Automatic Quality Assessment of Transperineal Ultrasound Images of the Male Pelvic Region, Using Deep Learning

S. M. Camps; T. Houben; G. Carneiro; C. Edwards; M. Antico; M. Dunnhofer; E. G.H.J. Martens; J. A. Baeza; B. G.L. Vanneste; E. J. van Limbergen; P. H.N. de With; F. Verhaegen; D. Fontanarosa

doi:10.1016/j.ultrasmedbio.2019.10.027

Automatic Quality Assessment of Transperineal Ultrasound Images of the Male Pelvic Region, Using Deep Learning

S. M. Camps, T. Houben, G. Carneiro, C. Edwards, M. Antico, M. Dunnhofer, E. G.H.J. Martens, J. A. Baeza, B. G.L. Vanneste, E. J. van Limbergen, P. H.N. de With, F. Verhaegen, D. Fontanarosa

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

11 Citations (Scopus)

31 Downloads (Pure)

Abstract

Ultrasound guidance is not in widespread use in prostate cancer radiotherapy workflows. This can be partially attributed to the need for image interpretation by a trained operator during ultrasound image acquisition. In this work, a one-class regressor, based on DenseNet and Gaussian processes, was implemented to automatically assess the quality of transperineal ultrasound images of the male pelvic region. The implemented deep learning approach was tested on 300 transperineal ultrasound images and it achieved a scoring accuracy of 94%, a specificity of 95% and a sensitivity of 92% with respect to the majority vote of 3 experts, which was comparable with the results of these experts. This is the first step toward a fully automatic workflow, which could potentially remove the need for ultrasound image interpretation and make real-time volumetric organ tracking in the radiotherapy environment using ultrasound more appealing.

Original language	English
Pages (from-to)	445-454
Number of pages	10
Journal	Ultrasound in Medicine and Biology
Volume	46
Issue number	2
DOIs	https://doi.org/10.1016/j.ultrasmedbio.2019.10.027
Publication status	Published - Feb 2020
Externally published	Yes

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.ultrasmedbio.2019.10.027

ArticleAccepted author manuscript, 661 KBLicence: CC BY-NC-ND

Cite this

Camps, S. M., Houben, T., Carneiro, G., Edwards, C., Antico, M., Dunnhofer, M., Martens, E. G. H. J., Baeza, J. A., Vanneste, B. G. L., van Limbergen, E. J., de With, P. H. N., Verhaegen, F., & Fontanarosa, D. (2020). Automatic Quality Assessment of Transperineal Ultrasound Images of the Male Pelvic Region, Using Deep Learning. Ultrasound in Medicine and Biology, 46(2), 445-454. https://doi.org/10.1016/j.ultrasmedbio.2019.10.027

@article{47a319ba10d64eaf98dadd34f1d3484f,

title = "Automatic Quality Assessment of Transperineal Ultrasound Images of the Male Pelvic Region, Using Deep Learning",

abstract = "Ultrasound guidance is not in widespread use in prostate cancer radiotherapy workflows. This can be partially attributed to the need for image interpretation by a trained operator during ultrasound image acquisition. In this work, a one-class regressor, based on DenseNet and Gaussian processes, was implemented to automatically assess the quality of transperineal ultrasound images of the male pelvic region. The implemented deep learning approach was tested on 300 transperineal ultrasound images and it achieved a scoring accuracy of 94%, a specificity of 95% and a sensitivity of 92% with respect to the majority vote of 3 experts, which was comparable with the results of these experts. This is the first step toward a fully automatic workflow, which could potentially remove the need for ultrasound image interpretation and make real-time volumetric organ tracking in the radiotherapy environment using ultrasound more appealing.",

keywords = "Deep learning, Image-guided radiotherapy, Prostate, Radiotherapy, Transperineal ultrasound imaging, Ultrasound",

author = "Camps, {S. M.} and T. Houben and G. Carneiro and C. Edwards and M. Antico and M. Dunnhofer and Martens, {E. G.H.J.} and Baeza, {J. A.} and Vanneste, {B. G.L.} and {van Limbergen}, {E. J.} and {de With}, {P. H.N.} and F. Verhaegen and D. Fontanarosa",

note = "Funding Information: The computational resources and services used in this work were provided by the HPC and Research Support Group, Queensland University of Technology, Brisbane, Australia. This research forms part of a project supported by an Australia-India strategic research fund (AISRF53820). G.C. acknowledges the support received by the Australian Research Council's Discovery Projects funding scheme (project DP180103232). Funding Information: The computational resources and services used in this work were provided by the HPC and Research Support Group, Queensland University of Technology , Brisbane, Australia. This research forms part of a project supported by an Australia-India strategic research fund ( AISRF53820 ). G.C. acknowledges the support received by the Australian Research Council's Discovery Projects funding scheme (project DP180103232 ). Publisher Copyright: {\textcopyright} 2019 World Federation for Ultrasound in Medicine & Biology",

year = "2020",

month = feb,

doi = "10.1016/j.ultrasmedbio.2019.10.027",

language = "English",

volume = "46",

pages = "445--454",

journal = "Ultrasound in Medicine and Biology",

issn = "0301-5629",

publisher = "Elsevier USA",

number = "2",

}

Camps, SM, Houben, T, Carneiro, G, Edwards, C, Antico, M, Dunnhofer, M, Martens, EGHJ, Baeza, JA, Vanneste, BGL, van Limbergen, EJ, de With, PHN, Verhaegen, F & Fontanarosa, D 2020, 'Automatic Quality Assessment of Transperineal Ultrasound Images of the Male Pelvic Region, Using Deep Learning', Ultrasound in Medicine and Biology, vol. 46, no. 2, pp. 445-454. https://doi.org/10.1016/j.ultrasmedbio.2019.10.027

TY - JOUR

T1 - Automatic Quality Assessment of Transperineal Ultrasound Images of the Male Pelvic Region, Using Deep Learning

AU - Camps, S. M.

AU - Houben, T.

AU - Carneiro, G.

AU - Edwards, C.

AU - Antico, M.

AU - Dunnhofer, M.

AU - Martens, E. G.H.J.

AU - Baeza, J. A.

AU - Vanneste, B. G.L.

AU - van Limbergen, E. J.

AU - de With, P. H.N.

AU - Verhaegen, F.

AU - Fontanarosa, D.

N1 - Funding Information: The computational resources and services used in this work were provided by the HPC and Research Support Group, Queensland University of Technology, Brisbane, Australia. This research forms part of a project supported by an Australia-India strategic research fund (AISRF53820). G.C. acknowledges the support received by the Australian Research Council's Discovery Projects funding scheme (project DP180103232). Funding Information: The computational resources and services used in this work were provided by the HPC and Research Support Group, Queensland University of Technology , Brisbane, Australia. This research forms part of a project supported by an Australia-India strategic research fund ( AISRF53820 ). G.C. acknowledges the support received by the Australian Research Council's Discovery Projects funding scheme (project DP180103232 ). Publisher Copyright: © 2019 World Federation for Ultrasound in Medicine & Biology

PY - 2020/2

Y1 - 2020/2

N2 - Ultrasound guidance is not in widespread use in prostate cancer radiotherapy workflows. This can be partially attributed to the need for image interpretation by a trained operator during ultrasound image acquisition. In this work, a one-class regressor, based on DenseNet and Gaussian processes, was implemented to automatically assess the quality of transperineal ultrasound images of the male pelvic region. The implemented deep learning approach was tested on 300 transperineal ultrasound images and it achieved a scoring accuracy of 94%, a specificity of 95% and a sensitivity of 92% with respect to the majority vote of 3 experts, which was comparable with the results of these experts. This is the first step toward a fully automatic workflow, which could potentially remove the need for ultrasound image interpretation and make real-time volumetric organ tracking in the radiotherapy environment using ultrasound more appealing.

AB - Ultrasound guidance is not in widespread use in prostate cancer radiotherapy workflows. This can be partially attributed to the need for image interpretation by a trained operator during ultrasound image acquisition. In this work, a one-class regressor, based on DenseNet and Gaussian processes, was implemented to automatically assess the quality of transperineal ultrasound images of the male pelvic region. The implemented deep learning approach was tested on 300 transperineal ultrasound images and it achieved a scoring accuracy of 94%, a specificity of 95% and a sensitivity of 92% with respect to the majority vote of 3 experts, which was comparable with the results of these experts. This is the first step toward a fully automatic workflow, which could potentially remove the need for ultrasound image interpretation and make real-time volumetric organ tracking in the radiotherapy environment using ultrasound more appealing.

KW - Deep learning

KW - Image-guided radiotherapy

KW - Prostate

KW - Radiotherapy

KW - Transperineal ultrasound imaging

KW - Ultrasound

UR - http://www.scopus.com/inward/record.url?scp=85076208417&partnerID=8YFLogxK

U2 - 10.1016/j.ultrasmedbio.2019.10.027

DO - 10.1016/j.ultrasmedbio.2019.10.027

M3 - Article

C2 - 31780240

AN - SCOPUS:85076208417

SN - 0301-5629

VL - 46

SP - 445

EP - 454

JO - Ultrasound in Medicine and Biology

JF - Ultrasound in Medicine and Biology

IS - 2

ER -

Automatic Quality Assessment of Transperineal Ultrasound Images of the Male Pelvic Region, Using Deep Learning

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this