<![CDATA[This study develops a pulmonary edema detection model based on chest x-ray images using both classical Convolutional Neural Network (CNN) and Quantum Convolutional Neural Network (QCNN) approaches. The dataset consists of chest x-ray images labeled as positive and negative for pulmonary edema and is divided into training and testing sets with an 80:20 ratio. To obtain the best performance, both models were optimized through hyperparameter tuning. The classical CNN model employed 3×3 filters, four convolutional layers, and was trained for 10 epochs. The QCNN model was designed with a comparable architecture incorporating quantum gate modifications and was also trained for 10 epochs. The performances of these models were assessed based on accuracy, sensitivity, and precision in order to measure their classification capacities. The QCNN was developed under the same experimental conditions to enable a fair performance comparison with the optimized classical CNN model. The results show that the classical CNN achieved better performance than the QCNN. This lower QCNN performance is likely due to the current limitations of quantum architectures and hardware, which are not yet able to extract and process image features as effectively as classical CNNs. However, this study was conducted using low-resolution medical images and a preliminary QCNN framework under limited computational resources.]]>
Copyrights © 2026
