Feasibility study of semi-automated measurements of finger joint space widths

• Published in: Rheumatology international 2011-10, Vol.31 (10), p.1349-1354
• Main Authors: Pfeil, Alexander, Sommerfeld, Julia, Fröber, Rosemarie, Lehmann, Gabriele, Malich, Ansgar, Hansch, Andreas, Wolf, Gunter, Böttcher, Joachim
• Format: Article
• Language: English
• Subjects: Arthritis, Rheumatoid - diagnostic imaging; Arthrography - methods; Arthrography - standards; Cadaver; Diagnosis, Computer-Assisted - methods; Diagnosis, Computer-Assisted - standards; Feasibility Studies; Finger Joint - diagnostic imaging; Humans; Male; Medicine; Medicine & Public Health; Original Article; Osteoarthritis - diagnostic imaging; Radiation Dosage; Rheumatology; X-Ray Film - standards
• ISSN: 0172-8172; 1437-160X
• DOI: 10.1007/s00296-010-1468-4

The purpose of this study is to evaluate technical feasibility based on image capturing conditions (film-focus distance (FFD), film sensitivity, film brand, exposure level and tube voltage) that potentially alter radiographs and consequently may influence the semi-automated measurement of joint space distance (JSD) by computer-aided joint space analysis (CAJSA) in rheumatoid arthritis and osteoarthritis. The radiographs of a left hand (deceased man) were acquired under systematically changing image capturing conditions (exposure level: 4–8 mAs; FFD: 90–130 cm; film sensitivity: 200/400 and tube voltage: 40–52 kV with different image modalities: conventional radiographs, original digital radiographs, digital print-outs). All JSD-measurements were performed with the CAJSA-technology (Radiogrammetry Kit, Version 1.3.6; Sectra; Sweden) at the metacarpal-phalangeal articulation. JSD-analysis was not influenced by changes of FFD, exposure level, film sensitivity or film brand. JSD showed significant variation caused by tube voltage (conventional: CV = 1.913% for Agfa and CV = 2.448% for Kodak; digital: CV = 0.741% for Philips print-outs and CV = 0.620% with original digital images versus CV = 2.185% for Siemens print-outs and 0.951% with original digital images). Computer-aided joint space analysis for JSD-measurements is unaffected by the following image capturing parameters: film-focus distance, film sensitivity, film brand and exposure level. An influence of tube voltage was detected in a lesser extent for original digital images compared to the printed digital as well as conventional versions. Consequently, a standardized tube voltage is essential for accurate reproductions of CAJSA-measurements in rheumatoid arthritis and osteoarthritis.