High Accuracy Diagnosis of Pediatric Pneumonia: Achieved Using Convolutional Neural Network

The objective of this research is to employ Convolutional Neural Network (CNN) models for the purpose of diagnosing pneumonia in children, with varying convolutional blocks and training epochs used. The performance of six different CNN models is evaluated using various metrics, including Accuracy, Recall, Precision, F1score, and Area Under the Curve (AUC). Based on these measures, the best-performing model among the six models assessed is determined. Using Jupyter Notebook as the main tool and programing platform, a best model with an accuracy of 93.59% was achieved. A comparison study with other comparable research in the literature is performed to acquire more insights into the performance of the best model. The models used in other research are examined, and their accuracy ratings are compared to the best model from this study. This examination gives a thorough grasp of the suggested CNN model's diagnostic capabilities in comparison to previous techniques. Finally, this study demonstrates the use of CNN models in the diagnosis of pediatric pneumonia. The comparison of multiple models and benchmarking against other similar research improves knowledge of the diagnostic performance obtained. The findings advance the area of pediatric pneumonia diagnosis and can be used to guide future research and therapeutic applications.