Osteoporosis Detection using Important Shape-Based Features of the Porous Trabecular Bone on the Dental X-Ray Images

Osteoporosis screening using dental X-Ray images has been growing an interesting research. Existing methods for osteoporosis screening have been performed using the dental periapical or panoramic in X-Ray images. There was limited research using both the periapical and panoramic due to the expensive cost of obtaining data. This paper presents a combination of the periapical and panoramic images for osteoporosis detection. The image processing was performed to obtain the shape-based features of the porous trabecular bone on both the dental radiograph images. The important features were selected from the extracted features. These selected features were chosen for osteoporosis detection using the decision tree. The quantitative evaluation used confusion matrix. It was found accuracy rate to be 73.33%, sensitivity rate to be 72.23, and specificity rate to be 72.23% for data testing. Keywords—dental X-Ray; feature selection; osteoporosis Detection; porous trabecular bone