A Predictive Analytical Model for Sickle Cell Anemia using Hybrid SVM and Random Forest Algorithms
Naredla Laharika, Hameed Hassan Khalaf, Maki Mahdi Abdulhasan, Wael Dheaa Kadhim, D. Naresh, Yuvraj Potulgude, B. Akshitha · 2024
A genetic condition known as sickle cell anemia is typified by aberrant hemoglobin, which results in malformed red blood cells that can lead to several health issues. For sickle cell anemia to be effectively managed and treated, early detection and monitoring are essential. The "Sickle Anemia Detector" research uses clinical data and analysis techniques to create a machine learning-based tool that can identify and predict sickle cell anemia. Algorithms, like SVM, random forest and the research uses a Decision Tree Classifier to train a prediction model by utilizing a dataset that includes data on hemoglobin levels, sex, and pixel values that indicate blood cell properties. The research workflow consists of multiple important components. First, the data is loaded and goes through preprocessing steps like target assignment and feature extraction. To guarantee representative samples in each subset, the dataset is next split using a stratified split into training and testing sets. After that, the model is trained using the training set of data, fine-tuning its parameters to yield the best prediction accuracy. Training and testing data are used to evaluate the model's performance and gauge how well it generalizes. With the use of an intuitive stream-lit interface, users may interactively enter data and receive predictions for the existence of sickle cell anemia with this newly designed technology. To promote openness and confidence in the forecasts, users are shown the model's performance indicators, which include training and testing results. All things considered, the Sickle Anemia Detector research advances medical technology by providing a dependable and easily accessible method for the early diagnosis and tracking of sickle cell anemia, which may enhance patient outcomes by allowing prompt intervention and treatment.