Model-based erosion spotting and visualization in rheumatoid arthritis.

RATIONALE AND OBJECTIVES A method for the automatic detection and the visualization of erosions caused by rheumatoid arthritis is investigated. Erosion-enhanced viewing is a contribution to the computer-aided diagnosis of rheumatoid arthritis. It supports the clinician by providing the automatic marking of erosions and the visualization of any deviations from intact anatomy for a concise reviewing interface. MATERIALS AND METHODS A generative appearance model is used to capture the variability of intact bone and erosions. The algorithm marks erosions on hand radiographs using this model, and visualizes these erosions with the help of the residual appearance error after fitting the model built from intact bone texture. The algorithm was evaluated on 17 hand radiographs. The standard of reference was an annotation of the erosions by a musculoskeletal radiologist. RESULTS Detection results from the algorithm are reported for a set of 17 radiographs of moderately diseased hands. With a specificity of 84%, the detection of unequivocal erosions achieved a sensitivity of 85%. A receiver operating characteristic analysis yields an area under the curve of 0.92. The visualization provided a clear representation of the erosions as determined by two musculoskeletal radiologists. CONCLUSION The automatic spotting of erosions provides promising results, and the visualization of the deviation from healthy anatomy aids clinicians in the evaluation of the erosions and in the reviewing of automatic detection results.

[1]  C. Taylor,et al.  Active shape models - 'Smart Snakes'. , 1992 .

[2]  Timothy F. Cootes,et al.  A minimum description length approach to statistical shape modeling , 2002, IEEE Transactions on Medical Imaging.

[3]  A. Graser,et al.  Effect of computer-aided detection as a second reader in multidetector-row CT colonography , 2007, European Radiology.

[4]  K. Kaufman,et al.  Semi-automated digital image analysis of joint space width in knee radiographs , 2005, Skeletal Radiology.

[5]  Harry K Genant,et al.  Effects of anakinra monotherapy on joint damage in patients with rheumatoid arthritis. Extension of a 24-week randomized, placebo-controlled trial. , 2004, The Journal of rheumatology.

[6]  H K Genant,et al.  Assessment of rheumatoid arthritis using a modified scoring method on digitized and original radiographs. , 1998, Arthritis and rheumatism.

[7]  P. Peloschek,et al.  Quantification of axial alignment of the lower extremity on conventional and digital total leg radiographs , 2004, European Radiology.

[8]  A. Larsen,et al.  Radiographic Evaluation of Rheumatoid Arthritis and Related Conditions by Standard Reference Films , 1977, Acta radiologica: diagnosis.

[9]  H K Genant,et al.  Neural network based algorithm to quantify joint space width in joints of the hand for arthritis assessment. , 2000, Medical physics.

[10]  S. van der Linden,et al.  Reading radiographs in chronological order, in pairs or as single films has important implications for the discriminative power of rheumatoid arthritis clinical trials. , 1999, Rheumatology.

[11]  E. Keystone,et al.  Radiographic, clinical, and functional outcomes of treatment with adalimumab (a human anti-tumor necrosis factor monoclonal antibody) in patients with active rheumatoid arthritis receiving concomitant methotrexate therapy: a randomized, placebo-controlled, 52-week trial. , 2004, Arthritis and rheumatism.

[12]  Anthony J. Sherbondy,et al.  Pulmonary nodules on multi-detector row CT scans: performance comparison of radiologists and computer-aided detection. , 2005, Radiology.

[13]  J. Moreland,et al.  Methods of scoring the progression of radiologic changes in rheumatoid arthritis. Correlation of radiologic, clinical and laboratory abnormalities. , 1971, Arthritis and rheumatism.

[14]  Milan Sonka,et al.  Object localization and border detection criteria design in edge-based image segmentation: automated learning from examples , 2000, IEEE Transactions on Medical Imaging.

[15]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[16]  Wolfgang Dihlmann,et al.  Präerosion, Erosion und ihre Kollaborateure: Die arthritischen Direktzeichen und ihre Differentialdiagnose , 1995 .

[17]  Maarten Boers,et al.  Variability of precision in scoring radiographic abnormalities in rheumatoid arthritis by experienced readers. , 2004, The Journal of rheumatology.

[18]  H. Bischof,et al.  Differential diagnosis of rheumatic diseases using conventional radiography. , 2004, Best practice & research. Clinical rheumatology.

[19]  J C Gardner,et al.  Computer-based methods for measuring joint space and estimating erosion volume in the finger and wrist joints of patients with rheumatoid arthritis. , 2000, Arthritis and rheumatism.

[20]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[21]  Horst Bischof,et al.  ASM Driven Snakes in Rheumatoid Arthritis Assessment , 2003, SCIA.

[22]  Franz Kainberger,et al.  An automatic model-based system for joint space measurements on hand radiographs: initial experience. , 2007, Radiology.

[23]  Timothy F. Cootes,et al.  Use of active shape models for locating structures in medical images , 1994, Image Vis. Comput..

[24]  Timothy F. Cootes,et al.  Training Models of Shape from Sets of Examples , 1992, BMVC.

[25]  Jane Angwin,et al.  Radiographic hand joint space width assessed by computer is a sensitive measure of change in early rheumatoid arthritis. , 2004, The Journal of rheumatology.