A deep learning and machine learning approach to predict neonatal death in the context of São Paulo

2024-03-01

International Journal of Public Health Science (IJPHS)

AI Summary

This study addresses neonatal mortality risk prediction in the São Paulo region using a large-scale, real-world birth dataset comprising 1.4 million records. To enable early identification of high-risk newborns, we establish a multi-model comparative framework and—novelty for this region—introduce Long Short-Term Memory (LSTM) networks for neonatal mortality prediction, overcoming performance limitations of conventional machine learning approaches. Experimental results demonstrate that the LSTM model achieves 99% accuracy, substantially outperforming baseline models including logistic regression, k-nearest neighbors (KNN), random forest (94%), XGBoost (94%), and convolutional neural networks (CNN). The findings empirically validate the efficacy of temporal modeling for perinatal risk prediction and yield a clinically deployable, high-accuracy early-warning system. This work establishes a new paradigm for neonatal mortality intervention in resource-constrained settings.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Large language models for search
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Neonatal death is still a concerning reality for underdeveloped and even for some of the developed countries. Worldwide data indicate that 26.693 babies out of 1,000 births according to Macro Trades. To reduce the death early prediction of endangered baby is crucial. An early prediction enables the opportunity to take ample care of the child and mother so that an early child death can be avoided. Machine learning was used to figure out whether a newborn baby is at risk. To train the predictive model historical data of 1.4 million newborn child data was used. Machine learning and deep learning techniques such as Logical regression, K nearest neighbor, Random Forest classifier, Extreme gradient boosting (XGboost), convolutional neural network, long short-term memory (LSTM). were implemented using the dataset to find out the most robust model which model is the most accurate to identify the mortality of a newborn. From all the machine learning algorithms, the XGboost and random classifier had the best accuracy with 94%, and from the deep learning model, the LSTM had the best outcome with 99% accuracy. Thus, using LSTM of the model shall be best suited to predict whether precaution for a child is necessary.

Problem

• Predict neonatal death using machine learning
• Identify at-risk newborns with high accuracy
• Compare ML and DL models for mortality prediction

Innovation

• Used deep learning and machine learning techniques
• Trained model with 1.4 million newborns data
• LSTM achieved highest accuracy with 99%