Feature Extraction and Fusion Method for Detection and Classification of Gastrointestinal Diseases Using Deep Learning Algorithms

The diagnosis of digestive system problems necessitates specialized knowledge and skills, and manual image processing is time-consuming, error-prone, and knowledge-intensive. As a result, there is a desire for automated systems capable of analyzing endoscopic images swiftly and accurately to diagnose gastrointestinal disorders. Deep learning approach and convolutional neural networks (CNN) have demonstrated promising improvements in medical image processing challenges. By merging a pre-trained VGG19 model with CNN, this work proposes an improved technique for diagnosing diseases in the gastrointestinal tract using endoscopic images. The VGG19 model has been trained to extract useful characteristics from photographs, which may be used in a variety of image analysis applications, including medical image analysis. The proposed technique extracts high-level features from endoscopic images using the pre-trained VGG19 model, and these features are then used to categorize the data by feeding them into a CNN. The CNN is trained on a labeled dataset of endoscopic images to classify each image into one of the disease categories.