Authors : O. D. Salunkhe; P. H. Mahadik

Volume/Issue : Volume 9 - 2024, Issue 11 - November

Google Scholar : https://tinyurl.com/4vjp6m4v

DOI : https://doi.org/10.5281/zenodo.14442787

Abstract : Deaths in incubator is a major problem worldwide which is the leading cause of death in infants. Early diagnosis of infant death in an incubator helps to improve the survival chances of neonatal babies in an infant incubator. Early diagnosis of death in an incubator can save the infant or treat him/her better than standard treatment. In this paper we proposed a different ML algorithm to predict infant death in an incubator. This research was carried out in the Satara district, Maharashtra. In this study, we have taken factors that were identified by medical professionals for infant incubation. The machine learning algorithms can help medical professionals to predict infant deaths in an incubator as well as identify factors that cause the death of the infant in the incubator. The various data imbalancing techniques, such as the synthetic oversampling technique (SMOTE) and adaptive synthetic (ADYSN) have been implemented to improve the performance of models. The XG Boost classifier (F1 score = 0.88) and Random Forest classifier (F1 score = 0.89) with ADYSN give us the better performance than other classifiers.

Keywords : Infant Deaths, Incubator, Machine Learning Algorithm, Imbalanced Data.

References :

1. https://www.who.int/health-topics/newborn-health#tab=tab_1
2. Horbar JD, Edwards EM, Greenberg LT, et al. Variation in Performance of Neonatal Intensive Care Units in the United States. JAMA Pediatr. 2017;171(3):e164396. doi:10.1001/jamapediatrics.2016.4396
3. Haibo He, Yang Bai, E. A. Garcia and Shutao Li, "ADASYN: Adaptive synthetic sampling approach for imbalanced learning," 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), Hong Kong, 2008, pp. 1322-1328, doi: 10.1109/IJCNN.2008.4633969.
4. H. A. Gameng, B. B. Gerardo and R. P. Medina, "Modified Adaptive Synthetic SMOTE to Improve Classification Performance in Imbalanced Datasets," 2019 IEEE 6th International Conference on Engineering Technologies and Applied Sciences (ICETAS), Kuala Lumpur, Malaysia, 2019, pp. 1-5, doi: 10.1109/ICETAS48360.2019.9117287
5. Saroj, R.K., Yadav, P.K., Singh, R. et al. Machine Learning Algorithms for understanding the determinants of under-five Mortality. BioData Mining 15, 20 (2022).
6. I. Hingorani, R. Khara, D. Pomendkar and N. Raul, "Police Complaint Management System using Blockchain Technology," 2020 3rd International Conference on Intelligent Sustainable Systems (ICISS), Thoothukudi, India, 2020, pp. 1214-1219, doi: 10.1109/ICISS49785.2020.9315884.
7. Chawla, Nitesh & Bowyer, Kevin & Hall, Lawrence & Kegelmeyer, W.. (2002). SMOTE: Synthetic Minority Over-sampling Technique. J. Artif. Intell. Res. (JAIR). 16. 321-357. 10.1613/jair.953.
8. https://medium.com/@nandiniverma78988/understanding-k-nearest-neighbors-knn-regression-in-machine-learning-c751a7cf516c