: Since the COVID-19 outbreak, Reverse Transcription Polymerase Chain Re-action (RT-PCR) has limitations of long cycles and low sensitivity, making medical imaging (CT/X-ray) critical for early auxiliary diagnosis. However, manual image interpretation by radiologists is inefficient and prone to errors due to subjective factors.To address this, this paper compares two automated solutions using the CoronaHack dataset with preprocessing: LBP-based SVM and transfer learning-enabled MobileNetV2. Experimental results indicated that the optimized LBP-SVM model attained a test accuracy of 80.65%, whereas the MobileNetV2 model achieved a test accuracy of 95.36% and a recall rate of 98.40 — with only 4 out of 250 true positive samples being missed. Because the application of data augmentation and regularization techniques, the training and validation loss curves of MobileNetV2 con-verged effectively, which in turn suppressed the issue of overfitting. This confirms the superiority of MobileNetV2 in end-to-end feature extraction for COVID-19 imaging.
Pneumonia is a significant cause of mortality, particularly in children under five. Accurate detection of pneumonia from Chest X-ray (CXR) images is crucial in mitigating diagnostic errors common in manual radiographic analysis. This study leverages deep learning models to enhance the detection of multiclass pneumonia (normal, bacterial, and viral) using CXR images. We utilized a dataset comprising 5,863 multiclass pneumonia CXR samples. Data augmentation and regularization techniques were applied to address class imbalance and overfitting. Pre-trained models, including EfficientNet, MobileNet, RegNet, and ViT, were fine-tuned using the PyTorch framework, with transfer learning employed to optimize training. Model performance was assessed using accuracy, precision, recall, and specificity. The fine-tuned models achieved high classification accuracy, with EfficientNet and ConvNext models achieving accuracy scores of 83% and 82%, respectively. Data augmentation and regularization significantly improved the models' generalization, reducing overfitting and improving predictive accuracy. The proposed deep learning models provide an efficient and accurate tool for multiclass pneumonia detection from CXRs. These models have the potential to support healthcare professionals in making more accurate diagnoses.
One of the most common and deadly infectious illnesses in the world is still tuberculosis (TB), especially in developing nations with inadequate healthcare systems. In order to stop the spread of tuberculosis and enhance patient outcomes, early identification and diagnosis are essential. In this study, we present a deep learning-based system that uses chest X-ray pictures to automatically detect tuberculosis. Despite the difficulties of limited dataset availability, the system uses transfer learning using MobileNetV2 and DenseNet architectures to classify chest Xrays as either TB-positive or Healthy, reaching notable accuracy. To increase model generalisation and image quality, pre-processing methods like Contrast Limited Adaptive Histogram Equalisation (CLAHE) and sophisticated data augmentation approaches are used. The trained model is then implemented as a Flask web application, offering a user-friendly interface with features like secure login, image upload and preview, prediction results with probability scores, and performance metrics visualisation like accuracy curves, confusion matrices, and ROC curves. The suggested framework shows how deep learning can be used to create scalable, dependable, and affordable diagnostic tools to help radiologists and other medical professionals with TB screening and diagnosis.
Background: COVID-19, pneumonia, and TB (tuberculosis) are still the big killers of people suffering from chest disease and continue to be a serious health challenge globally. A timely diagnosis leads to timely treatment and improved patient outcomes. Chest X-ray (CXR) imaging is widely used for diagnostic purposes due to its speed, low cost, and availability in most healthcare facilities. Manual reading of CXR images is, however, challenging because of the similarity in the presentation of radiographic features across various chest diseases. Methods: This research introduces a hybrid ensemble learning approach to classify chest X-ray images into four classes—Normal, COVID-19, Pneumonia, and Tuberculosis. Three Deep CNN network models, namely Xception, AlexNet, and EfficientNet-B0, were used for deep feature extraction. Additionally, texture features of the images were extracted using Gabor filters. The deep and texture features were combined and classified using logistic regression and a stacking ensemble learning approach. A publicly available chest X-ray image database containing 7,135 X-rays was used, with six-fold stratified cross-validation to assess the proposed approach. Results: The ensemble models outperformed the individual CNN models. The Average Ensemble produced the best results with an accuracy of 91.18%, an Average Precision (AP) of 96.75%, and an Area Under the ROC Curve (AUC) of 98.87%. The proposed model performs well across all four disease classes. It exhibited high sensitivity in detecting Tuberculosis with considerable stability in classifying Normal, COVID-19, and Pneumonia. Conclusion: The proposed framework demonstrates the effectiveness of integrating deep learning features, Gabor texture features, and ensemble learning for improved chest X-ray image classification. This can help computer-assisted diagnostics systems and aid medical workers in identifying chest diseases early.
Pneumonia is a respiratory condition, which is critical and needs an accurate diagnosis to minimize death. The following paper introduces a deep learning model, which can be explained, to identify pneumonia in the X-ray images of the chest, using the Convolutional Neural Networks (CNN) architecture. The image resizing, normalization and data augmentation are some of the preprocessing strategies included in the proposed approach to improve the quality of data and the generalization of the model. CNN model is created to automatically identify discriminative features and binomial classification between normal and pneumonia cases with the help of a binary cross-entropy loss and a sigmoid activation function. Gradient-weighted Class Activation Mapping (Grad-CAM) is added to enhance the interpretability of the model by producing visual explanations by highlighting the areas that do affect the model predictions. The results of the experiment prove that the proposed model can show high classification rates (up to 97 percent) and can compete with those of the pretrained models like VGG16, ResNet50, and DenseNet121, at the same time, having lower computational complexity. The suggested framework offers high precision, as well as interpretability, and is an effective instrument to use in the process of computer-aided diagnosis and assist in clinical decision-making in the field of medical imaging.
In response to the challenge of balancing accuracy and generalization in medical image classification using convolutional neural networks (CNNs), this paper proposes ResKAN18, a hybrid structure that embeds the learnable spline function of the Kolmogorov–Arnold network (KAN) into the ResNet18 classification head for intelligent diagnosis of COVID-19 in chest X-ray images. ResKAN18 includes three variants: ResKAN18—Large (four layers of KAN, hidden-layer dimensions [256, 128, 64]), ResKAN18—Standard (four layers of KAN, hidden-layer dimensions [128, 64, 32]), and ResKAN18—Simple (three layers of KAN, hidden-layer dimensions [64, 32]), which can achieve a flexible balance between accuracy and efficiency with different depths of KAN. A systematic comparison has been conducted between four classic CNN baselines including ResNet18, VGG16, DenseNet121, ShuffleNetV2, and three ResKAN18 variants on a benchmark dataset containing 3880 chest X-rays (COVID-19, normal, viral pneumonia). The results have shown that ResKAN18—Large can achieve an accuracy of 98.80% on the independent test set, which is 1.21% higher than ResNet18 and 0.69% higher than DenseNet121—its parameter count is 13.97M, inference delay is 8.25 ms/image, and training–validation accuracy difference is only 1.50%. The accuracy and performance stability of the dataset under random partitioning conditions are superior to the other two variants and all classic CNN baselines. All ResKAN18 variants have achieved zero missed diagnoses for COVID-19, while ResNet18 has shown missed diagnoses (0.9944). Taking into account the trade-off between accuracy, generalization, and inference efficiency, ResKAN1—Large is recommended as the default configuration, while for edge deployment scenarios with severely limited resources, ResKAN18—Simple can provide a cost-effective alternative with an extremely low latency of 2.50 ms/image and only 3.1% parameter increment compared to ResNet18.