Computer-Aided Polyps Classification from Colonoscopy Using
Stacking-Based Deep Learning Model
Colorectal cancer is responsible for a high proportion of cancer
mortality. The most effective way to avoid colorectal cancer is to have
a colonoscopy. However, not every polyp in the colon is prone to cancer.
As a result, different techniques are employed to classify polyps. Video
endoscopy can diagnose stomach ulcers, bleeding, and polyps. Doctors
spend a lot of time reviewing medical video endoscopy images.
Computer-aided diagnosis to analyze all images quickly and accurately.
Diagnosing gastrointestinal problems with the proposed strategy is
novel. Following the classification of the gastrointestinal disorder,
the further therapy and surgeries will be determined by its
classification. The 5000 images in the Kvasir dataset are evenly
distributed across five different digestive tract-related categories:
ulcerative colitis, dye-lifted polyps, resection margins, normal cecum,
polyps, and ulcerative polyps. The images used in the deep learning
networks went through enhancements and noise reduction. We suggested a
stacking ensemble approach to boost the model’s accuracy for prediction.
The ensemble approach included five meticulously tuned deep
convolutional neural network architectures, namely Xception, ResNet-101,
VGG-19, EfficientNetB2v3, and MobineNetV2. These models were trained
using weights obtained from the ImageNet dataset. Accuracy of 96.50%
was achieved using meta models based on logistic regression, K-NN,
Decision Tree, Support vector machine, and Naive Bayes classifier.