The impact of multicentric datasets for the automated tumor delineation in primary prostate cancer using convolutional neural networks on 18F-PSMA-1007 PET

Research output: Contribution to journalResearch articleContributedpeer-review

Contributors

  • Julius C. Holzschuh - , University Medical Center Freiburg, German Cancer Research Center (DKFZ) (Author)
  • Michael Mix - , University Medical Center Freiburg (Author)
  • Martin T. Freitag - , University Medical Center Freiburg (Author)
  • Tobias Hölscher - , Department of Radiotherapy and Radiooncology, University Hospital Carl Gustav Carus Dresden (Author)
  • Anja Braune - , Department of Nuclear Medicine, University Hospital Carl Gustav Carus Dresden (Author)
  • Jörg Kotzerke - , Department of Nuclear Medicine, University Hospital Carl Gustav Carus Dresden (Author)
  • Alexis Vrachimis - , European University Cyprus (Author)
  • Paul Doolan - , European University Cyprus (Author)
  • Harun Ilhan - , Ludwig Maximilian University of Munich (Author)
  • Ioana M. Marinescu - , University Medical Center Freiburg (Author)
  • Simon K.B. Spohn - , University Medical Center Freiburg (Author)
  • Tobias Fechter - , University Medical Center Freiburg (Author)
  • Dejan Kuhn - , University Medical Center Freiburg (Author)
  • Christian Gratzke - , University Medical Center Freiburg (Author)
  • Radu Grosu - , Vienna University of Technology, Stony Brook University (Author)
  • Anca Ligia Grosu - , University Medical Center Freiburg (Author)
  • C. Zamboglou - , University Medical Center Freiburg, European University Cyprus (Author)

Abstract

Purpose: Convolutional Neural Networks (CNNs) have emerged as transformative tools in the field of radiation oncology, significantly advancing the precision of contouring practices. However, the adaptability of these algorithms across diverse scanners, institutions, and imaging protocols remains a considerable obstacle. This study aims to investigate the effects of incorporating institution-specific datasets into the training regimen of CNNs to assess their generalization ability in real-world clinical environments. Focusing on a data-centric analysis, the influence of varying multi- and single center training approaches on algorithm performance is conducted. Methods: nnU-Net is trained using a dataset comprising 161 18F-PSMA-1007 PET images collected from four distinct institutions (Freiburg: n = 96, Munich: n = 19, Cyprus: n = 32, Dresden: n = 14). The dataset is partitioned such that data from each center are systematically excluded from training and used solely for testing to assess the model's generalizability and adaptability to data from unfamiliar sources. Performance is compared through a 5-Fold Cross-Validation, providing a detailed comparison between models trained on datasets from single centers to those trained on aggregated multi-center datasets. Dice Similarity Score, Hausdorff distance and volumetric analysis are used as primary evaluation metrics. Results: The mixed training approach yielded a median DSC of 0.76 (IQR: 0.64–0.84) in a five-fold cross-validation, showing no significant differences (p = 0.18) compared to models trained with data exclusion from each center, which performed with a median DSC of 0.74 (IQR: 0.56–0.86). Significant performance improvements regarding multi-center training were observed for the Dresden cohort (multi-center median DSC 0.71, IQR: 0.58–0.80 vs. single-center 0.68, IQR: 0.50–0.80, p < 0.001) and Cyprus cohort (multi-center 0.74, IQR: 0.62–0.83 vs. single-center 0.72, IQR: 0.54–0.82, p < 0.01). While Munich and Freiburg also showed performance improvements with multi-center training, results showed no statistical significance (Munich: multi-center DSC 0.74, IQR: 0.60–0.80 vs. single-center 0.72, IQR: 0.59–0.82, p > 0.05; Freiburg: multi-center 0.78, IQR: 0.53–0.87 vs. single-center 0.71, IQR: 0.53–0.83, p = 0.23). Conclusion: CNNs trained for auto contouring intraprostatic GTV in 18F-PSMA-1007 PET on a diverse dataset from multiple centers mostly generalize well to unseen data from other centers. Training on a multicentric dataset can improve performance compared to training exclusively with a single-center dataset regarding intraprostatic 18F-PSMA-1007 PET GTV segmentation. The segmentation performance of the same CNN can vary depending on the dataset employed for training and testing.

Details

Original languageEnglish
Article number106
JournalRadiation oncology
Volume19
Issue number1
Publication statusPublished - Dec 2024
Peer-reviewedYes

External IDs

PubMed 39113123

Keywords

Sustainable Development Goals