A comprehensive study on feature types for osteoporosis classification in dental panoramic radiographs

• Published in: Computer methods and programs in biomedicine 2020-05, Vol.188, p.105301-105301, Article 105301
• Main Authors: Alzubaidi, Mohammad A., Otoom, Mwaffaq
• Format: Article
• Language: English
• Subjects: Absorptiometry, Photon; Algorithms; Bone Density; Computer vision; Dental radiograph; Diagnosis, Computer-Assisted; False Positive Reactions; Humans; Image processing; Image Processing, Computer-Assisted; Learning vector quantization; Machine Learning; Mandibular Fractures - diagnostic imaging; Osteoporosis - classification; Osteoporosis - diagnostic imaging; Osteoporosis detection and classification; Osteoporotic Fractures - diagnostic imaging; Pattern Recognition, Automated; Radiographic Image Interpretation, Computer-Assisted - methods; Radiography, Panoramic; Self organizing map; Sensitivity and Specificity; Support Vector Machine; Wavelet Analysis
• ISSN: 0169-2607; 1872-7565
• DOI: 10.1016/j.cmpb.2019.105301

•Osteoporosis early detection and classification.•Feature types in dental radiograph.•Self organizing map with learning vector quantization. Osteoporosis is a disease characterized by a decrease in bone density. It is often associated with fractures and severe pain. Previous studies have shown a high correlation between the density of the bone in the hip and in the mandibular bone in the jaw. This suggests that dental radiographs might be useful for detecting osteoporosis. Use of dental radiographs for this purpose would simplify early detection of osteoporosis. However, dental radiographs are not normally examined by radiologists. This paper explores the use of 13 different feature extractors for detection of reduced bone density in dental radiographs. The computed feature vectors are then processed with a Self-Organizing Map and Learning Vector Quantization as well as Support Vector Machines to produce a set of 26 predictive models. The results show that the models based on Self-Organizing Map and Learning Vector Quantization using Gabor Filter, Edge Orientation Histogram, Haar Wavelet, and Steerable Filter feature extractors outperform the rest of the 22 models in detecting osteoporosis. The proposed Gabor-based algorithm achieved an accuracy of 92.6%, a sensitivity of 97.1%, and a specificity of 86.4%. The oriented edges and textures in the upper and lower jaw regions are useful for distinguishing normal patients from patients with osteoporosis.