B2-Net: an artificial intelligence powered machine learning framework for the classification of pneumonia in chest x-ray images

Abstract

A chest X-ray radiograph is still the global standard for diagnosing pneumonia and helps distinguish between bacterial and viral pneumonia. Despite several studies, radiologists and physicians still have trouble correctly diagnosing and classifying pneumonia without false negatives. Modern mathematical modeling and artificial intelligence could help to reduce false-negative rates and improve diagnostic accuracy. This research aims to create a novel and efficient multiclass machine learning framework for analyzing and classifying chest X-ray images on a graphics processing unit. Researchers initially applied a geometric augmentation using a positional transformation function to the original dataset to enhance the sample size and aid future transfer learning. Models with the best accuracy, AUROC, F1 score, precision, recall, and specificity are chosen from a pool of nine state-of-the-art neural network models. The best-performing models are then retrained using an ensemble technique using depth-wise convolutions, demonstrating significant improvements over the baseline models employed in this research. With a remarkable 97.69% accuracy, 100% recall, and 0.9977 AUROC scores, the proposed B2-Net (Bek-Bas Network) model can differentiate between normal, bacterial, and viral pneumonia in chest X-ray images. A superior model is retrained using the chosen DenseNet-160, ResNet-121, and VGGNet-16 ensemble models. The diagnostic accuracy of the X-ray classification unit is enhanced by the newly designed multiclass network, the B2-Net model. The developed GPU-based framework has been examined and tested to the highest clinical standards. After extensive clinical testing, the final B2-Net model is implemented on an NVIDIA Jetson Nano graphics processing unit computer. Healthcare facilities have confirmed the B2-Net is the most effective framework for identifying bacterial and viral pneumonia in chest X-rays.

Joint Diagnosis of Pneumonia, COVID-19, and Tuberculosis from Chest X-ray Images: A Deep Learning Approach

“…Another celebrated study in [12] presented a machine learning framework for pneumonia detection from chest X-ray images comprising dense CNN-160, ResNet-121, and VGG-16 ensemble models. With 97.69% accuracy, 100% recall, and 0.9977 area under the curve scores, the scheme was promising for multivariate classification of normal, bacterial, and viral pneumonia in chest X-ray images.…”

Fighting against COVID‐19: Innovations and applications

“…An x‐ray radiograph is a medical imaging technique that uses ionizing electromagnetic radiation to create images of internal body structures 9 . The x‐ray radiation is absorbed differently by different tissues within the body, creating a contrast that can be captured on film or digitally 10 .…”

Whale Optimization Algorithm with Fuzzy Wavelet Neural Network for Pneumonia Detection and Classification

“…According to this study, the authors developed DL for the classification task that can be trained with altered images by multi-stages of preprocessing. Abubeker and Baskar [13] targeted to make an innovative and effectual multi-class ML method for detecting and classifying CXR images on GPU. The authors primarily used a geometric extension by a positional transforming function to the unique database for increasing the sample size and aiding forthcoming TL.…”