Cargando…

HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy

Antral follicle Count (AFC) is a non-invasive biomarker used to assess ovarian reserves through transvaginal ultrasound (TVUS) imaging. Antral follicles’ diameter is usually in the range of 2–10 mm. The primary aim of ovarian reserve monitoring is to measure the size of ovarian follicles and the num...

Descripción completa

Detalles Bibliográficos
Autores principales: Singh, Vivek Kumar, Yousef Kalafi, Elham, Cheah, Eugene, Wang, Shuhang, Wang, Jingchao, Ozturk, Arinc, Li, Qian, Eldar, Yonina C., Samir, Anthony E., Kumar, Viksit
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9777827/
https://www.ncbi.nlm.nih.gov/pubmed/36553220
http://dx.doi.org/10.3390/diagnostics12123213
_version_ 1784856202818093056
author Singh, Vivek Kumar
Yousef Kalafi, Elham
Cheah, Eugene
Wang, Shuhang
Wang, Jingchao
Ozturk, Arinc
Li, Qian
Eldar, Yonina C.
Samir, Anthony E.
Kumar, Viksit
author_facet Singh, Vivek Kumar
Yousef Kalafi, Elham
Cheah, Eugene
Wang, Shuhang
Wang, Jingchao
Ozturk, Arinc
Li, Qian
Eldar, Yonina C.
Samir, Anthony E.
Kumar, Viksit
author_sort Singh, Vivek Kumar
collection PubMed
description Antral follicle Count (AFC) is a non-invasive biomarker used to assess ovarian reserves through transvaginal ultrasound (TVUS) imaging. Antral follicles’ diameter is usually in the range of 2–10 mm. The primary aim of ovarian reserve monitoring is to measure the size of ovarian follicles and the number of antral follicles. Manual follicle measurement is inhibited by operator time, expertise and the subjectivity of delineating the two axes of the follicles. This necessitates an automated framework capable of quantifying follicle size and count in a clinical setting. This paper proposes a novel Harmonic Attention-based U-Net network, HaTU-Net, to precisely segment the ovary and follicles in ultrasound images. We replace the standard convolution operation with a harmonic block that convolves the features with a window-based discrete cosine transform (DCT). Additionally, we proposed a harmonic attention mechanism that helps to promote the extraction of rich features. The suggested technique allows for capturing the most relevant features, such as boundaries, shape, and textural patterns, in the presence of various noise sources (i.e., shadows, poor contrast between tissues, and speckle noise). We evaluated the proposed model on our in-house private dataset of 197 patients undergoing TransVaginal UltraSound (TVUS) exam. The experimental results on an independent test set confirm that HaTU-Net achieved a Dice coefficient score of [Formula: see text] for ovaries and [Formula: see text] for antral follicles, an improvement of [Formula: see text] and [Formula: see text] , respectively, when compared to a standard U-Net. Further, we accurately measure the follicle size, yielding the recall, and precision rates of [Formula: see text] and [Formula: see text] , respectively.
format Online
Article
Text
id pubmed-9777827
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-97778272022-12-23 HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy Singh, Vivek Kumar Yousef Kalafi, Elham Cheah, Eugene Wang, Shuhang Wang, Jingchao Ozturk, Arinc Li, Qian Eldar, Yonina C. Samir, Anthony E. Kumar, Viksit Diagnostics (Basel) Article Antral follicle Count (AFC) is a non-invasive biomarker used to assess ovarian reserves through transvaginal ultrasound (TVUS) imaging. Antral follicles’ diameter is usually in the range of 2–10 mm. The primary aim of ovarian reserve monitoring is to measure the size of ovarian follicles and the number of antral follicles. Manual follicle measurement is inhibited by operator time, expertise and the subjectivity of delineating the two axes of the follicles. This necessitates an automated framework capable of quantifying follicle size and count in a clinical setting. This paper proposes a novel Harmonic Attention-based U-Net network, HaTU-Net, to precisely segment the ovary and follicles in ultrasound images. We replace the standard convolution operation with a harmonic block that convolves the features with a window-based discrete cosine transform (DCT). Additionally, we proposed a harmonic attention mechanism that helps to promote the extraction of rich features. The suggested technique allows for capturing the most relevant features, such as boundaries, shape, and textural patterns, in the presence of various noise sources (i.e., shadows, poor contrast between tissues, and speckle noise). We evaluated the proposed model on our in-house private dataset of 197 patients undergoing TransVaginal UltraSound (TVUS) exam. The experimental results on an independent test set confirm that HaTU-Net achieved a Dice coefficient score of [Formula: see text] for ovaries and [Formula: see text] for antral follicles, an improvement of [Formula: see text] and [Formula: see text] , respectively, when compared to a standard U-Net. Further, we accurately measure the follicle size, yielding the recall, and precision rates of [Formula: see text] and [Formula: see text] , respectively. MDPI 2022-12-18 /pmc/articles/PMC9777827/ /pubmed/36553220 http://dx.doi.org/10.3390/diagnostics12123213 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Singh, Vivek Kumar
Yousef Kalafi, Elham
Cheah, Eugene
Wang, Shuhang
Wang, Jingchao
Ozturk, Arinc
Li, Qian
Eldar, Yonina C.
Samir, Anthony E.
Kumar, Viksit
HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy
title HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy
title_full HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy
title_fullStr HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy
title_full_unstemmed HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy
title_short HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy
title_sort hatu-net: harmonic attention network for automated ovarian ultrasound quantification in assisted pregnancy
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9777827/
https://www.ncbi.nlm.nih.gov/pubmed/36553220
http://dx.doi.org/10.3390/diagnostics12123213
work_keys_str_mv AT singhvivekkumar hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT yousefkalafielham hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT cheaheugene hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT wangshuhang hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT wangjingchao hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT ozturkarinc hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT liqian hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT eldaryoninac hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT samiranthonye hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy
AT kumarviksit hatunetharmonicattentionnetworkforautomatedovarianultrasoundquantificationinassistedpregnancy