Research
Print page Print page
Switch language
The Capital Region of Denmark - a part of Copenhagen University Hospital
Published

Comparison of 11 automated PET segmentation methods in lymphoma

Research output: Contribution to journalJournal articleResearchpeer-review

DOI

  1. Simulating effects of brain atrophy in longitudinal PET imaging with an anthropomorphic brain phantom

    Research output: Contribution to journalJournal articleResearchpeer-review

  2. Repeated diffusion MRI reveals earliest time point for stratification of radiotherapy response in brain metastases

    Research output: Contribution to journalJournal articleResearchpeer-review

  3. Optimized MLAA for quantitative non-TOF PET/MR of the brain

    Research output: Contribution to journalJournal articleResearchpeer-review

  4. Comment on: 'A Poisson resampling method for simulating reduced counts in nuclear medicine images'

    Research output: Contribution to journalJournal articleResearchpeer-review

  • Amy J Weisman
  • Minnie W Kieler
  • Scott Perlman
  • Martin Hutchings
  • Robert Jeraj
  • Lale Kostakoglu
  • Tyler J Bradshaw
View graph of relations

Segmentation of lymphoma lesions in FDG PET/CT images is critical in both assessing individual lesions and quantifying patient disease burden. Simple thresholding methods remain common despite the large heterogeneity in lymphoma lesion location, size, and contrast. Here, we assess 11 automated PET segmentation methods for their use in two scenarios: individual lesion segmentation and patient-level disease quantification in lymphoma. Lesions on 18F-FDG PET/CT scans of 90 lymphoma patients were contoured by a nuclear medicine physician. Thresholding, active contours, clustering, adaptive region-growing, and convolutional neural network (CNN) methods were implemented on all physician-identified lesions. Lesion-level segmentation was evaluated using multiple segmentation performance metrics (Dice, Hausdorff Distance). Patient-level quantification of total disease burden (SUVtotal) and metabolic tumor volume (MTV) was assessed using Spearman's correlation coefficients between the segmentation output and physician contours. Lesion segmentation and patient quantification performance was compared to inter-physician agreement in a subset of 20 patients segmented by a second nuclear medicine physician. In total, 1223 lesions with median tumor-to-background ratio of 4.0 and volume of 1.8 cm3, were evaluated. When assessed for lesion segmentation, a 3D CNN, DeepMedic, achieved the highest performance across all evaluation metrics. DeepMedic, clustering methods, and an iterative threshold method had lesion-level segmentation performance comparable to the degree of inter-physician agreement. For patient-level SUVtotal and MTV quantification, all methods except 40% and 50% SUVmax and adaptive region-growing achieved a performance that was similar the agreement of the two physicians. Multiple methods, including a 3D CNN, clustering, and an iterative threshold method, achieved both good lesion-level segmentation and patient-level quantification performance in a population of 90 lymphoma patients. These methods are thus recommended over thresholding methods such as 40% and 50% SUVmax, which were consistently found to be significantly outside the limits defined by inter-physician agreement.

Original languageEnglish
JournalPhysics in Medicine and Biology
Volume65
Issue number23
Pages (from-to)235019
ISSN0031-9155
DOIs
Publication statusPublished - 27 Nov 2020

ID: 62112065