Pneumonia remains a leading cause of mortality worldwide, particularly affecting vulnerable populations such as children and elderly patients. Early and accurate diagnosis through chest X-ray analysis is crucial for effective treatment and improved patient outcomes. Traditional diagnostic methods rely heavily on radiologists’ expertise, which can be time-consuming and subject to human error, especially in resource-limited settings. This study presents a novel deep learning approach for automated pneumonia detection from chest X-ray images using an enhanced EfficientNet-B3 architecture integrated with triple attention mechanisms and multi-scale feature fusion. The primary objective is to develop a robust, accurate, and interpretable model that demonstrates superior performance in binary classification tasks for pneumonia detection. A Model was developed that combines EfficientNet-B3 as the backbone network with three distinct attention mechanisms, including channel attention, spatial attention, and self-attention. The model incorporates multi-scale feature extraction from different network layers, progressive feature refinement, and adaptive pooling strategies. The system was trained and evaluated on a comprehensive chest X-ray dataset containing both normal and pneumonia cases to improve generalization. The proposed model achieved exceptional performance with a test accuracy of 95.2%, precision of 94.8%, recall of 95.6%, and F1-score of 95.2%. The AUC score reached 0.987, demonstrating excellent discriminative ability. The triple attention mechanism significantly improved feature representation, while multi-scale fusion enhanced the model’s ability to capture both local pathological patterns and global anatomical structures. The integration of multiple attention mechanisms with multi-scale feature fusion provides a powerful framework for pneumonia detection from chest X-rays.