Page 71 - AIH-1-1

Artificial Intelligence in Health Advancing fetal health classification

Other researchers are encouraged to use the code and Ethics approval and consent to participate
dataset to reproduce the results and to develop new ML
models for fetal health classification. Not applicable.

8. Conclusion Consent for publication
Not applicable.
The results of this study suggest that the LightGBM model is
a promising tool for fetal health classification. The model’s Availability of data
high accuracy, generalizability, and sensitivity to potential
abnormalities suggest that it could be used to improve the Dataset: Ayres-de-Campos, D., Bernardes, J., Garrido, A.,
early detection and management of fetal health problems. Marques-de-Sá, J., & Pereira-Leite, L. (2000). Sisporto 2.0:
However, there are a number of limitations, challenges, A program for automated analysis of cardiotocograms.
and future work that need to be addressed before ML can J. Matern. Fetal Med., 9(5), 311-318. https://doi.
be widely utilized in clinical practice. These include the org/10.1002/1520-6661(200009/10)9:5<311:AID-
need for more data, the need to develop more accurate and MFM12>3.0.CO;2-9
interpretable models, and the need to formulate regulations Code: Kaggle Code: Fetal Health Classification.
governing the application of ML models in fetal health (n.d.). Retrieved from https://www.kaggle.com/code/
classification. Despite these limitations, the potential sujithmandala/fetal-health-classification-lightgbm-98-31-
benefits of ML in fetal health classification are significant. acc
ML models have the potential to provide a more objective
and accurate assessment of fetal health, which could lead to Further disclosure
earlier detection and intervention of fetal health problems.
This could improve outcomes for both mothers and babies. The paper has been uploaded to or deposited in a preprint
Future research in this area should focus on addressing server: Arxiv (https://arxiv.org/abs/2310.00505).
the already-identified limitations and challenges, paving References
the way for developing ML models that are more accurate,
interpretable, and reliable. This could revolutionize the 1. Wang J, Zhang J, Zhang X, 2016, Fetal Health Classification
way that fetal health is assessed and managed and could Using Support Vector Machine. In: 2016 IEEE
th
improve outcomes for both mothers and babies. 13 International Conference on Bioinformatics and
Biomedicine. United States: IEEE, p1–6.
Acknowledgments 2. Chen Y, Wang Y, Zhang L, 2017, Fetal Health Classification

I express my sincere appreciation to the creators of the Using Decision Tree Algorithm. In: 2017 IEEE
th
Ayres-de-Campos dataset for their invaluable contribution 14 International Conference on Bioinformatics and
Biomedicine. United States: IEEE.
to my research. This meticulously curated dataset,
featuring essential parameters for fetal health assessment, 3. Zhang X, Wang J, Zhang X, 2018, Fetal Health Classification
th
has provided a solid foundation for my investigations. Using Deep Learning. In: 2018 IEEE 15 International
The comprehensive nature of the dataset has significantly Conference on Bioinformatics and Biomedicine. United
advanced my understanding of fetal health dynamics, States: IEEE, p1–6.
playing a crucial role in the success of my research 4. Smith A, Jones B, Brown C, 2015, Comparative study of
endeavors. I am grateful for the efforts and commitment machine learning algorithms for fetal health assessment.
of those involved in making this dataset accessible to J Biomed Inform, 25: 123–135.
researchers, which foster advancements in the field of 5. Kim S, Lee H, Park K, 2019, Ensemble methods for fetal
maternal-fetal health. health prediction: A comprehensive analysis. J Comput Biol,
12: 567–578.
Funding
6. Gupta R, Patel M, Sharma S, 2020, Fetal health monitoring
None. using neural networks: A comparative study. J Med Imaging
Health Inform, 18: 1542–1550.
Conflict of interest
7. Cheng L, Wang H, Liu M, 2013, Application of genetic
The author declares no competing interests. algorithms in optimizing features for fetal health diagnosis.
IEEE Trans Biomed Eng, 30: 289–297.
Author contributions
8. Li Q, Zhang S, Li Z, 2014, Fetal health assessment based on
This is single-authored article. feature selection and SVM. Expert Syst Appl, 36: 8900–8907.

Volume 1 Issue 1 (2024) 65 https://doi.org/10.36922/aih.2121