Morphological components detection for super-depth-of-field bio-micrograph based on deep learning

Xiaohui Du; Xiangzhou Wang; Fan Xu; Jing Zhang; Yibo Huo; Guangmin Ni; Ruqian Hao; Juanxiu Liu; Lin Liu

doi:10.1093/jmicro/dfab033

Morphological components detection for super-depth-of-field bio-micrograph based on deep learning

Xiaohui Du, Xiangzhou Wang, Fan Xu^*, Jing Zhang^*, Yibo Huo, Guangmin Ni, Ruqian Hao, Juanxiu Liu, Lin Liu

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

3 Citations (Scopus)

Abstract

Accompanied with the clinical routine examination demand increase sharply, the efficiency and accuracy are the first priority. However, automatic classification and localization of cells in microscopic images in super depth of Field (SDoF) system remains great challenges. In this paper, we advance an object detection algorithm for cells in the SDoF micrograph based on Retinanet model. Compared with the current mainstream algorithm, the mean average precision (mAP) index is significantly improved. In the experiment of leucorrhea samples and fecal samples, mAP indexes are 83.1% and 88.1%, respectively, with an average increase of 10%. The object detection model proposed in this paper can be applied to feces and leucorrhea detection equipment, and significantly improve the detection efficiency and accuracy.

Original language	English
Pages (from-to)	50-59
Number of pages	10
Journal	Microscopy (Oxford, England)
Volume	71
Issue number	1
DOIs	https://doi.org/10.1093/jmicro/dfab033
Publication status	Published - 1 Jan 2022
Externally published	Yes

Keywords

Ritinanet
microscopy
object detection
super-depth-of-field

Access to Document

10.1093/jmicro/dfab033

Cite this

@article{648f4665b486457a9da8c87b83b13355,

title = "Morphological components detection for super-depth-of-field bio-micrograph based on deep learning",

abstract = "Accompanied with the clinical routine examination demand increase sharply, the efficiency and accuracy are the first priority. However, automatic classification and localization of cells in microscopic images in super depth of Field (SDoF) system remains great challenges. In this paper, we advance an object detection algorithm for cells in the SDoF micrograph based on Retinanet model. Compared with the current mainstream algorithm, the mean average precision (mAP) index is significantly improved. In the experiment of leucorrhea samples and fecal samples, mAP indexes are 83.1% and 88.1%, respectively, with an average increase of 10%. The object detection model proposed in this paper can be applied to feces and leucorrhea detection equipment, and significantly improve the detection efficiency and accuracy.",

keywords = "Ritinanet, microscopy, object detection, super-depth-of-field",

author = "Xiaohui Du and Xiangzhou Wang and Fan Xu and Jing Zhang and Yibo Huo and Guangmin Ni and Ruqian Hao and Juanxiu Liu and Lin Liu",

note = "Publisher Copyright: {\textcopyright} 2021 The Author(s).",

year = "2022",

month = jan,

day = "1",

doi = "10.1093/jmicro/dfab033",

language = "English",

volume = "71",

pages = "50--59",

journal = "Microscopy (Oxford, England)",

issn = "2050-5698",

publisher = "Oxford University Press",

number = "1",

}

TY - JOUR

T1 - Morphological components detection for super-depth-of-field bio-micrograph based on deep learning

AU - Du, Xiaohui

AU - Wang, Xiangzhou

AU - Xu, Fan

AU - Zhang, Jing

AU - Huo, Yibo

AU - Ni, Guangmin

AU - Hao, Ruqian

AU - Liu, Juanxiu

AU - Liu, Lin

PY - 2022/1/1

Y1 - 2022/1/1

N2 - Accompanied with the clinical routine examination demand increase sharply, the efficiency and accuracy are the first priority. However, automatic classification and localization of cells in microscopic images in super depth of Field (SDoF) system remains great challenges. In this paper, we advance an object detection algorithm for cells in the SDoF micrograph based on Retinanet model. Compared with the current mainstream algorithm, the mean average precision (mAP) index is significantly improved. In the experiment of leucorrhea samples and fecal samples, mAP indexes are 83.1% and 88.1%, respectively, with an average increase of 10%. The object detection model proposed in this paper can be applied to feces and leucorrhea detection equipment, and significantly improve the detection efficiency and accuracy.

AB - Accompanied with the clinical routine examination demand increase sharply, the efficiency and accuracy are the first priority. However, automatic classification and localization of cells in microscopic images in super depth of Field (SDoF) system remains great challenges. In this paper, we advance an object detection algorithm for cells in the SDoF micrograph based on Retinanet model. Compared with the current mainstream algorithm, the mean average precision (mAP) index is significantly improved. In the experiment of leucorrhea samples and fecal samples, mAP indexes are 83.1% and 88.1%, respectively, with an average increase of 10%. The object detection model proposed in this paper can be applied to feces and leucorrhea detection equipment, and significantly improve the detection efficiency and accuracy.

KW - Ritinanet

KW - microscopy

KW - object detection

KW - super-depth-of-field

UR - http://www.scopus.com/inward/record.url?scp=85123879322&partnerID=8YFLogxK

U2 - 10.1093/jmicro/dfab033

DO - 10.1093/jmicro/dfab033

M3 - Article

C2 - 34417804

AN - SCOPUS:85123879322

SN - 2050-5698

VL - 71

SP - 50

EP - 59

JO - Microscopy (Oxford, England)

JF - Microscopy (Oxford, England)

IS - 1

ER -

Morphological components detection for super-depth-of-field bio-micrograph based on deep learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this