One-class Gaussian process regressor for quality assessment of transperineal ultrasound images

Saskia Camps; Tim Houben; Davide Fontanarosa; Christopher Edwards; Maria Antico; Matteo Dunnhofer; Esther Martens; Jose Baeza; Ben Vanneste; Evert van Limbergen; Peter de With; Frank Verhaegen; Gustavo Carneiro

One-class Gaussian process regressor for quality assessment of transperineal ultrasound images

Saskia Camps, Tim Houben, Davide Fontanarosa, Christopher Edwards, Maria Antico, Matteo Dunnhofer, Esther Martens, Jose Baeza, Ben Vanneste, Evert van Limbergen, Peter de With, Frank Verhaegen, Gustavo Carneiro

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Hoofdstuk › Academic › peer review

1 Downloads (Pure)

Samenvatting

The use of ultrasound guidance in prostate cancer radiotherapy workflows is not widespread. 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 assess automatically the quality of transperineal ultrasound images of the male pelvic region. The implemented deep learning approach achieved a scoring accuracy of 94 a specificity of 953 which was comparable with the results of these experts. This is the first step towards a fully automatic workflow, which could potentially remove the need for image interpretation and thereby make the use of ultrasound imaging, which allows real-time volumetric organ tracking in the RT environment, more appealing for hospitals.

Originele taal-2	Engels
Titel	Proceedings of the 1st International Conference on Medical Imaging with Deep Learning 2018
Redacteuren	I. Isgum, C. Sanchez, G. Litjens
Aantal pagina's	10
Status	Gepubliceerd - 2018
Evenement	1st Conference on Medical Imaging with Deep Learning (MIDL 2018) - Amsterdam, Nederland Duur: 4 jul. 2018 → 6 jul. 2018 https://midl.amsterdam

Congres

Congres	1st Conference on Medical Imaging with Deep Learning (MIDL 2018)
Verkorte titel	MIDL 2018
Land/Regio	Nederland
Stad	Amsterdam
Periode	4/07/18 → 6/07/18
Internet adres	https://midl.amsterdam

Duurzame ontwikkelingsdoelstellingen van de VN

Deze output draagt bij aan de volgende duurzame ontwikkelingsdoelstelling(en)

Toegang tot document

https://eprints.qut.edu.au/120113/

Andere bestanden en links

poster paper

Citeer dit

Camps, S., Houben, T., Fontanarosa, D., Edwards, C., Antico, M., Dunnhofer, M., Martens, E., Baeza, J., Vanneste, B., Limbergen, E. V., With, P. D., Verhaegen, F., & Carneiro, G. (2018). One-class Gaussian process regressor for quality assessment of transperineal ultrasound images. In I. Isgum, C. Sanchez, & G. Litjens (editors), Proceedings of the 1st International Conference on Medical Imaging with Deep Learning 2018 https://eprints.qut.edu.au/120113/

@inbook{f36fbae19b084ca09ce1f38c616667f4,

title = "One-class Gaussian process regressor for quality assessment of transperineal ultrasound images",

abstract = "The use of ultrasound guidance in prostate cancer radiotherapy workflows is not widespread. 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 assess automatically the quality of transperineal ultrasound images of the male pelvic region. The implemented deep learning approach achieved a scoring accuracy of 94 a specificity of 953 which was comparable with the results of these experts. This is the first step towards a fully automatic workflow, which could potentially remove the need for image interpretation and thereby make the use of ultrasound imaging, which allows real-time volumetric organ tracking in the RT environment, more appealing for hospitals.",

author = "Saskia Camps and Tim Houben and Davide Fontanarosa and Christopher Edwards and Maria Antico and Matteo Dunnhofer and Esther Martens and Jose Baeza and Ben Vanneste and Limbergen, {Evert van} and With, {Peter de} and Frank Verhaegen and Gustavo Carneiro",

year = "2018",

language = "English",

editor = "I. Isgum and C. Sanchez and G. Litjens",

booktitle = "Proceedings of the 1st International Conference on Medical Imaging with Deep Learning 2018",

note = "1st Conference on Medical Imaging with Deep Learning (MIDL 2018), MIDL 2018 ; Conference date: 04-07-2018 Through 06-07-2018",

url = "https://midl.amsterdam",

}

Camps, S, Houben, T, Fontanarosa, D, Edwards, C, Antico, M, Dunnhofer, M, Martens, E, Baeza, J, Vanneste, B, Limbergen, EV, With, PD, Verhaegen, F & Carneiro, G 2018, One-class Gaussian process regressor for quality assessment of transperineal ultrasound images. in I Isgum, C Sanchez & G Litjens (uitgave), Proceedings of the 1st International Conference on Medical Imaging with Deep Learning 2018. 1st Conference on Medical Imaging with Deep Learning (MIDL 2018), Amsterdam, Nederland, 4/07/18. <https://eprints.qut.edu.au/120113/>

One-class Gaussian process regressor for quality assessment of transperineal ultrasound images. / Camps, Saskia; Houben, Tim; Fontanarosa, Davide et al.
Proceedings of the 1st International Conference on Medical Imaging with Deep Learning 2018. uitgave / I. Isgum; C. Sanchez; G. Litjens. 2018.

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Hoofdstuk › Academic › peer review

TY - CHAP

T1 - One-class Gaussian process regressor for quality assessment of transperineal ultrasound images

AU - Camps, Saskia

AU - Houben, Tim

AU - Fontanarosa, Davide

AU - Edwards, Christopher

AU - Antico, Maria

AU - Dunnhofer, Matteo

AU - Martens, Esther

AU - Baeza, Jose

AU - Vanneste, Ben

AU - Limbergen, Evert van

AU - With, Peter de

AU - Verhaegen, Frank

AU - Carneiro, Gustavo

PY - 2018

Y1 - 2018

N2 - The use of ultrasound guidance in prostate cancer radiotherapy workflows is not widespread. 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 assess automatically the quality of transperineal ultrasound images of the male pelvic region. The implemented deep learning approach achieved a scoring accuracy of 94 a specificity of 953 which was comparable with the results of these experts. This is the first step towards a fully automatic workflow, which could potentially remove the need for image interpretation and thereby make the use of ultrasound imaging, which allows real-time volumetric organ tracking in the RT environment, more appealing for hospitals.

AB - The use of ultrasound guidance in prostate cancer radiotherapy workflows is not widespread. 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 assess automatically the quality of transperineal ultrasound images of the male pelvic region. The implemented deep learning approach achieved a scoring accuracy of 94 a specificity of 953 which was comparable with the results of these experts. This is the first step towards a fully automatic workflow, which could potentially remove the need for image interpretation and thereby make the use of ultrasound imaging, which allows real-time volumetric organ tracking in the RT environment, more appealing for hospitals.

UR - https://openreview.net/forum?id=r1s0gx3iG

M3 - Chapter

BT - Proceedings of the 1st International Conference on Medical Imaging with Deep Learning 2018

A2 - Isgum, I.

A2 - Sanchez, C.

A2 - Litjens, G.

T2 - 1st Conference on Medical Imaging with Deep Learning (MIDL 2018)

Y2 - 4 July 2018 through 6 July 2018

ER -

One-class Gaussian process regressor for quality assessment of transperineal ultrasound images

Samenvatting

Congres

Duurzame ontwikkelingsdoelstellingen van de VN

Toegang tot document

Andere bestanden en links

Vingerafdruk

Citeer dit