Abstract In the era of modern medicine, with the continuous advancement of targeted therapies, rapid and accurate diagnosis of colorectal cancer plays a pivotal role in designing personalized treatment regimens. However, traditional diagnostic methods, which rely on visual examination of tissue samples by pathologists, are not only time-consuming but also face challenges such as personnel fatigue and human error. This study introduces an intelligent deep learning-based framework for the automated detection of colorectal cancer, leveraging a combination of Convolutional Neural Networks (CNNs) and transfer learning. The proposed model, trained on a dataset of digital images of colorectal tissues, not only achieves high accuracy in distinguishing cancerous cells from healthy tissues but also precisely classifies various tissue subtypes, including: TUMOR, STROMA, NORM, LYMPHO, DEBRIS, MUCOSA, ADIPOSE, EMPTY and MUS. Evaluation results demonstrate that the proposed model outperforms conventional methods in terms of higher sensitivity, improved accuracy, and better generalization capability. This system can serve as an intelligent assistant to pathologists, enhancing diagnostic speed and precision, while playing a crucial role in early screening and effective management of colorectal cancer. The deployment of such systems could revolutionize digital pathology, reduce diagnostic time, optimize treatment costs, and ultimately, significantly improve patients' survival rates and quality of life.
ghahremani et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: