The rapid spread of COVID-19 has necessitated the development of accurate, automated, and computationally efficient diagnostic systems based on medical imaging. This study proposes an intelligent COVID-19 classification framework leveraging multi-resolution Curvelet Transform (CT) feature extraction combined with an optimized Support Vector Machine (SVM) classifier. The Curvelet Transform is employed due to its superior capability in capturing anisotropic edges and fine-grained texture information in chest X-ray images, which are critical for identifying COVID-19-related pulmonary abnormalities. The extracted features are further optimized to reduce redundancy and enhance classification performance. The SVM model, known for its robustness in high-dimensional feature spaces, is fine-tuned to achieve optimal separation between COVID-19, pneumonia, and normal cases.

The proposed framework is evaluated conceptually against existing machine learning and deep learning approaches, demonstrating improved interpretability and computational efficiency compared to deep neural architectures. Prior studies such as CoroDet and hybrid CNN-based frameworks have shown strong performance; however, they often suffer from high computational complexity and limited feature interpretability (Hussain et al., 2021; Islam et al., 2020). Similarly, traditional machine learning approaches combined with handcrafted features provide better explainability but struggle with feature representation limitations (Kassani et al., 2021; Saygili, 2021).

The integration of Curvelet-based multi-resolution analysis with optimized SVM presents a balanced trade-off between accuracy and computational cost. The study highlights the potential of transform-domain feature engineering in enhancing diagnostic reliability for COVID-19 detection systems. Experimental comparisons from related literature indicate that texture-sensitive transforms significantly improve classification performance in medical imaging datasets. The proposed system contributes to the growing body of intelligent diagnostic tools aimed at supporting radiological decision-making in pandemic scenarios.