Histopathological Colorectal Cancer Image Classification Based On Convolutional Neural Network

Colorectal cancer(CRC)also known as bowl cancer is one of the leading deaths causes worldwide.Early diagnosis has become vital for a successful treatment.Now days with the new advancements in Convolutional Neural networks(CNNs)it’s possible to classify different images of CRC into different classes.Today It is crucial for physician to take advantage of the new advancements in deep learning,since classification methods are becoming more and more accurate and efficient.This article uses the National Cancer Center(NCT)data set.This paper combines transfer learning and fine-tuning technology with ResNet-50 model to train and classify CRC histopathological images.The experimental results show that in all CNN network architectures,based on the NCT data set,the ResNet＿50 model can classify CRC histopathological images with high accuracy reaching 97.7%.Based on the model designed and technique used in this paper,a colorectal cancer histology image classification software(AI pathologist)was developed to automatically classify CRC images.AI pathologist software can classify CRC images and show the probability of each classification.For uncertain categories,the tool will mark the image,and the pathologist can manually extract the CRC image for manual classification.This tool can efficiently and accurately classify CRC images,improve the efficiency of pathological diagnosis of CRC,and has great application value for clinical diagnosis of CRC.