Automatic classification of cervical cancer from cytological images by using convolutional neural network.

Miao Wu,Huiqiang Liu,Yi Yin,Qian Liu,Chuanbo Yan

doi:10.1042/bsr20181769

Abstract

Cervical cancer (CC) is one of the most common gynecologic malignancies in the world. The incidence and mortality keep high in some remote and poor medical condition regions in China. In order to improve the current situation and promote the pathologists’ diagnostic accuracy of CC in such regions, we tried to propose an intelligent and efficient classification model for CC based on convolutional neural network (CNN) with relatively simple architecture compared with others. The model was trained and tested by two groups of image datasets, respectively, which were original image group with a volume of 3012 datasets and augmented image group with a volume of 108432 datasets. Each group has a number of fixed-size RGB images (227*227) of keratinizing squamous, non-keratinizing squamous, and basaloid squamous. The method of three-folder cross-validation was applied to the model. And the classification accuracy of the models, overall, 93.33% for original image group and 89.48% for augmented image group. The improvement of 3.85% has been achieved by using augmented images as input data for the model. The results got from paired-samples ttest indicated that two models’ classification accuracy has a significant difference (P<0.05). The developed scheme we proposed was useful for classifying CCs from cytological images and the model can be served as a pathologist assistance to improve the doctor’s diagnostic level of CC, which has a great meaning and huge potential application in poor medical condition areas in China.

Highlights

Cervical cancer (CC) remains one of the leading causes of cancer-related deaths in women worldwide [1], with 80% of the cases occurring in developing countries [2]
Teramoto et al [16] developed an automated classification scheme for lung cancers presented in microscopic images using deep convolutional neural network (DCNN) and the results showed that approximately 71% of images were correctly classified, which is at par with the accuracy of cytotechnologists and pathologists
By reading the extensive literature, we found researchers around the world had carried out various studies on computer-aided diagnosis (CADx) on CC recent years

Summary

Introduction

Cervical cancer (CC) remains one of the leading causes of cancer-related deaths in women worldwide [1], with 80% of the cases occurring in developing countries [2]. The incidence of this cancer in young Chinese women (≤30 years old) is increasing by 2–3% yearly [5]. In some remote districts like Xinjiang Uyghur Autonomous Region in Northwest China, which has poor medical conditions like insufficient healthcare accessibility and qualified medical staff, CC incidence and mortality are even higher. It has already been an extremely important public health issue in Xinjiang area. The data from Chinese Health Statistics Yearbook published in recent years indicated that the overall level of Xinjiang public medical and health conditions, such as healthcare facilities, health funds, health technicians, medical service etc. Due to the historical reasons and unbalanced development in China, the situation of medical and health condition in Southern Xinjiang is even worse, which is urgently needed to improve

Methods

Results

Discussion

Conclusion