Improving the detection of osteoporosis from dental radiographs using Active Appearance Models

We describe improvements to a method of detecting patients at risk of osteoporosis from automatic measurement of the inferior mandibular cortex on panoramic dental tomograms. Previous work had used an Active Shape Model (ASM) to locate the mandibular edges. However, the edge-based ASM has little lateral positioning information and in osteoporotic cases the superior border is often poorly defined. These problems can degrade the accuracy of the measurements of cortical width. We have obtained superior accuracy using an Active Appearance Model incorporating a complex texture model derived from an osteoporotic-enriched training set. This leads to improvements in diagnostic accuracy, when applied to a large dataset of 663 subjects with known Bone Mineral Density.