Prediction of maternal health risk levels based on multiple machine learning algorithms

Abstract

With the development and popularisation of the Internet of Things (IoT) technology, more and more medical devices have begun to connect with the Internet, forming IoT medical devices. Among them, for pregnant women's health risk monitoring, IoT devices can achieve remote monitoring, data transmission, intelligent analysis and other functions to provide more comprehensive health management services for pregnant women. To evaluate the model's effectiveness in predicting the health risk level of pregnant women, we use four metrics: precision, accuracy, recall, and F1 score. Where precision indicates the proportion of correctly predicted samples to the total number of samples; accuracy indicates the proportion of true cases to the number of all samples predicted to be positive; recall indicates the proportion of true cases to the number of all samples that are actually positive; and F1 score is the reconciled average of precision and recall. By using these metrics, we can more fully assess the predictive effectiveness of the model. In this paper, we use three models, decision tree, random forest and logistic regression, for training and testing. All three models are common machine learning algorithms, but they each have different strengths and weaknesses. By training and testing the three models, we found that the random forest model performed the best, with a prediction accuracy of 80%. In conclusion, this paper describes an IoT-based health risk monitoring system for pregnant women, trained and tested using three models: decision tree, random forest and logistic regression, it can provide more comprehensive health management services for pregnant women.