Deep Learning–Driven Visualization Framework for Osteo Carcinoma Detection in X-Ray and MRI Images

Osteo carcinoma (osteosarcoma) is an aggressive primary bone malignancy that requires early and accurate diagnosis to improve patient survival and treatment outcomes. Radiological imaging modalities such as X-ray and Magnetic Resonance Imaging (MRI) play a crucial role in detecting structural and soft tissue abnormalities associated with bone tumors. However, manual interpretation of these images is time-consuming and subject to variability among radiologists. This study proposes a visualization-driven computational framework that integrates advanced image processing techniques with deep learning algorithms to enhance the detection of osteo carcinoma from X-ray and MRI images. The proposed system employs preprocessing methods including noise filtering, contrast enhancement, and normalization to improve image clarity and highlight pathological features. Texture and shape-based feature extraction techniques are combined with Convolutional Neural Network (CNN)–based classification to distinguish between normal and malignant bone tissues. Furthermore, Gradient-weighted Class Activation Mapping (Grad-CAM) is incorporated to generate interpretable heatmaps that localize tumor regions, thereby increasing clinical transparency and diagnostic confidence. Experimental results demonstrate that the integration of visualization techniques with deep learning significantly improves classification accuracy, sensitivity, and specificity compared to conventional methods. MRI images showed enhanced soft tissue delineation, while X-ray images provided effective preliminary structural assessment. The proposed framework offers a cost-effective, scalable, and clinically supportive diagnostic tool for early osteo carcinoma detection. This research highlights the potential of artificial intelligence–driven visualization systems in advancing medical imaging diagnostics and improving healthcare decision-making processes.