Literature DB >> 33407588

Deep learning model for classifying endometrial lesions.

YunZheng Zhang1, ZiHao Wang1, Jin Zhang1, CuiCui Wang1, YuShan Wang1, Hao Chen1, LuHe Shan1, JiaNing Huo1, JiaHui Gu1, Xiaoxin Ma2.   

Abstract

BACKGROUND: Hysteroscopy is a commonly used technique for diagnosing endometrial lesions. It is essential to develop an objective model to aid clinicians in lesion diagnosis, as each type of lesion has a distinct treatment, and judgments of hysteroscopists are relatively subjective. This study constructs a convolutional neural network model that can automatically classify endometrial lesions using hysteroscopic images as input.
METHODS: All histopathologically confirmed endometrial lesion images were obtained from the Shengjing Hospital of China Medical University, including endometrial hyperplasia without atypia, atypical hyperplasia, endometrial cancer, endometrial polyps, and submucous myomas. The study included 1851 images from 454 patients. After the images were preprocessed (histogram equalization, addition of noise, rotations, and flips), a training set of 6478 images was input into a tuned VGGNet-16 model; 250 images were used as the test set to evaluate the model's performance. Thereafter, we compared the model's results with the diagnosis of gynecologists.
RESULTS: The overall accuracy of the VGGNet-16 model in classifying endometrial lesions is 80.8%. Its sensitivity to endometrial hyperplasia without atypia, atypical hyperplasia, endometrial cancer, endometrial polyp, and submucous myoma is 84.0%, 68.0%, 78.0%, 94.0%, and 80.0%, respectively; for these diagnoses, the model's specificity is 92.5%, 95.5%, 96.5%, 95.0%, and 96.5%, respectively. When classifying lesions as benign or as premalignant/malignant, the VGGNet-16 model's accuracy, sensitivity, and specificity are 90.8%, 83.0%, and 96.0%, respectively. The diagnostic performance of the VGGNet-16 model is slightly better than that of the three gynecologists in both classification tasks. With the aid of the model, the overall accuracy of the diagnosis of endometrial lesions by gynecologists can be improved.
CONCLUSIONS: The VGGNet-16 model performs well in classifying endometrial lesions from hysteroscopic images and can provide objective diagnostic evidence for hysteroscopists.

Entities:  

Keywords:  Computer-aided diagnosis; Convolutional neural network; Endometrial lesion; Hysteroscopy; VGGNet

Mesh:

Year:  2021        PMID: 33407588      PMCID: PMC7788977          DOI: 10.1186/s12967-020-02660-x

Source DB:  PubMed          Journal:  J Transl Med        ISSN: 1479-5876            Impact factor:   5.531


  22 in total

1.  Cost-effectiveness of diagnostic strategies for the management of abnormal uterine bleeding (heavy menstrual bleeding and post-menopausal bleeding): a decision analysis.

Authors:  Natalie A M Cooper; Pelham M Barton; Maria Breijer; Orla Caffrey; Brent C Opmeer; Anne Timmermans; Ben W J Mol; Khalid S Khan; T Justin Clark
Journal:  Health Technol Assess       Date:  2014-04       Impact factor: 4.014

2.  [Introduction of WHO classification of tumours of female reproductive organs, fourth edition].

Authors:  Zhaohui Lu; Jie Chen
Journal:  Zhonghua Bing Li Xue Za Zhi       Date:  2014-10

3.  Comparison of transvaginal ultrasonography and hysteroscopy in the diagnosis of uterine pathologies.

Authors:  Ali Babacan; Ismet Gun; Cem Kizilaslan; Okan Ozden; Murat Muhcu; Ercument Mungen; Vedat Atay
Journal:  Int J Clin Exp Med       Date:  2014-03-15

Review 4.  Fertility and infertility: Definition and epidemiology.

Authors:  Mélodie Vander Borght; Christine Wyns
Journal:  Clin Biochem       Date:  2018-03-16       Impact factor: 3.281

5.  Computer-Aided Diagnosis Based on Convolutional Neural Network System for Colorectal Polyp Classification: Preliminary Experience.

Authors:  Yoriaki Komeda; Hisashi Handa; Tomohiro Watanabe; Takanobu Nomura; Misaki Kitahashi; Toshiharu Sakurai; Ayana Okamoto; Tomohiro Minami; Masashi Kono; Tadaaki Arizumi; Mamoru Takenaka; Satoru Hagiwara; Shigenaga Matsui; Naoshi Nishida; Hiroshi Kashida; Masatoshi Kudo
Journal:  Oncology       Date:  2017-12-20       Impact factor: 2.935

6.  Dermatologist-level classification of skin cancer with deep neural networks.

Authors:  Andre Esteva; Brett Kuprel; Roberto A Novoa; Justin Ko; Susan M Swetter; Helen M Blau; Sebastian Thrun
Journal:  Nature       Date:  2017-01-25       Impact factor: 49.962

7.  Modeling Depression Symptoms from Social Network Data through Multiple Instance Learning.

Authors:  Akkapon Wongkoblap; Miguel A Vadillo; Vasa Curcin
Journal:  AMIA Jt Summits Transl Sci Proc       Date:  2019-05-06

8.  Automatic Recognition of Laryngoscopic Images Using a Deep-Learning Technique.

Authors:  Jianjun Ren; Xueping Jing; Jing Wang; Xue Ren; Yang Xu; Qiuyun Yang; Lanzhi Ma; Yi Sun; Wei Xu; Ning Yang; Jian Zou; Yongbo Zheng; Min Chen; Weigang Gan; Ting Xiang; Junnan An; Ruiqing Liu; Cao Lv; Ken Lin; Xianfeng Zheng; Fan Lou; Yufang Rao; Hui Yang; Kai Liu; Geoffrey Liu; Tao Lu; Xiujuan Zheng; Yu Zhao
Journal:  Laryngoscope       Date:  2020-02-18       Impact factor: 3.325

9.  Identification of cecum time-location in a colonoscopy video by deep learning analysis of colonoscope movement.

Authors:  Hyoun-Joong Kong; Sungwan Kim; Minwoo Cho; Jee Hyun Kim; Kyoung Sup Hong; Joo Sung Kim
Journal:  PeerJ       Date:  2019-07-29       Impact factor: 2.984

10.  Support System of Cystoscopic Diagnosis for Bladder Cancer Based on Artificial Intelligence.

Authors:  Atsushi Ikeda; Hirokazu Nosato; Yuta Kochi; Takahiro Kojima; Koji Kawai; Hidenori Sakanashi; Masahiro Murakawa; Hiroyuki Nishiyama
Journal:  J Endourol       Date:  2020-01-14       Impact factor: 2.942

View more
  1 in total

Review 1.  Machine Learning for Endometrial Cancer Prediction and Prognostication.

Authors:  Vipul Bhardwaj; Arundhiti Sharma; Snijesh Valiya Parambath; Ijaz Gul; Xi Zhang; Peter E Lobie; Peiwu Qin; Vijay Pandey
Journal:  Front Oncol       Date:  2022-07-27       Impact factor: 5.738

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.