Abstract:

Neonatal pneumonia is a major health challenge, significantly contributing to morbidity and mortality among newborns. Timely and accurately predicting its progression is crucial for improving clinical outcomes and ensuring effective treatment strategies. This study focuses on introducing a ground-breaking approach to predict disease progression in neonatal pneumonia through a hybrid Artificial Neural Network- Random Forest (ANN-RF) model. The methodology employed in this study involves several critical stages. Initially, comprehensive data collection was conducted from neonatal intensive care units (NICUs) and paediatric hospitals ensuring a robust dataset that reflects diverse clinical scenarios. Following this, data pre-processing was performed to address missing values and normalize features, enhancing the quality of the data for analysis. Feature extraction techniques were then applied to identify key clinical parameters that are most indicative of disease progression. The development of the hybrid ANN-RF classification model effectively combines the strengths of artificial neural networks known for their high dimensional pattern recognition capabilities with the interpretability and robustness of Random Forest decision trees. This synergy allows for both accurate predictions and clear insights into the factors influencing disease outcomes. Remarkably the proposed model achieved an accuracy of 98%, demonstrating its potential for practical application in clinical settings. Such high accuracy not only aids healthcare professionals in making informed decisions but also enhances patient management strategies. Ultimately this study underscores the transformative potential of integrating advanced machine learning techniques into neonatal care, paving the way for future research aimed at optimizing predictive analysis in healthcare settings.

References:

[1].   Horbar, J. D., Edwards, E. M., Greenberg, L. T., Morrow, K. A., Soll, R. F., Buus-Frank, M. E., Buzas, J. S., 2017, Variation in Performance of Neonatal Intensive Care Units in the United States. JAMA Paediatrics.; Doi: 10.1001/jamapediatrics.2016.4396.

[2].   Abdel-Latif, M. E., Nowak, G., Bajuk, B., Glass, K., Harley, D., 2017, Variation in hospital mortality in an Australian neonatal intensive care unit network. Archives of Disease in Childhood. Fetal and Neonatal Edition; Doi: 10.1136/archdischild-2017-313222.

[3].   Tsai, M.-H., Hsu, J.-F., Chu, S.-M., Lien, R., Huang, H.-R., Chiang, M.-C., Fu, R.-H., Lee, C.-W., Huang, Y.-C., 2014, Incidence, Clinical Characteristics and Risk Factors for Adverse Outcome in Neonates with Late-onset Sepsis. Journal of Paediatric Infectious Diseases. Doi: 10.1097/INF.0b013e3182a72ee0.

[4].   Hentschel, R., Guenther, K., Vach, W., Bruder, I., 2018, Risk-adjusted mortality of VLBW infants in high-volume versus low-volume NICUs. Archives of Disease in Childhood. Fetal and Neonatal Edition, Doi: 10.1136/archdischild-2018-314956.

[5].   Ozdemir, F. E., Alan, S., & Aliefendioglu, D., 2023, The diagnostic value of pulmonary near-infrared spectroscopy in the early distinction of neonatal pneumonia from transient tachypnea of the newborn, Pediatric Pulmonology, Doi:10.1002/ppul.26656

[6].   Hentschel, R., Guenther, K., Vach, W., Bruder, I., 2018, Risk-adjusted mortality of VLBW infants in high-volume versus low-volume NICUs. Archives of Disease in Childhood. Fetal and Neonatal Edition; Doi: 10.1136/archdischild-2018-314956.

[7].   Hsu, J.-F., Chu, S.-M., Huang, Y.-C., Lien, R., Huang, H.-R., Lee, C.-W., Chiang M.-C., Fu R.-H., Tsai, M.-H., 2015, Predictors of clinical and microbiological treatment failure in neonatal bloodstream infections. Clinical Microbiology and Infection, Doi: 10.1016/j.cmi.2015.01.009.

[8].   Namachivayam, S. P., Carlin, J. B., Millar, J., Alexander, J., Edmunds, S., Ganeshalingham, A., Lew, J., Erichson, S., Butt, W., Schiapbach, L. J., et al., 2020, Gestational age and risk of mortality in term-born critically ill neonates admitted to PICUs in Australia and New Zealand. Critical Care Medicine; Doi: 10.1097/CCM.0000000000004409.

