Previous studies have shown an association between osteoporosis and automatic measurements of mandibular cortical width on dental panoramic radiographs (DPRs). In this study, we show that additional image texture features increase this association and propose the combined features as a potential biomarker for osteoporosis. We used an existing dataset of 663 DPRs of female patients with bone mineral density (BMD) measurements. The mandibular cortex was located using a previously described computer algorithm. Texture features, based on co-occurrence matrices and fractal dimension, were measured in the bone within the cortex and also in the superior basal bone above the cortex. These, augmented by cortical width measurements, were used by a random forest classifier to identify osteoporosis at femoral neck, total hip, and lumbar spine. Classification performance was assessed by ROC analysis. Area-under-curve (AUC) values for identifying osteoporosis at femoral neck were 0.830, 0.824, and 0.872 using, respectively, cortical width alone, cortical texture (co-occurrence matrix features) alone, and combined width and texture. At 80% sensitivity, these classifiers produced specificity values of 74.4%, 73.6%, and 80.0%, respectively. Fractal dimension was a less effective texture feature. Prediction of osteoporosis at the lumbar spine was poorer, but a combined width and superior basal bone texture classifier gave a significant improvement in AUC at over the use of width alone.
The pattern of decrease in mandibular cortical width with age was similar to the known pattern of bone loss from the hip, accelerating in women after the age of 42.5 years.
Abstract. We provide a fully automatic method of segmenting vertebrae in DXA images. This is of clinical relevance to the diagnosis of osteoporosis by vertebral fracture, and to grading fractures in clinical trials. In order to locate the vertebrae we train detectors for the upper and lower vertebral endplates. Each detector uses random forest regressor voting applied to Haar-like input features. The regressors are applied at a grid of points across the image, and each tree votes for an endplate centre position. Modes in the smoothed vote image are endplate candidates, some of which are the neighbouring vertebrae of the one sought. The ambiguity is resolved by applying geometric constraints to the connections between vertebrae, although there can be some ambiguity about where the sequence starts (e.g. is the lowest vertebra L4 or L5, Fig 2a). The endplate centres are used to initialise a final phase of Active Appearance Model search for a detailed solution. The method is applied to a dataset of 320 DXA images. Accuracy is comparable to manually initialised AAM segmentation in 91% of images, but multiple grade 3 fractures can cause some edge confusion in severely osteoporotic cases.
Specificity and sensitivity are improved by using an appearance-based classifier compared to standard height ratio morphometry. An overall sensitivity loss of 7% occurs (at 95% specificity) when using a semi-automatic (AAM) segmentation compared to expert annotation, due to segmentation error. However, the classifier sensitivity is still adequate for a computer-assisted diagnosis system for vertebral fracture, especially if used in a triage approach.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.