Growing Region Segmentation Software (GRES) for quantitative magnetic resonance imaging of multiple sclerosis: Intra- and inter-observer agreement variability: A comparison with manual contouring method

Roberto C. Parodi, Francesco Sardanelli, Paolo Renzetti, Elisabetta Rosso, Caterina Losacco, Alessandra Ferrari, Fabrizio Levrero, Alberto Pilot, Matilde Inglese, Giovanni L. Mancardi

Research output: Contribution to journalArticle

Abstract

Lesion area measurement in multiple sclerosis (MS) is one of the key points in evaluating the natural history and in monitoring the efficacy of treatments. This study was performed to check the intra- and inter-observer agreement variability of a locally developed Growing Region Segmentation Software (GRES), comparing them to those obtained using manual contouring (MC). From routine 1.5-T MRI study of clinically definite multiple sclerosis patients, 36 lesions seen on proton-density-weighted images (PDWI) and 36 enhancing lesion on Gd-DTPA-BMA-enhanced T1-weighted images (Gd-T1WI) were randomly chosen and were evaluated by three observers. The mean range of lesion size was 9.9-536.0 mm2 on PDWI and 3.6-57.2 mm2 on Gd-T1WI. The median intra- and inter-observer agreement were, respectively, 97.1 and 90.0% using GRES on PDWI, 81.0 and 70.0% using MC on PDWI, 88.8 and 80.0% using GRES on Gd-T1WI, and 85.8 and 70.0% using MC on Gd-T1WI. The intra- and inter-observer agreements were significantly greater for GRES compared with MC (P <0.0001 and P = 0.0023, respectively) for PDWI, while no difference was found between GRES an MC for Gd-T1WI. The intra-observer variability for GRES was significantly lower on both PDWI (P = 0.0001) and Gd-T1WI (P = 0.0067), whereas for MC the same result was found only for PDWI (P = 0.0147). These data indicate that GRES reduces both the intra- and the inter-observer variability in assessing the area of MS lesions on PDWI and may prove useful in multicentre studies.

Original languageEnglish
Pages (from-to)866-871
Number of pages6
JournalEuropean Radiology
Volume12
Issue number4
DOIs
Publication statusPublished - Apr 1 2002

Keywords

  • Brain MRI
  • Intra- and inter-observer agreement
  • Lesion volume measurements
  • Multiple sclerosis

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology

Fingerprint Dive into the research topics of 'Growing Region Segmentation Software (GRES) for quantitative magnetic resonance imaging of multiple sclerosis: Intra- and inter-observer agreement variability: A comparison with manual contouring method'. Together they form a unique fingerprint.

  • Cite this