Affiliation:
1. Department of Computer Engineering, College of Computer and Information Sciences King Saud University Riyadh Saudi Arabia
2. Department of Electronics Engineering Sejong University Seoul South Korea
Abstract
ABSTRACTThe Internet of Medical Things (IoMT) has revolutionized healthcare, particularly in ambient assisted living (AAL). Deep learning has emerged as a powerful tool for identifying disorders and making health‐related decisions. Pneumonia, a dangerous and contagious disease, has a significant global impact. Prompt and accurate diagnosis is crucial, but traditional methods are time‐consuming and require specialized expertise. This research introduces PneumNet, a novel deep‐learning model. PneumNet consists of two versions: PneumNet v1.0 and PneumNet v2.0. The comparative analysis demonstrates PneumNet's exceptional performance. The top model achieves 99.84% accuracy, 99.87% F1‐score, 99.74% sensitivity, 100% specificity, 100% positive predictive value (PPV), and 99.58% negative predictive value (NPV). PneumNet outperforms other methods, accurately diagnosing pneumonia and improving treatment outcomes. By leveraging deep convolutional neural networks (D‐CNNs), PneumNet provides an efficient and accurate solution for pneumonia detection. These findings highlight the significance of D‐CNNs, particularly the proposed PneumNet model, in enhancing pneumonia detection accuracy and reducing mortality rates. IoMT and deep learning pave the way for transformative advancements in healthcare.