A hybrid framework for enhanced segmentation and classification of colorectal cancer histopathology.

Introduction: Colorectal cancer (CRC) remains one of the leading causes of cancer-related deaths globally. Early detection and precise diagnosis are crucial in improving patient outcomes. Traditional histological evaluation through manual inspection of stained tissue slides is time-consuming, prone to observer variability, and susceptible to inconsistent diagnoses. Methods: To address these challenges, we propose a hybrid deep learning system combining Swin Transformer, EfficientNet, and ResUNet-A. This model integrates self-attention, compound scaling, and residual learning to enhance feature extraction, global context modeling, and spatial categorization. The model was trained and evaluated using a histopathological dataset that included serrated adenoma, polyps, adenocarcinoma, high-grade and low-grade intraepithelial neoplasia, and normal tissues. Results: Our hybrid model achieved impressive results, with 93% accuracy, 92% precision, 93% recall, and 93% F1-score. It outperformed individual architectures in both segmentation and classification tasks. Expert annotations and segmentation masks closely matched, demonstrating the model's reliability. Discussion: The proposed hybrid design proves to be a robust tool for the automated analysis of histopathological features in CRC, showing significant promise for improving diagnostic accuracy and efficiency in clinical settings.