Cover Image

Sperm-cell detection using YOLOv5 architecture

Infertility has become a severe health issue in recent years. Sperm morphology, sperm motility, and sperm density are the most critical factors in male infertility. As a result, sperm motility, density, and morphology are examined in semen analysis carried out by laboratory professionals. However, a...

Full description

Saved in:
Bibliographic Details
Main Authors: Dobrovolny, Michal, Benes, Jakub, Krejcar, Ondrej, Selamat, Ali
Format: Article
Published: Springer Science and Business Media Deutschland GmbH 2022
Subjects:
T Technology (General)
Online Access:http://eprints.utm.my/id/eprint/100500/
http://dx.doi.org/10.1007/978-3-031-07802-6_27
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.utm.100500
record_format eprints
spelling my.utm.1005002023-04-14T02:17:21Z http://eprints.utm.my/id/eprint/100500/ Sperm-cell detection using YOLOv5 architecture Dobrovolny, Michal Benes, Jakub Krejcar, Ondrej Selamat, Ali T Technology (General) Infertility has become a severe health issue in recent years. Sperm morphology, sperm motility, and sperm density are the most critical factors in male infertility. As a result, sperm motility, density, and morphology are examined in semen analysis carried out by laboratory professionals. However, applying a subjective analysis based on laboratory observation is easy to make a mistake. To reduce the effect of specialists in semen analysis, a computer-aided sperm count estimation approach is proposed in this work. The quantity of active sperm in the semen is determined using object detection methods focusing on sperm motility. The proposed strategy was tested using data from the Visem dataset provided by Association for Computing Machinery. We created a small sample custom dataset to prove that our network will be able to detect sperms in images. The best not-super tuned result is mAP 72.15. Springer Science and Business Media Deutschland GmbH 2022 Article PeerReviewed Dobrovolny, Michal and Benes, Jakub and Krejcar, Ondrej and Selamat, Ali (2022) Sperm-cell detection using YOLOv5 architecture. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 13347 (NA). pp. 319-330. ISSN 0302-9743 http://dx.doi.org/10.1007/978-3-031-07802-6_27 DOI : 10.1007/978-3-031-07802-6_27
institution Universiti Teknologi Malaysia
building UTM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Malaysia
content_source UTM Institutional Repository
url_provider http://eprints.utm.my/
topic T Technology (General)
spellingShingle T Technology (General)
Dobrovolny, Michal
Benes, Jakub
Krejcar, Ondrej
Selamat, Ali
Sperm-cell detection using YOLOv5 architecture
description Infertility has become a severe health issue in recent years. Sperm morphology, sperm motility, and sperm density are the most critical factors in male infertility. As a result, sperm motility, density, and morphology are examined in semen analysis carried out by laboratory professionals. However, applying a subjective analysis based on laboratory observation is easy to make a mistake. To reduce the effect of specialists in semen analysis, a computer-aided sperm count estimation approach is proposed in this work. The quantity of active sperm in the semen is determined using object detection methods focusing on sperm motility. The proposed strategy was tested using data from the Visem dataset provided by Association for Computing Machinery. We created a small sample custom dataset to prove that our network will be able to detect sperms in images. The best not-super tuned result is mAP 72.15.
format Article
author Dobrovolny, Michal
Benes, Jakub
Krejcar, Ondrej
Selamat, Ali
author_facet Dobrovolny, Michal
Benes, Jakub
Krejcar, Ondrej
Selamat, Ali
author_sort Dobrovolny, Michal
title Sperm-cell detection using YOLOv5 architecture
title_short Sperm-cell detection using YOLOv5 architecture
title_full Sperm-cell detection using YOLOv5 architecture
title_fullStr Sperm-cell detection using YOLOv5 architecture
title_full_unstemmed Sperm-cell detection using YOLOv5 architecture
title_sort sperm-cell detection using yolov5 architecture
publisher Springer Science and Business Media Deutschland GmbH
publishDate 2022
url http://eprints.utm.my/id/eprint/100500/
http://dx.doi.org/10.1007/978-3-031-07802-6_27
_version_ 1764222575676751872
score 13.211869

Similar Items