A Method for Identifying Cervical Abnormal Cells Based on Sample Benchmark Values

The identification of cervical abnormal cells using deep learning methods usually requires a large amount of training data, but these data inevitably use different samples of cervical abnormal cells to participate in model training, and naturally miss the positive and abnormal intracellular controls...

Full description

Saved in:
Bibliographic Details
Main Authors: ZHAO Si-qi, LIANG Yi-qin, QIN Jian, HE Yong-jun
Format: Article
Language:zho
Published: Harbin University of Science and Technology Publications 2022-12-01
Series:Journal of Harbin University of Science and Technology
Subjects:
Online Access:https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2164
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849387569404444672
author ZHAO Si-qi
LIANG Yi-qin
QIN Jian
HE Yong-jun
author_facet ZHAO Si-qi
LIANG Yi-qin
QIN Jian
HE Yong-jun
author_sort ZHAO Si-qi
collection DOAJ
description The identification of cervical abnormal cells using deep learning methods usually requires a large amount of training data, but these data inevitably use different samples of cervical abnormal cells to participate in model training, and naturally miss the positive and abnormal intracellular controls of a single sample, resulting in the fact that recognition accuracy of cervical abnormal cells is not high, and the false positive rate is high. To solve this problem, this paper proposes a method for identifying cervical abnormal cells based on sample benchmark values. Firstly, this method identifies cervical cells and segments cervical cell nucleus by Mask R-CNN model. Then we calculate the key cervical nucleus indicators, propose the concept of benchmark cells, define the sample benchmark values, and quantify the diagnostic criteria.Finally, the abnormal nucleus indicator and model information are used to complete the reclassification of abnormal cervical cells, and the abnormal cells were identified by simulating a doctor comparing the morphology of the normal cells in the sample. Experiments show that the positive cell completion rate, positive cell detection accuracy and sample detection accuracy rate on the cervical cell smear dataset reached 84.7%, 94.6% and 92.4%, respectively.
format Article
id doaj-art-b8a84f2ce2e649f1bc2635dbec1de2d9
institution Kabale University
issn 1007-2683
language zho
publishDate 2022-12-01
publisher Harbin University of Science and Technology Publications
record_format Article
series Journal of Harbin University of Science and Technology
spelling doaj-art-b8a84f2ce2e649f1bc2635dbec1de2d92025-08-20T03:51:40ZzhoHarbin University of Science and Technology PublicationsJournal of Harbin University of Science and Technology1007-26832022-12-01270610311410.15938/j.jhust.2022.06.013A Method for Identifying Cervical Abnormal Cells Based on Sample Benchmark ValuesZHAO Si-qi0LIANG Yi-qin1QIN Jian2HE Yong-jun3School of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, ChinaThe identification of cervical abnormal cells using deep learning methods usually requires a large amount of training data, but these data inevitably use different samples of cervical abnormal cells to participate in model training, and naturally miss the positive and abnormal intracellular controls of a single sample, resulting in the fact that recognition accuracy of cervical abnormal cells is not high, and the false positive rate is high. To solve this problem, this paper proposes a method for identifying cervical abnormal cells based on sample benchmark values. Firstly, this method identifies cervical cells and segments cervical cell nucleus by Mask R-CNN model. Then we calculate the key cervical nucleus indicators, propose the concept of benchmark cells, define the sample benchmark values, and quantify the diagnostic criteria.Finally, the abnormal nucleus indicator and model information are used to complete the reclassification of abnormal cervical cells, and the abnormal cells were identified by simulating a doctor comparing the morphology of the normal cells in the sample. Experiments show that the positive cell completion rate, positive cell detection accuracy and sample detection accuracy rate on the cervical cell smear dataset reached 84.7%, 94.6% and 92.4%, respectively.https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2164cervical abnormal cell recognitionsample benchmark valuesdeep learningmachine learning
spellingShingle ZHAO Si-qi
LIANG Yi-qin
QIN Jian
HE Yong-jun
A Method for Identifying Cervical Abnormal Cells Based on Sample Benchmark Values
Journal of Harbin University of Science and Technology
cervical abnormal cell recognition
sample benchmark values
deep learning
machine learning
title A Method for Identifying Cervical Abnormal Cells Based on Sample Benchmark Values
title_full A Method for Identifying Cervical Abnormal Cells Based on Sample Benchmark Values
title_fullStr A Method for Identifying Cervical Abnormal Cells Based on Sample Benchmark Values
title_full_unstemmed A Method for Identifying Cervical Abnormal Cells Based on Sample Benchmark Values
title_short A Method for Identifying Cervical Abnormal Cells Based on Sample Benchmark Values
title_sort method for identifying cervical abnormal cells based on sample benchmark values
topic cervical abnormal cell recognition
sample benchmark values
deep learning
machine learning
url https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2164
work_keys_str_mv AT zhaosiqi amethodforidentifyingcervicalabnormalcellsbasedonsamplebenchmarkvalues
AT liangyiqin amethodforidentifyingcervicalabnormalcellsbasedonsamplebenchmarkvalues
AT qinjian amethodforidentifyingcervicalabnormalcellsbasedonsamplebenchmarkvalues
AT heyongjun amethodforidentifyingcervicalabnormalcellsbasedonsamplebenchmarkvalues
AT zhaosiqi methodforidentifyingcervicalabnormalcellsbasedonsamplebenchmarkvalues
AT liangyiqin methodforidentifyingcervicalabnormalcellsbasedonsamplebenchmarkvalues
AT qinjian methodforidentifyingcervicalabnormalcellsbasedonsamplebenchmarkvalues
AT heyongjun methodforidentifyingcervicalabnormalcellsbasedonsamplebenchmarkvalues