A Machine Learning Approach for Detection and Classification of Colon Cancer using Convolutional Neural Network Architecture

Main Article Content

Subrata Sinha, Saurav Mali, Utpala Borgohain, Ujjal Saikia, Gunadeep Chetia, Smriti Priya Medhi, Debashree Borthakur, Jayanta Aich, Gunjan Mukherjee, Dolan Ghosh, Rajeev Sarmah


In both developing and developed countries, cancer has become one of the most fearsome health ailments in the 21st century. Among the different variants of cancer, colon cancer is the 3rd most prevalent cancer globally and the 2nd deadliest form of cancer. Early diagnosis is very critical for the successful treatment of colon cancer.  In recent times, machine-learning approaches have made significant strides in IoT-assisted healthcare systems. Computer-aided diagnostic systems driven by deep learning algorithms can detect colon cancer with great accuracy, assisting the medical profession in quick diagnosis and developing quick remedies against it. In our approach, we have developed a deep learning model based on convolutional neural network architecture to accurately classify and detect colon cancer.  The developed model has been trained with 10,000 histopathological images of the colon, divided into two classes: colon adenocarcinoma and colon benign tissues, each containing 5000 images. We have also employed various performance metrics in our research to monitor the performance of our machine-learning model.  The proposed model has been trained for 30 epochs with a batch size of 32 and achieved an overall accuracy of 98.79%.

Article Details