Education, Science, Technology, Innovation and Life
Open Access
Sign In

Clinical Value Analysis of “Internet +” Artificial Intelligence Assisted Diagnosis of Cervical Intraepithelial Lesions

Download as PDF

DOI: 10.23977/medsc.2022.030111 | Downloads: 16 | Views: 1004

Author(s)

Qingling Qi 1, Qianyin Wang 1, Wenjun Gao 1, Xue Gong 1, Hui Zeng 1, Sha Zhang 1, Liehong Wang 1

Affiliation(s)

1 Qinghai Reo Cross Hospital, Xining Qinghai, 810000, China

Corresponding Author

Liehong Wang

ABSTRACT

Artificial intelligence (AI) can automatically detect abnormalities in digital cytology images, but its effectiveness in cervical cytology remains to be studied. Our objective was to evaluate the performance and clinical value of “Internet +” AI-assisted cytology in the detection of cervical intraepithelial lesions. A total of 7,225 women were tested for cervical cancer at Qinghai Red Cross Hospital. Cervical epithelial cells were collected with a cervical brush. After fixation, the cervical cell samples were prepared by ThinPrep method in the laboratory and then stained with Feulgen+EA50. After staining, cell images and other data were sent to Ali Cloud by cell scanner. The AI-assisted cytology system in the cloud was used for analysis, and each tested cell was scored. The cell technician first classified the cell samples as negative or positive based on the score of each sample, and then made the diagnosis by reviewing 100% positive samples and 10% negative samples. Each woman with abnormal cytology (including low-grade squamous intraepithelial lesion [LSIL], atypical squamous cells where it was not possible to exclude high grade squamous intraepithelial lesion [ASC-H] and high squamous intraepithelial lesion [HSIL]) identified by either AI or cytologists was refered to colposcopy and biopsy for histological confirmation, and women with atypical squamous cells of undetermined significance (ASC-US) were recommended for an reexamination during 6-12 months. Of these, 698 women underwent colposcopy and histopathology. Results Colposcopy-directed biopsies were performed in 698 women with abnormal cytology diagnoed by either AI or cytologists. The biopsy identified 67 invasive cancer, 64 cervical intraepithelial neoplasia grade 3 (CIN3), 43 cervical intraepithelial neoplasia grade 2 (CIN2), 98 cervical intraepithelial neoplasia grade 1 (CIN1), and 426 cervicitis. By comparing the coincidence rate of cervical cytology and histopathological diagnosis, it could conclude that the detection of CIN2+ among women with HSIL, ASC-H, LSIL, and ASC-US was 92.31%, 77.55%, 32.26% and 7.63%, respectively. If LSIL was used as the positive criterion, the sensitivity of AI in diagnosing CIN1+, CIN2+ and CIN3+ was 69.49%, 82.76% and 83.97%, respectively. In addition, it was found that with the increase of cytological diagnostic level by AI technology, the pathological level of patients also gradually increased. The “Internet+” AI diagnostic technology can assist cytologists in diagnosis, and it has certain clinical value for the early diagnosis of cervical cancer. Further research is needed to apply this technique to more population samples.

KEYWORDS

“internet +”, Artificial intelligence, Cervical cancer detection, Histopathologic

CITE THIS PAPER

Qingling Qi, Qianyin Wang, Wenjun Gao, Xue Gong, Hui Zeng, Sha Zhang, Liehong Wang, Clinical Value Analysis of “Internet +” Artificial Intelligence Assisted Diagnosis of Cervical Intraepithelial Lesions. MEDS Clinical Medicine (2022) Vol. 3: 59-64. DOI: http://dx.doi.org/10.23977/medsc.2022.030111.

REFERENCES

[1] Big Data [J]. Nature, 2008, 455(7209): 1-136.
[2] Bishop C M, Nasrabadi N M. Pattern Recognition and Machine Learning [J]. J Elect Imaging, 2007, 16(4): 20-25.
[3] Buyya R, Yeo C S, Venugopal S, et al. Cloud computing and emerging IT platforms: vision, hype, and reality for delivering computing as the 5th ulity [J]. Future Generation Computer Systems, 2009, 25(6): 599-616.
[4] Cukier K. Data, data everywhere. Economist, 2010, 394: 3-16.
[5] Dealing with Data [J]. Science, 2011, 331(6018): 639-806.
[6] de Grey A D. Artificial intelligence and medical research: time to aim higher? [J]. Rejuvenation Res, 2016, 19(2): 105-106.
[7] Dilsizian S E, Siegel E L. Artificial intelligence in medicine and cardiac imaging: harnessing big data and advanced computing to provide personalized medical diagnosis and treatment [J]. Curr Cardiol Rep, 2013, 16(1): 1-8.
[8] Franc S, Daoudi A, Mounier S, et al. Telemedicine: what more is needed for its integration in everyday life? [J]. Diabetes & Metabolism, 2011, 37(6): 71-77.
[9] Gong Yang, Zhang Jiajie. Toward a Human-centered Hyper-lipidemia Management System: the interaction between internal and external information on relational data search [J]. Journal of Medical Systems, 2011, 35(2): 169-177.
[10] Hastie T, Tibshirani R, Friedman J H, et al. The elements of statistical learning: data mining, inference, and prediction [J]. Math Intell, 2001, 27(2): 83-85.
[11] Lawrence D R, Palacios-Gonzalez C, Harris J. Artificial Intelligence [J]. Camb Q Healthc Ethics, 2016, 25(2): 250-261.
[12] Lerouge C, Garfield M J, Collins R W. Telemedicine: Technology mediated service relationship, encounter, or something else? [J]. International Journal of Medical Infornatics, 2012, 81(9): 622.]
[13] Rosenbloom S T, Denny J C, Xu H, et al. Data from Clinical Notes: a Perspective on the ension between structure and flexible documentation [J]. Journal of the American Medical Informatics Association, 2011, 18(2): 181-186.
[14] Saitwal Hinali, Feng Xuan, Walji Muhammad, et al. Assessing Performance of an Electronic Health Record (EHR) Using Cognitive Task Analysis [J]. International Journal of Medical Informatics, 2010, 79(7): 501-506.
[15] Wyatt J C, Liu J L. Basic Concepts in Medical Informatics [J]. J Epidemiol Community Health, 2002, 56(11): 808-812.
[16] Yan Dong, Jigeng Bai, Yuping Zhang, et al. Automated Quantitative Cytology Imaging Analysis System in Cervical Cancer Screening in Shanxi Province, China [J]. Cancer and Clinical Oncology, 2017, 6 (2): 51-59.
[17] Hua Chen, Juan Liu, Qing-Man Wen, et al. CytoBrain: Cervical Cancer Screening System Based on Deep Learning Technology [J]. J. Comput. Sci. & Technol., 2021, 36(2): 347-360.
[18] Heling Bao, Xiaorong Sun, Yi Zhang, et al. The artificial intelligence-assisted cytology diagnostic system in large-scale cervical cancer screening: A population-based cohort study of 0.7 million women [J]. Cancer Medicine, 2020, 9: 6896-6906.
[19] Heling Bao, Hui Bi, Xiaosong Zhang, et al. Artificial intelligence-assisted cytology for detection of cervical intraepithelial neoplasia or invasive cancer: A multicenter, clinical-based, observational study [J]. Gynecologic Oncology, 2020, 159: 171-178.

Downloads: 3870
Visits: 166989

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.