Cardiovascular disease (CVD) is one of the foremost reasons behind the death of people worldwide. Prevention and early diagnosis are the only ways to control its progression and onset. Thus, there is an urgent need for a detection model comprising intelligent technologies, including Machine Learning (ML) and deep learning, to predict the future state of an individual suffering from cardiovascular disease by effectively analyzing patient data. This study aims to propose a hybrid model that provides a deep insight into the data under consideration to enhance model accuracy for effectively detecting cardiovascular disease. This current research proposes a hybrid model comprising four stages. In the first stage of the proposed hybrid model, the data imbalance problem is solved using a hybrid sampling technique named Synthetic Minority Oversampling Technique-Edited Nearest Neighbors Rule. In the second stage, the Chi-square is applied as a feature selection method to select the highly relevant features from the records of 1190 with 11 clinical features, curated by combining the 5 most popular datasets, including Long Beach VA, Hungarian, Switzerland, and Statlog (Heart). In the third stage, the preprocessed dataset is passed to a stacking ensemble model comprising three base learners: Random Forest Tree (RFT), K-Nearest Neighbor (K-NN), and AdaBoost classifier and one meta-learner: Logistic Regression (LR), optimized with Grid Search Cross-Validation (GSCV) optimization approach, whose performance is evaluated against individual classifier. In the fourth stage, the performance is evaluated in terms of accuracy, sensitivity, specificity, F1 score, and ROC_AUC score.. The comparative results prove that the proposed hybrid model scored the highest accuracy of 97.8%, 96.15% sensitivity, and 96.75% specificity and 98.6% ROC_AUC score when compared with the existing techniques and models after applying the SMOTE–ENN (for data balancing) and Chi-square (for feature selection) methods for the efficient detection of cardiovascular disease. The implementation results demonstrate that the suggested hybrid model may accurately identify cardiovascular disease among patients. It facilitates the application of robust clinical treatment strategies.

Freezing of gait (FoG) is a common and debilitating symptom of Parkinson’s disease (PD), and its early detection is crucial for timely intervention and improved patient outcomes. Traditional approaches often rely on manually extracted gait features, which may overlook subtle yet clinically significant patterns. This study presents a hybrid model that integrates manual gait feature extraction with deep learning-based analysis using a 3D-ResNext model to enhance the detection of FoG. In the first stage, both manual and deep learning features are extracted, and in the second stage, these features are fed into a long short-term memory (LSTM) network to classify gait patterns into walk, Pre-FoG, and FoG. The model was evaluated using publicly available Daphnet dataset, which comprises gait data from ten PD patients, eight of whom exhibited FoG events. Experimental analysis with varying segment lengths and Pre-FoG periods revealed that the optimal performance was achieved with a 1-s segment length and a 5-s Pre-FoG window. The proposed hybrid model achieved an accuracy of 98.87%, specificity of 96.67%, F1-score of 92.29%, and a loss value of 0.0597. These results demonstrate that the proposed approach outperforms existing state-of-the-art models and offers a robust solution for accurate FoG classification and PD prediction.