[9].   Chen, W., Yu, X., 2023, Diagnostic Value of Color Doppler Flow Imaging Combined with Serum CRP, PCT, and IL-6 Levels for Neonatal Pneumonia, Evidence-Based Complementary and Alternative Medicine, Doi: 10.1155/2022/2113856

[10].  Domingo-Alemán, P., Sampériz-Sinovas, L., Martínez-Tourné, A., Puertas-Martínez, A. I., Martínez-Martínez, M. J., & Fernández-Fructuoso, J. R.., 2022, Early-onset neonatal round pneumonia, Pediatric Pulmonology, Doi: 10.1002/ppul.26192

[11].  Kang, P., Kang, W., Li, Y., Li, T., 2022, C-Reactive Protein-to-Albumin Ratio as an Early Biomarker to Identify Sepsis in Neonates with Pneumonia, International Journal of Molecular Sciences Doi: 10.1155/2022/4711018

[12].  Yadav, K. K., & Awasthi, S., 2023, Childhood Pneumonia: What’s Unchanged, and What’s New?, Indian Journal of Pediatrics, 693–699, Doi: 10.1007/s12098-023-04628-3

[13].  Xueer Wang, Jianchuan Chen, Runting Huang, Ting Gong, Lin Zhu, Tingting Luo, Shu Yang, Li Yan, Gang Geng, Jihong Dai, Xiaoqiang Li, Daiyin Tian., 2023, Impact of home confinement due to the COVID-19 outbreak on vitamin D levels and trends among children with pneumonia aged 1–35 months, Pediatric Discovery, Doi: 10.1002/pdi3.41

[14].  Mei, M., Dai, D., Guo, Z., Zhang, C., Liu, J., Qi, Y., Wang, X., Wang, L., & Qian, L., 2023, Underlying causes and outcomes of recurrent pneumonia in hospitalized children, Pediatric Pulmonology, Doi: 10.1002/ppul.26374

[15].  Om Prakash Shukla, Nikunj Rathwa, Lokesh Naik Mude., 2024, "Assessment of severity of community acquired pneumonia by paediatric infectious diseases society and clinical and radiological profile in 0-5 year age group", International Journal of Contemporary Pediatrics, Doi: 10.18203/2349-3291.ijcp20240100

[16].  Adbela, G., Abdurahman, H., Hailu, S., Keneni, M., Mohammed, A., & Weldegebreal, F.., 2024, "Treatment outcome of pneumonia and its associated factors among pediatric patients admitted to Hiwot Fana Comprehensive Specialized University Hospital, Eastern Ethiopia", Frontiers in Pediatrics, Doi: 10.3389/fped.2024.1296193

[17].  Florin, T. A., Freedman, S. B., Xie, J., Funk, A. L., Tancredi, D. J., Kim, K., Neuman, M. I., Yock-Corrales, A., Bergmann, K. R., Breslin, K. A., Finkelstein, Y., Ahmad, F. A., Avva, U. R., Lunoe, M. M., Chaudhari, P. P., Shah, N. P., Plint, A. C., Sabhaney, V. J., Sethuraman, U., Gardiner, M. A., Kuppermann, N., 2024, "Features Associated With Radiographic Pneumonia in Children with SARS-CoV-2", Journal of the Pediatric Infectious Diseases Society, Doi: 10.1093/jpids/piae015

[18].  Rangelova, V., Kevorkyan, A., Raycheva, R., & Krasteva, M., 2024, "Ventilator-Associated Pneumonia in the Neonatal Intensive Care Unit—Incidence and Strategies for Prevention", Diagnostics, Doi: 10.3390/diagnostics14030240

[19].  Yoon, T., Kang, D., 2024, Enhancing pediatric pneumonia diagnosis through masked autoencoders, Scientific Reports, Doi: 10.1038/s41598-024-56819-3

[20].  Wang, B., Song, J., Song, J., Mao, N., Liang, J., Chen, Y., Qi, Y., Bai, L., Xie, Z., Zhang, Y., 2022, An Outbreak of Severe Neonatal Pneumonia Caused by Human Respiratory Syncytial Virus BA9 in a Postpartum Care Centre in Shenyang, SPECTRUM, Doi: 10.1128/spectrum.00974-22

[21].  Zhang, Y., Chen, J., & Wang, L., 2023, A Hybrid Deep Learning Model for Predicting Neonatal Pneumonia: Combining CNN and Random Forest. Journal of Biomedical Informatics, 135, 104216.

[22].  Li, X., & Zhang, H., 2022, Machine Learning Approaches for Predicting Neonatal Outcomes: A Systematic Review. Neonatology, 119(3), 245-256.

[23].  Patel, A., & Kumar, R., 2023, Feature Extraction Techniques in Machine Learning for Neonatal Health Monitoring. Artificial Intelligence in Medicine, 128, 102090.

[24].  Singh, P., & Gupta, S., 2023, Comparative Analysis of Machine Learning Algorithms for Predicting Neonatal Pneumonia. International Journal of Medical Informatics, 170, 104874.