Clinical and computerized assessment of mandibular asymmetry.

The aim of this study was to investigate assessment of mandibular asymmetry by clinicians and to evaluate a new computerized system. Eight experienced clinicians (four maxillofacial surgeons and four orthodontists) assessed 12 standardized facial photographs of patients selected to provide a range, from normality to marked mandibular asymmetry. Photographs were taken under standardized conditions; analysis of photographs repeated after one month showed close agreement. The computerized system used four different methods for quantifying asymmetry based on right/left differences in perimeter, area, compactness, and centre of area ratios. Good agreement (kappa = 0.77) was achieved by the clinicians in assessing the likelihood of treatment need; however, when assessing an 'acceptable' level of asymmetry, agreement was moderate (kappa = 0.46). Differences in perimeter ratios did not compare well with clinical assessment, but those for both area and compactness showed 100 per cent sensitivity and specificity to clinical assessment at ratio differences of 0.05 and 0.03 (deviation from 1), respectively. A centre of area difference ratio greater than 1 showed 75 per cent sensitivity and 85 per cent specificity to clinical assessment. These parameters could prove useful in quantifying change in asymmetries brought about by growth, treatment, or any subsequent relapse.