RT Journal Article
SR Electronic
T1 Enhancing Histopathological Colorectal Cancer Image Classification by using Convolutional Neural Network
JF medRxiv
FD Cold Spring Harbor Laboratory Press
SP 2021.03.17.21253390
DO 10.1101/2021.03.17.21253390
A1 Radwan Al.Shawesh
A1 Yi Xiang Chen
YR 2021
UL http://medrxiv.org/content/early/2021/03/24/2021.03.17.21253390.abstract
AB Colorectal cancer (CRC) also known as bowl cancer is one of the leading death 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 advancement’s in deep learning, since classification methods are becoming more and more accurate and efficient. In this study, we introduce a method to improve the classification accuracy from previous studies that used the National Center for Tumor diseases (NCT) data sets. We adapt the ResNet-50 model in our experiment to classify the CRC histopathological images. Furthermore, we utilize transfer learning and fine-tunning techniques to improve the accuracy. Our Experiment results show that ResNet_50 network is the best CNN architecture so far for classifying CRC histopathological images on the NCT Biobank open source dataset. In addition to that using transfer learning allow us to obtain 97.7% accuracy on the validation dataset, which is better than all previous results we found in literature.Competing Interest StatementThe authors have declared no competing interest.Funding StatementThis work was supported in part by the National Key Research 177 and Development Project of China (2018YFB2101300)Author DeclarationsI confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.YesThe details of the IRB/oversight body that provided approval or exemption for the research described are given below:Yixiang Chen, yxchen@sei.ecnu.edu.cnAll necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived.YesI understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).YesI have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.YesYes - all data are fully available without restriction http://dx.doi.org/10.5281/zenodo.1214456 https://drive.google.com/drive/folders/1Xcu1LPBV5FKoe2-HW6OmgCpw7Uz2WLX5?usp=sharing