Performance Analysis of Rabbit Algorithm with Gaussian Process Classifier for Detection of Adenocarcinoma from Lung Cancer Histopathological Images

Lung and colon cancers are among the most common and fatal tumors. In 2022, 4.27 million people were diagnosed with lung and colon cancer, and more than 2.74 million died worldwide. Parts of the lungs affected by cancer are located utilizing image processing techniques such as noise reduction, feature extraction, damaged region identification, and possibly a comparison to a medical history of lung cancer. The primary goal of this study is to employ Rabbit Algorithm with Gaussian process classifier to detect and classify cancer using histopathology images of lung. The LC25000 collection contains $\mathbf{2 5, 0 0 0}$ histopathological images from the human lung and colon, including both malignant and normal scans. Classifier performance parameters like Accuracy, Error rates, F1 score, Sensitivity, Specificity, Critical Success Index (CSI) and Youden’s Index (YI) are used to assess how well a classifier can distinguish between classes. Results indicate that an accuracy rate of $\mathbf{9 1. 6 7 \%}$ in detecting Adenocarcinoma can be achieved using the Rabbit Algorithm which is also characterized by the Gaussian Process Classifier.