Histopathological Cancer Detection with Deep Neural Networks
Ninad Sail, Swati Nadkarni · 2023
Cancer is one of the most dreaded diseases faced by mankind today, largely because of the huge medical cost & high mortality rate. It is a serious and potentially life-threatening medical condition that can affect different parts of the body. The treatment and severity of any sort of cancer is determined by its accurate diagnosis for an affected person. There were approx. 10 million deaths recorded in the year 2020 due to cancer. Also, the study of histopathological cancer using modern technologies such as machine learning and deep neural networks became an area of interest. Research on histopathological cancer is quite popular among the researchers using different machine learning techniques such as Artificial Neural Network (ANN), Support Vector Machine (SVM), Self-supervised Learning (SSL), Feed-forward Neural Network (FNN), etc. These techniques are very prominent in detecting cancer at its early stages. Hence the survival rate of patients can be increased as the spreading of malignant cells can be reduced, upon detection. The research conducted on Convolutional Neural Network (CNN) shows the highest accuracy in classifying the cancer cells & non-cancer cells. The accuracy recorded was between 93%-96%. The paper aims to explore the use of Convolutional Neural Network (CNN) as a potential model in detection of histopathological cancer with its accuracy and its ability to detect at the early stage for an affected person.