Enhancing Heart Disease Diagnosis Through Particle Swarm Optimization and Ensemble Deep Learning Models

Advances in computer and electrical engineering book series, Journal Year: 2024, Volume and Issue: unknown, P. 313 - 330

Published: Dec. 6, 2024

The present research focused on combining Particle Swarm Optimization (PSO) based hybrid deep learning models to classify heart disease images and patient sequences. This study employs Convolutional Neural Networks (CNNs), including VGG 16, 19 ResNet 50, as well Recurrent Neu-ral (RNNs), whereby their performance is optimized by PSO im-prove the accuracy in diagnosing from CT together with associated medical history. experienced a significant increase classification performance, using manual hyperparameters tuning PSO. combined algorithm RNN model performed best, achieving precision of 97.78% becoming highest recall testing. that we propose uses modern feature extraction an take into consideration sequential nature data, making it very accurate while keeping loss minimal. 16 also another robust 94.5%.

Language: Английский

Combination of XGBoost - Grid Search with SVM for Diabetes Diagnostics

Published: Aug. 25, 2023

Our study focuses on using data mining methods, such as XGBoost, Hyperparameter Grid Search, and SVM, to diagnose diabetes. Diabetes is a degenerative disease characterized by metabolic disorders diabetic levels that exceed normal limits. In an effort find the most effective model in diagnosing this disease, uses available dataset divides it into training test with ratio of 80%:20%. Through testing, accuracy results obtained for XGBoost without Search were 87.0%, whereas implementation best was 98.5%. addition, SVM RBF kernel also evaluated provided 84.5%.

Language: Английский

Secure Health Management and Prediction System for Chronic Diseases

2022 IEEE 6th Conference on Information and Communication Technology (CICT), Journal Year: 2023, Volume and Issue: unknown, P. 1 - 6

Published: Dec. 15, 2023

The Secure Health Management and Prediction System for Chronic illnesses has transformed healthcare by predicting controlling chronic using user-provided data. It lowers expenses combining machine learning, data analytics, cloud computing. With strong safeguards such as encryption authentication, the system protects privacy security. Its HIPAA compliance focus on patient make it helpful in rural Sri Lanka, where experts are few. Using computing, research study proposes a secure health management prediction diseases. Based medical information, algorithm forecasts diabetes cardiovascular disorders. For disease prediction, employs techniques Decision Tree, Linear Regression, Random Forest, Support V ector Machine, Logistic Naive Bayes. feedback, also assesses insurance costs suggests providers.

Language: Английский