PMID- 37928618
OWN - NLM
STAT- PubMed-not-MEDLINE
LR  - 20231107
IS  - 2405-6316 (Electronic)
IS  - 2405-6316 (Linking)
VI  - 28
DP  - 2023 Oct
TI  - Prior knowledge based deep learning auto-segmentation in magnetic resonance 
      imaging-guided radiotherapy of prostate cancer.
PG  - 100498
LID - 10.1016/j.phro.2023.100498 [doi]
LID - 100498
AB  - BACKGROUND AND PURPOSE: Automation is desirable for organ segmentation in 
      radiotherapy. This study compared deep learning methods for auto-segmentation of 
      organs-at-risk (OARs) and clinical target volume (CTV) in prostate cancer 
      patients undergoing fractionated magnetic resonance (MR)-guided adaptive 
      radiation therapy. Models predicting dense displacement fields (DDFMs) between 
      planning and fraction images were compared to patient-specific (PSM) and baseline 
      (BM) segmentation models. MATERIALS AND METHODS: A dataset of 92 patients with 
      planning and fraction MR images (MRIs) from two institutions were used. DDFMs 
      were trained to predict dense displacement fields (DDFs) between the planning and 
      fraction images, which were subsequently used to propagate the planning contours 
      of the bladder, rectum, and CTV to the daily MRI. The training was performed 
      either with true planning-fraction image pairs or with planning images and their 
      counterparts deformed by known DDFs. The BMs were trained on 53 planning images, 
      while to generate PSMs, the BMs were fine-tuned using the planning image of a 
      given single patient. The evaluation included Dice similarity coefficient (DSC), 
      the average (HD(avg)) and the 95th percentile (HD(95)) Hausdorff distance (HD). 
      RESULTS: The DDFMs with DSCs for bladder/rectum of 0.76/0.76 performed worse than 
      PSMs (0.91/0.90) and BMs (0.89/0.88). The same trend was observed for HDs. For 
      CTV, DDFM and PSM performed similarly yielding DSCs of 0.87 and 0.84, 
      respectively. CONCLUSIONS: DDFMs were found suitable for CTV delineation after 
      rigid alignment. However, for OARs they were outperformed by PSMs, as they 
      predicted only limited deformations even in the presence of substantial 
      anatomical changes.
CI  - (c) 2023 The Author(s).
FAU - Kawula, Maria
AU  - Kawula M
AD  - Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, 
      Germany.
FAU - Vagni, Marica
AU  - Vagni M
AD  - Fondazione Policlinico Universitario "Agostino Gemelli" IRCCS, Rome, Italy.
FAU - Cusumano, Davide
AU  - Cusumano D
AD  - Fondazione Policlinico Universitario "Agostino Gemelli" IRCCS, Rome, Italy.
AD  - Mater Olbia Hospital, Olbia (SS), Italy.
FAU - Boldrini, Luca
AU  - Boldrini L
AD  - Fondazione Policlinico Universitario "Agostino Gemelli" IRCCS, Rome, Italy.
FAU - Placidi, Lorenzo
AU  - Placidi L
AD  - Fondazione Policlinico Universitario "Agostino Gemelli" IRCCS, Rome, Italy.
FAU - Corradini, Stefanie
AU  - Corradini S
AD  - Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, 
      Germany.
FAU - Belka, Claus
AU  - Belka C
AD  - Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, 
      Germany.
AD  - German Cancer Consortium (DKTK), Partner Site Munich, A Partnership Between DKFZ 
      and LMU University Hospital Munich, Germany.
AD  - Bavarian Cancer Research Center (BZKF), Munich, Germany.
FAU - Landry, Guillaume
AU  - Landry G
AD  - Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, 
      Germany.
FAU - Kurz, Christopher
AU  - Kurz C
AD  - Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, 
      Germany.
LA  - eng
PT  - Journal Article
DEP - 20231010
PL  - Netherlands
TA  - Phys Imaging Radiat Oncol
JT  - Physics and imaging in radiation oncology
JID - 101704276
PMC - PMC10624570
OTO - NOTNLM
OT  - Auto-segmentation
OT  - Deep learning
OT  - MR-Linac
OT  - MRgRT
OT  - Patient-specific models
OT  - Prostate cancer
OT  - Spatial transformer layer
COIS- The authors declare that they have no known competing financial interests or 
      personal relationships that could have appeared to influence the work reported in 
      this paper. The Department of Radiation Oncology of the University Hospital of 
      LMU Munich has a research agreement with ViewRay. ViewRay did not fund this study 
      and was not involved and had no influence on the study design, the collection or 
      analysis of data, or on the writing of the manuscript.
EDAT- 2023/11/06 06:42
MHDA- 2023/11/06 06:43
PMCR- 2023/10/10
CRDT- 2023/11/06 04:37
PHST- 2023/05/30 00:00 [received]
PHST- 2023/10/03 00:00 [revised]
PHST- 2023/10/04 00:00 [accepted]
PHST- 2023/11/06 06:43 [medline]
PHST- 2023/11/06 06:42 [pubmed]
PHST- 2023/11/06 04:37 [entrez]
PHST- 2023/10/10 00:00 [pmc-release]
AID - S2405-6316(23)00089-1 [pii]
AID - 100498 [pii]
AID - 10.1016/j.phro.2023.100498 [doi]
PST - epublish
SO  - Phys Imaging Radiat Oncol. 2023 Oct 10;28:100498. doi: 
      10.1016/j.phro.2023.100498. eCollection 2023 Oct.