• Title of article

    Computer-aided quantification of focal cartilage lesions of osteoarthritic knee using MRI

  • Author/Authors

    Lee، نويسنده , , Keh-Yang and Dunn، نويسنده , , Timothy C. and Steinbach، نويسنده , , Lynne S. and Ozhinsky، نويسنده , , Eugene and Ries، نويسنده , , Michael D. and Majumdar، نويسنده , , Sharmila، نويسنده ,

  • Issue Information
    روزنامه با شماره پیاپی سال 2004
  • Pages
    11
  • From page
    1105
  • To page
    1115
  • Abstract
    Noninvasive assessment of articular cartilage using magnetic resonance imaging (MRI) has gained popularity in the diagnosis of osteoarthritis (OA), a condition that affects 20 million Americans. Focal cartilage lesions, a defect found in roughly 19% of the OA population, currently can only be evaluated with confidence using minimally invasive arthroscopy. This article presents a computer-aided procedure using MRI to quantify focal cartilage lesions and aims to support clinical practices of diagnosis and monitoring of lesion progress. Upon a local minima search for identifying focal lesions, the proposed gradient peak method outlines lesion boundaries and then generates morphological properties, such as lesion volume and lesion area. The procedure was evaluated using simulated and in vivo data. First, a simulated lesion was created and analyzed, and the results were compared with the exact solutions. Second, an in vivo evaluation was carried out on seven human knees in which nine focal lesions were identified and quantified. Three of the subjects had follow-up analyses, at either 1 or 2 years. Finally, in an attempt to characterize local biochemical changes underlying focal lesions, MR-derived T2 values of defective cartilage within the lesion boundaries were examined and compared with the values of adjacent cartilage compartments.
  • Keywords
    Focal cartilage lesion , MAGNETIC RESONANCE IMAGING , Quantification , Osteoarthritis
  • Journal title
    Magnetic Resonance Imaging
  • Serial Year
    2004
  • Journal title
    Magnetic Resonance Imaging
  • Record number

    1832020