TB-CXRNet is a deep learning framework for detecting tuberculosis (TB) and drug-resistant TB using chest X-ray (CXR) images. The framework uses a dataset of 40,000 CXR images to classify TB, non-TB (other lung infections), and healthy patients with an accuracy of 93.32%. It also implements a stacking machine learning model to detect drug-resistant TB with 87.48% accuracy for binary classification (drug-resistant vs drug-sensitive TB) and 79.59% for three-class classification (multi-drug resistant (MDR), extreme drug-resistant (XDR), and sensitive TB). The proposed approach outperforms existing methods in accuracy and is the best reported result compared to the literature. The framework also includes a Score-CAM-based visualization technique to make the model interpretable. The model can detect TB and drug-resistant TB from chest X-rays quickly and reliably, helping to reduce disease complications and spread. The largest drug-resistant TB dataset will be released to develop a machine learning model for drug-resistant TB detection and stratification. The study also discusses the challenges of detecting drug-resistant TB and the importance of early detection. The framework uses a pre-trained CNN model (CheXNet) for feature extraction and a self-MLP classifier for classification. The model is trained on a large and robust dataset, which confirms its reliability and generalizability. The study also highlights the importance of using Score-CAM-based visualization to understand how the model makes decisions. The model is able to distinguish between TB and non-TB cases with high accuracy and can detect MDR-TB reliably. The study also discusses the challenges of detecting drug-resistant TB and the importance of early detection. The framework is able to detect TB and drug-resistant TB from chest X-rays with high accuracy and can help reduce disease complications and spread. The study also discusses the importance of using Score-CAM-based visualization to understand how the model makes decisions. The model is able to distinguish between TB and non-TB cases with high accuracy and can detect MDR-TB reliably. The study also discusses the challenges of detecting drug-resistant TB and the importance of early detection.TB-CXRNet is a deep learning framework for detecting tuberculosis (TB) and drug-resistant TB using chest X-ray (CXR) images. The framework uses a dataset of 40,000 CXR images to classify TB, non-TB (other lung infections), and healthy patients with an accuracy of 93.32%. It also implements a stacking machine learning model to detect drug-resistant TB with 87.48% accuracy for binary classification (drug-resistant vs drug-sensitive TB) and 79.59% for three-class classification (multi-drug resistant (MDR), extreme drug-resistant (XDR), and sensitive TB). The proposed approach outperforms existing methods in accuracy and is the best reported result compared to the literature. The framework also includes a Score-CAM-based visualization technique to make the model interpretable. The model can detect TB and drug-resistant TB from chest X-rays quickly and reliably, helping to reduce disease complications and spread. The largest drug-resistant TB dataset will be released to develop a machine learning model for drug-resistant TB detection and stratification. The study also discusses the challenges of detecting drug-resistant TB and the importance of early detection. The framework uses a pre-trained CNN model (CheXNet) for feature extraction and a self-MLP classifier for classification. The model is trained on a large and robust dataset, which confirms its reliability and generalizability. The study also highlights the importance of using Score-CAM-based visualization to understand how the model makes decisions. The model is able to distinguish between TB and non-TB cases with high accuracy and can detect MDR-TB reliably. The study also discusses the challenges of detecting drug-resistant TB and the importance of early detection. The framework is able to detect TB and drug-resistant TB from chest X-rays with high accuracy and can help reduce disease complications and spread. The study also discusses the importance of using Score-CAM-based visualization to understand how the model makes decisions. The model is able to distinguish between TB and non-TB cases with high accuracy and can detect MDR-TB reliably. The study also discusses the challenges of detecting drug-resistant TB and the importance of early detection.