OBJECTIVES: This in vitro study examined radiographic fractal dimension changes in alveolar process bone during simulated osteoporosis. STUDY DESIGN: Ten specimens of human maxillary alveolar process bone were progressively decalcified, and the percentage of calcium lost at each decalcification stage was quantified. Four radiographs of each specimen, together with an aluminum step-wedge, were exposed at 70 kVp at each stage. The test set of 560 radiographs was digitized, identical bony regions of interest were selected from the density-corrected images of each specimen, the regions were digitally filtered to reduce film-grain noise, and fractal dimension was computed on a line-to-line basis. Correlation analysis quantified the relationship between calcium loss and fractal dimension change. Analysis of variance and Duncan's multiple range test determined whether a difference existed in fractal dimension computed from images at x-ray beam angulations of -5, 0, and +5 degrees. RESULTS: A strong correlation (average r = -0.94, p < or = 0.0037) was found between generalized demineralization and decreasing fractal dimension. In every bone sample fractal dimension changed significantly (p < or = 0.0189) with angular change. CONCLUSIONS: Radiographic fractal dimension holds promise for detecting simulated osteoporosis in the maxilla under ideal conditions, but the sensitivity of fractal dimension to small x-ray beam angular change renders its clinical application questionable.
OBJECTIVES: This in vitro study examined radiographic fractal dimension changes in alveolar process bone during simulated osteoporosis. STUDY DESIGN: Ten specimens of human maxillary alveolar process bone were progressively decalcified, and the percentage of calcium lost at each decalcification stage was quantified. Four radiographs of each specimen, together with an aluminum step-wedge, were exposed at 70 kVp at each stage. The test set of 560 radiographs was digitized, identical bony regions of interest were selected from the density-corrected images of each specimen, the regions were digitally filtered to reduce film-grain noise, and fractal dimension was computed on a line-to-line basis. Correlation analysis quantified the relationship between calcium loss and fractal dimension change. Analysis of variance and Duncan's multiple range test determined whether a difference existed in fractal dimension computed from images at x-ray beam angulations of -5, 0, and +5 degrees. RESULTS: A strong correlation (average r = -0.94, p < or = 0.0037) was found between generalized demineralization and decreasing fractal dimension. In every bone sample fractal dimension changed significantly (p < or = 0.0189) with angular change. CONCLUSIONS: Radiographic fractal dimension holds promise for detecting simulated osteoporosis in the maxilla under ideal conditions, but the sensitivity of fractal dimension to small x-ray beam angular change renders its clinical application questionable.