Xiangyu Tan1, Kexin Li1, Jiucheng Zhang2, Wenzhe Wang2, Bian Wu3, Jian Wu4, Xiaoping Li5, Xiaoyuan Huang6. 1. Department of Obstetrics and Gynecology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, 430030, Wuhan, Hubei, China. 2. College of Computer Science & Technology, Zhejiang University, 310027, Hangzhou, China. 3. Data Science and AI Lab, WeDoctor Group Limited, 311200, Hangzhou, China. 4. School of Public Health, Zhejiang University, 310027, Hangzhou, China. 5. Department of Obstetrics and Gynecology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, 430030, Wuhan, Hubei, China. xiaopingli11@163.com. 6. Department of Obstetrics and Gynecology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, 430030, Wuhan, Hubei, China. huangxy@tjh.tjmu.edu.cn.
Abstract
BACKGROUND: The incidence rates of cervical cancer in developing countries have been steeply increasing while the medical resources for prevention, detection, and treatment are still quite limited. Computer-based deep learning methods can achieve high-accuracy fast cancer screening. Such methods can lead to early diagnosis, effective treatment, and hopefully successful prevention of cervical cancer. In this work, we seek to construct a robust deep convolutional neural network (DCNN) model that can assist pathologists in screening cervical cancer. METHODS: ThinPrep cytologic test (TCT) images diagnosed by pathologists from many collaborating hospitals in different regions were collected. The images were divided into a training dataset (13,775 images), validation dataset (2301 images), and test dataset (408,030 images from 290 scanned copies) for training and effect evaluation of a faster region convolutional neural network (Faster R-CNN) system. RESULTS: The sensitivity and specificity of the proposed cervical cancer screening system was 99.4 and 34.8%, respectively, with an area under the curve (AUC) of 0.67. The model could also distinguish between negative and positive cells. The sensitivity values of the atypical squamous cells of undetermined significance (ASCUS), the low-grade squamous intraepithelial lesion (LSIL), and the high-grade squamous intraepithelial lesions (HSIL) were 89.3, 71.5, and 73.9%, respectively. This system could quickly classify the images and generate a test report in about 3 minutes. Hence, the system can reduce the burden on the pathologists and saves them valuable time to analyze more complex cases. CONCLUSIONS: In our study, a CNN-based TCT cervical-cancer screening model was established through a retrospective study of multicenter TCT images. This model shows improved speed and accuracy for cervical cancer screening, and helps overcome the shortage of medical resources required for cervical cancer screening.
BACKGROUND: The incidence rates of cervical cancer in developing countries have been steeply increasing while the medical resources for prevention, detection, and treatment are still quite limited. Computer-based deep learning methods can achieve high-accuracy fast cancer screening. Such methods can lead to early diagnosis, effective treatment, and hopefully successful prevention of cervical cancer. In this work, we seek to construct a robust deep convolutional neural network (DCNN) model that can assist pathologists in screening cervical cancer. METHODS: ThinPrep cytologic test (TCT) images diagnosed by pathologists from many collaborating hospitals in different regions were collected. The images were divided into a training dataset (13,775 images), validation dataset (2301 images), and test dataset (408,030 images from 290 scanned copies) for training and effect evaluation of a faster region convolutional neural network (Faster R-CNN) system. RESULTS: The sensitivity and specificity of the proposed cervical cancer screening system was 99.4 and 34.8%, respectively, with an area under the curve (AUC) of 0.67. The model could also distinguish between negative and positive cells. The sensitivity values of the atypical squamous cells of undetermined significance (ASCUS), the low-grade squamous intraepithelial lesion (LSIL), and the high-grade squamous intraepithelial lesions (HSIL) were 89.3, 71.5, and 73.9%, respectively. This system could quickly classify the images and generate a test report in about 3 minutes. Hence, the system can reduce the burden on the pathologists and saves them valuable time to analyze more complex cases. CONCLUSIONS: In our study, a CNN-based TCT cervical-cancer screening model was established through a retrospective study of multicenter TCT images. This model shows improved speed and accuracy for cervical cancer screening, and helps overcome the shortage of medical resources required for cervical cancer screening.
Entities:
Keywords:
Cervical cancer; Convolutional neural network (CNN); Deep leaning; ThinPrep cytologic test (TCT)
Authors: Diane Davis Davey; Rhona J Souers; Kelly Goodrich; Dina R Mody; Sana O Tabbara; Christine N Booth Journal: Arch Pathol Lab Med Date: 2019-04-25 Impact factor: 5.534
Authors: Clinton J Wang; Charlie A Hamm; Lynn J Savic; Marc Ferrante; Isabel Schobert; Todd Schlachter; MingDe Lin; Jeffrey C Weinreb; James S Duncan; Julius Chapiro; Brian Letzen Journal: Eur Radiol Date: 2019-05-15 Impact factor: 5.315
Authors: Susan J Curry; Alex H Krist; Douglas K Owens; Michael J Barry; Aaron B Caughey; Karina W Davidson; Chyke A Doubeni; John W Epling; Alex R Kemper; Martha Kubik; C Seth Landefeld; Carol M Mangione; Maureen G Phipps; Michael Silverstein; Melissa A Simon; Chien-Wen Tseng; John B Wong Journal: JAMA Date: 2018-08-21 Impact factor: 56.272