Lung and Colon Cancer Classification using EfficientNet B3 Transfer Learning Model
Rahul Singh, Neha Sharma, Rupesh Gupta
Abstract
Lung and colon cancers are among the most common kinds of cancer in the globe. Lung cancer affects the respiratory system, while colon cancer impacts the digestive system. Both cancers have a high mortality rate and are often diagnosed at an advanced stage. Early detection and treatment are critical in improving survival rates for both lung and colon cancer. The study conducted an experiment to classify lung and colon cancer into five distinct classes using an EfficientNet B7 model. The model has been trained using Adam as the optimizer for a total of 12 epochs along with the batch size of 128 and the resulting accuracy and loss plots were carefully examined. Following this, accuracy is calculated based on precision, recall, and f1 score, and the model’s accuracy is 98%. The study’s findings show that the model accurately classified the various types of lungs as well as the colon cancer with high precision and recall. These findings indicate that the EfficientNet B7 model is a good fit for accurately classifying lung and colon cancer.