Biomimetic Transfer Learning-Based Complex Gastrointestinal Polyp Classification.

(1) Background: This research investigates the application of Artificial Intelligence (AI), particularly biomimetic convolutional neural networks (CNNs), for the automatic classification of gastrointestinal (GI) polyps in endoscopic images. The study combines AI and Transfer learning techniques to support early detection of colorectal cancer by enhancing diagnostic accuracy with pre-trained models; (2) Methods: The Kvasir dataset, comprising 4000 annotated endoscopic images across eight polyp categories, was used. Images were pre-processed via normalisation, resizing, and data augmentation. Several CNN architectures, including state-of-the-art optimized ResNet50, DenseNet121, and MobileNetV2, were trained and evaluated. Models were assessed through training, validation, and testing phases, using performance metrics such as overall accuracy, confusion matrix, precision, recall, and F1 score; (3) Results: ResNet50 achieved the highest validation accuracy at 90.5%, followed closely by DenseNet121 with 87.5% and MobileNetV2 with 86.5%. The models demonstrated good generalisation, with small differences between training and validation accuracy. The average inference time was under 0.5 s on a computer with limited resources, confirming real-time applicability. Confusion matrix analysis indicates that common errors frequently occur between visually similar classes, particularly when reviewed by less-experienced medical physicians. These errors underscore the difficulty of distinguishing subtle features in gastrointestinal imagery and highlight the value of model-assisted diagnostics; (4) Conclusions: The obtained results confirm that Deep learning-based CNN architectures, combined with Transfer learning and optimisation techniques, can classify accurately endoscopic images and support medical diagnostics.