Answers are grounded in this paper and cite their sources.
Accurate and timely prognosis is essential for effective patient management and improved healthcare outcomes. This study introduces a novel ensemble framework that integrates Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks, optimized using the Integer-Based Linear Hopper Optimization Algorithm (ILHOA). The model leverages CNN for spatial feature extraction and LSTM for capturing temporal patterns in clinical and laboratory data. ILHOA enhances model efficiency and robustness by selecting the most relevant features and eliminating redundancy. The proposed system includes data preprocessing, ILHOA-based feature selection, and a hybrid CNN-LSTM structure. Predictions from both deep learning models are combined through an ensemble process to boost generalization and reduce overfitting. The model’s performance is evaluated using key metrics including accuracy, precision, recall, F1-score, and AUC-ROC. Experimental results demonstrate the superiority of the ILHOA-optimized CNN-LSTM model over traditional machine learning and standalone deep learning approaches, offering high classification accuracy and reduced computational complexity. The hybrid architecture also improves interpretability, making it suitable for real-time clinical decision-making. Future work will focus on integrating attention mechanisms and validating the model with real-world patient datasets to enhance generalizability and expand its applicability across broader healthcare diagnostics.
No takes yet. Share an insight, caveat, or question.
Parthiban et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: