Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Journal article
Title: Automatic quality assessment of transperineal ultrasound images of the male pelvic region, using deep learning
Author: 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.
Citation: Ultrasound in Medicine and Biology, 2020; 46(2):445-454
Publisher: Elsevier
Issue Date: 2020
ISSN: 0301-5629
Statement of
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 and D. Fontanarosa
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
Transperineal ultrasound imaging
Rights: © 2019 World Federation for Ultrasound in Medicine & Biology.
DOI: 10.1016/j.ultrasmedbio.2019.10.027
Grant ID:
Published version:
Appears in Collections:Aurora harvest 4
Computer Science publications

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.