Knee osteoarthritis (OA) is a prevalent degenerative joint disease that significantly impacts mobility and quality of life. Early and accurate diagnosis is crucial for effective treatment and management. In this study, we propose a deep learning-based approach for automated classification of knee osteoarthritis severity using grayscale X-ray images. A convolutional neural network (CNN) model was developed to classify knee X-ray images into multiple severity categories, leveraging a dataset of knee images organized into distinct classes representing different stages of OA. The dataset was preprocessed by resizing images to 256 × 256 pixels, converting them to grayscale, and normalizing pixel values. The CNN architecture comprised three convolutional layers with ReLU activation and max-pooling, followed by fully connected layers with dropout regularization to prevent overfitting. The dataset was split into training (90%) and testing (10%) sets to evaluate model performance. The proposed model achieved a higher accuracy in testing, demonstrating its potential for reliable classification of knee OA severity. Visualization of the model’s predictions on test images further validated its effectiveness. This study highlights the potential of deep learning techniques in medical image analysis, offering a scalable and efficient tool for assisting clinicians in diagnosing knee osteoarthritis. Future work will focus on expanding the dataset, incorporating additional imaging modalities, and improving the model’s generalizability for real-world clinical applications.