PneuXNet An Enhanced Feature Extraction and Transformation Approach for Pneumonia Detection

The rise of respiratory diseases such as Pneumonia, COVID-19, and Tuberculosis highlights the need for advanced, accurate diagnostic systems. This paper presents Hybrid Quantum, a novel hybrid deep learning framework designed for multi-class medical image classification and severity assessment. The system classifies chest X-ray images into four categories: Normal, Pneumonia, COVID-19, and Tuberculosis. The model combines Convolutional Neural Networks (CNN) for efficient feature extraction with Quantum Support Vector Machine (QSVM) for classification. CNN excels in automatically extracting relevant spatial features from X-ray images, such as lung structures and disease markers, using convolutional and pooling layers. QSVM is employed to perform the final classification, leveraging quantum computing principles to improve decision boundaries and speed up convergence compared to traditional SVMs. The framework also incorporates severity assessment by generating a confidence score for each classification, helping clinicians assess the certainty of the predictions. Furthermore, Explainable Quantum AI (XQAI) is integrated to enhance transparency and interpretability, allowing healthcare professionals to trust and understand the model's decision-making process. Initial results from testing on a diverse dataset of chest X-rays demonstrate promising classification accuracy and computational efficiency. This hybrid model presents significant potential for real-time clinical decision support and paves the way for future advancements in AI and quantum computing integration for healthcare applications.

Keywords: Multi-class classification, Convolutional Neural Networks (CNN), Quantum Support Vector Machine (QSVM), Explainable Quantum AI (XQAI), Pneumonia, COVID-19, Tuberculosis, Chest X-ray, Severity assessment, Prognosis prediction, Deep learning, Quantum computing, Medical image classification, AI in healthcare, Quantum-enhanced classification, Diagnostic systems.