Classification and detection of Colon Cancer from histopathological images with CNN model: EfficientNet

Abstract

Colon cancer, a common type of carcinoma affecting the large intestine, impacts over 1.9 million individuals worldwide annually, posing a significant burden on global health. Early detection is crucial, and advancements in medical image classification techniques play an essential role in diagnosis, monitoring, and treatment, as with other types of cancer. However, manually analyzing large sets of medical images entails complexities and potential errors, leading to inaccurate diagnoses and delaying effective treatments. To overcome these challenges, convolutional neural networks (CNNs) were implemented using the EfficientNet method, proving to be viable in detecting colon cancer-related images, achieving a notable accuracy rate of 99.92%. This technological innovation holds promise in identifying the disease in its early stages, potentially revolutionizing treatment effectiveness and patient survival. The successful application of machine learning models in precise medical image classification offers hope for improving healthcare and paving the way for faster and accurate colon cancer diagnosis, with the potential to positively transform the landscape of public health.