Mandibular Trabecular Bone Analysis Using Local Binary Pattern for Osteoporosis Diagnosis.

Area of Science:

• Radiology
• Biomedical Imaging
• Osteoporosis Research

Background:

• Osteoporosis is a skeletal disease marked by low bone density and deterioration, increasing fracture risk.
• Panoramic radiography, a common dental imaging technique, offers an opportunistic method for osteoporosis screening.
• Jaw trabecular pattern analysis from radiographs is crucial for assessing bone quality and strength.

Purpose of the Study:

• To investigate the correlation between jaw trabecular patterns in panoramic images and osteoporosis.
• To utilize texture analysis with local binary patterns for osteoporosis assessment from dental radiographs.

Main Methods:

• A dataset of panoramic radiographs from osteoporosis and non-osteoporosis patients was analyzed.
• A modified local binary pattern approach was employed to extract features from jaw radiographs.
• Various classifiers were used to diagnose osteoporosis based on extracted features.

Main Results:

• The proposed method achieved a classification accuracy of approximately 99.6%.
• This accuracy surpasses many existing state-of-the-art methods for osteoporosis detection.
• The approach demonstrated high efficacy in distinguishing between osteoporosis and non-osteoporosis cases.

Conclusions:

• The developed approach offers high classification accuracy for osteoporosis screening using panoramic images.
• Advantages include low computational complexity, multi-resolution capability, and rotation invariance.
• This method presents a promising, non-invasive tool for early osteoporosis detection.