Deep learning-based classification and structure name standardization for organ at risk and target delineations in prostate cancer radiotherapy

Christian Jamtheim Gustafsson, Michael Lempart, Johan Swärd, Emilia Persson, Tufve Nyholm, Camilla Thellenberg Karlsson, Jonas Scherman

Research output: Contribution to journalArticlepeer-review


Radiotherapy (RT) datasets can suffer from variations in annotation of organ at risk (OAR) and target structures. Annotation standards exist, but their description for prostate targets is limited. This restricts the use of such data for supervised machine learning purposes as it requires properly annotated data. The aim of this work was to develop a modality independent deep learning (DL) model for automatic classification and annotation of prostate RT DICOM structures. Delineated prostate organs at risk (OAR), support- and target structures (gross tumor volume [GTV]/clinical target volume [CTV]/planning target volume [PTV]), along with or without separate vesicles and/or lymph nodes, were extracted as binary masks from 1854 patients. An image modality independent 2D InceptionResNetV2 classification network was trained with varying amounts of training data using four image input channels. Channel 1–3 consisted of orthogonal 2D projections from each individual binary structure. The fourth channel contained a summation of the other available binary structure masks. Structure classification performance was assessed in independent CT (n = 200 pat) and magnetic resonance imaging (MRI) (n = 40 pat) test datasets and an external CT (n = 99 pat) dataset from another clinic. A weighted classification accuracy of 99.4% was achieved during training. The unweighted classification accuracy and the weighted average F1 score among different structures in the CT test dataset were 98.8% and 98.4% and 98.6% and 98.5% for the MRI test dataset, respectively. The external CT dataset yielded the corresponding results 98.4% and 98.7% when analyzed for trained structures only, and results from the full dataset yielded 79.6% and 75.2%. Most misclassifications in the external CT dataset occurred due to multiple CTVs and PTVs being fused together, which was not included in the training data. Our proposed DL-based method for automated renaming and standardization of prostate radiotherapy annotations shows great potential. Clinic specific contouring standards however need to be represented in the training data for successful use. Source code is available at

Original languageEnglish
Pages (from-to)51-63
JournalJournal of Applied Clinical Medical Physics
Issue number12
Early online date2021
Publication statusPublished - 2021

Subject classification (UKÄ)

  • Radiology, Nuclear Medicine and Medical Imaging

Free keywords

  • classification
  • deep learning
  • machine learning
  • radiotherapy
  • structure


Dive into the research topics of 'Deep learning-based classification and structure name standardization for organ at risk and target delineations in prostate cancer radiotherapy'. Together they form a unique fingerprint.

Cite this