Enhanced Pneumonia Detection Using Ensembled Deep Neural Networks

In order to effectively treat pneumonia, which is still a major worldwide health problem, rapid and precise diagnosis is essential. This paper introduces an ensemble strategy to improve pneumonia identification using chest X-ray images (CXM), utilising developments in deep learning. We propose an Ensemble Deep Neural Networks (EDNN), comprising cascaded ShuffleNet and Support Vector Machines (SVM), to harness diverse features and improve classification performance. The ensemble method combines the strengths of multiple models, mitigating individual weaknesses and enhancing overall diagnostic accuracy. Implementation is carried out using Python, and the proposed approach achieves an impressive accuracy of 97.89% on benchmark datasets. Through extensive experimentation and validation on benchmark datasets, our approach demonstrates superior performance compared to individual models and existing state-of-the-art methods. Additionally, we provide insights into the interpretability of ensemble predictions, enhancing the transparency and trustworthiness of automated pneumonia detection systems. The proposed ensemble framework holds promise for robust and reliable pneumonia detection in clinical settings, facilitating timely interventions and improving patient outcomes.