PT159 - Automatic segmentation of the prostate in transrectal ultrasound images using deep learning for application in MRI-TRUS fusion

R.J.G. van Sloun, R.R. Wildeboer, A.W. Postema, M. Gayet, C.K. Mannaerts, H.P. Beerlage, Georg Salomon, H. Wijkstra, M. Mischi (Corresponding author)

Research output: Contribution to journalConference articleAcademicpeer-review

Abstract

Introduction & Objectives: In recent years, prostate biopsy increasingly involves targeting magnetic resonance imaging (MRI)-suspicious lesions after fusion with real-time transrectal ultrasound (TRUS). Such fusion currently requires (semi)manual prostate delineation, burdening clinicians with this lengthy procedure. A reliable automatic prostate segmentation on TRUS is still an unsolved challenge; therefore, here we propose a real-time prostate segmentation algorithm through deep-learning that readily translates between different scanners and user settings.
Original languageEnglish
Pages (from-to)e1880-e1881
Number of pages2
JournalEuropean Urology Supplements
Volume18
Issue number1
DOIs
Publication statusPublished - 2019
Event34th Annual EAU Congress European Association of Urology (EAU19) - Fira Gran Via, Barcelona, Spain
Duration: 15 Mar 201919 Mar 2019
https://eaucongress.uroweb.org/

Bibliographical note

Abstracts EAU19 – 34th Annual EAU Congress

Fingerprint Dive into the research topics of 'PT159 - Automatic segmentation of the prostate in transrectal ultrasound images using deep learning for application in MRI-TRUS fusion'. Together they form a unique fingerprint.

  • Cite this