Year: 2024
Type: article
Abstract: Forecasting fetal health is essential for spotting issues throughout pregnancy. However, complicated data is often a challenge for conventional statistical techniques to cope with. This study used a publicly accessible fetal health dataset and optimized the dataset through preprocessing steps such as removing outliers, selecting relevant features, and normalizing the data. Subsequently, four models, including random forest (RF), support vector machine (SVM), lasso logistic regression (LLR), and neural network (NN), were selected to train the data. The results showed that the RF model performed best with an accuracy of 0.9494. The NN model performed the worst with an accuracy of 0.8752. At the same time, the results showed that RF can handle small datasets, provide feature importance, and resist overfitting. However, all models in this study have inherent limitations. Future research should focus on combining multiple models and using larger and more diverse datasets to further improve the prediction performance and effectiveness of clinical applications.
DOI: 10.54097/9az7vk94
Author: Shuqi Wu
Institution: King's College London
Language: English
Subfield: Health Information Management
Field: Health Professions
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4405810884