obstet Gynecol Scand 96:496–502. https://doi.org/10.1111/aogs.13090
24. 24. Pehrson C, Sorensen J, Amer-Wahlin I (2011) Evaluation and impact of cardiotocography training programmes : a systematic review. Br J Obstet Gynaecol 118:926–935. https://doi.org/10.1111/j.1471-0528.2011.03021.x
25. 25. Beksaç MS, Özdemir K, Karakaş Ü, et al (1990) Development and application of a simple expert system for the interpretation of the antepartum fetal heart rate tracings (version 88/2.29). Eur J Obstet Gynecol Reprod Biol 37:133–141. https://doi.org/10.1016/0028-2243(90)90106-B
26. 26. Sbrollini A, Agostinelli A, Marcantoni I, et al (2018) Computer Methods and Programs in Biomedicine eCTG : an automatic procedure to extract digital cardiotocographic signals from digital images R. Comput Methods Programs Biomed 156:133–139. https://doi.org/10.1016/j.cmpb.2017.12.030
27. 27. Guijarro-Berdias B, Alonso-Betanzos A, Fontenla-Romero O (2002) Intelligent analysis and pattern recognition in cardiotocographic signals using a tightly coupled hybrid system. Artif Intell 136:1–27. https://doi.org/10.1016/S0004-3702(01)00163-1
28. 28. Tang H, Wang T, Li M, Yang X (2018) The Design and Implementation of Cardiotocography Signals Classification Algorithm Based on Neural Network. Comput Math Methods Med 2018
29. 29. Nazari S, Hatami E, Tabatabayeechehr M, et al (2018) Diagnostic Value of Non-stress Test Interpreted by Smart Interpretive Software. J Midwifery Reprod Heal 6:1384–1389. https://doi.org/10.22038/jmrh.2018.21461.1228
30. 30. Petrozziello A, Redman CWG, Papageorghiou AT, et al (2019) Multimodal Convolutional Neural Networks to Detect Fetal Compromise During Labor and Delivery. IEEE Access 7:112026–112036. https://doi.org/10.1109/ACCESS.2019.2933368
31. 31. Weindling AM (2004) The confidential enquiry into maternal and child health (CEMACH). Arch Dis Child 89:1034–1037
32. 32. Thellesen L, Bergholt T, Led J, et al (2019) The impact of a national cardiotocography education program on neonatal and maternal outcomes : A historical cohort study. Acta Obstet Gynecol Scand 98:1258–1267. https://doi.org/10.1111/aogs.13666
33. 33. Young P, Hamilton R, Hodgett S, et al (2001) Reducing risk by improving standards of intrapartum fetal care. J R Soc Med 94:226–231. https://doi.org/10.1177/014107680109400507
34. 34. Nunes I, Ayres-de-campos D (2016) Computer analysis of foetal monitoring signals. Best Pract Res Clin Obstet Gynaecol 30:68–78. https://doi.org/10.1016/j.bpobgyn.2015.02.009
35. 35. Nandipati S, Ying C (2019) Classification and Feature Selection Approaches by Machine Learning Techniques: Heart Disease Prediction. Int J Innov Comput 9:7–14. https://doi.org/10.11113/ijic.v9n1.210