Analysis and Detection of Preeclampsia Using Machine Learning Techniques

Preeclampsia is a significant pregnancy complication marked by elevated blood pressure and the possibility of harm to various organs, posing risks to both the mother and the baby. Preeclampsia is a dangerous pregnancy complication that can harm other organs and raise blood pressure, endangering both the mother and the fetus. To improve outcomes for mothers and newborns, preeclampsia must be identified early and treated appropriately. This study proposes a hyperparameter-tuned K-Nearest Neighbors (KNN) classification model as a potential means of diagnosing preeclampsia. To effectively forecast the possibility of preeclampsia the model utilizes several clinical factors, such as blood pressure, body mass index (BMI), number of pregnancies, placental location, uterine artery pulsatility index (UTPI), and ultrasound imaging of the placenta. There are various important phases to this endeavor. First, a comprehensive dataset comprising patient records is preprocessed, standardizing continuous variables and encoding categorical variables. Next, to enhance the model's performance, adjustments to the KNN classifier are made based on research that methodically examines several hyperparameters like the number of neighbors, distance metrics, and weight functions. To enhance model performance, it's beneficial to utilize weight functions and various distance metrics. Metrics such as accuracy, precision, recall, F1 score, and the receiver operating characteristic (ROC) curve are employed to evaluate the enhanced model. In order to make real-time forecasts and enable healthcare practitioners to input patient data, a user-friendly interface has been built. Thorough testing is used to confirm the system's capacity to handle novel and unseen data, guaranteeing generalizability and reliability. With a reliable method for anticipating and treating this serious illness, this preeclampsia detection model seeks to promote early diagnosis, prompt intervention, and enhanced patient care.