<![CDATA[Osteoporosis and osteopenia are conditions that commonly affect bone health significantly, this is characterized by decreased bone density causing the risk of fractures especially in the femur and tibia. The prevalence rate of these diseases is calculated from 103,334,579 people between the ages of 15 and 105 years, with an overall prevalence of 18.3%. Fast and accurate detection is needed for the first line of defense for osteoporosis patients and potential patients. This study provides the development of a Convolutional Neural network (CNN) model trained to predict osteoporosis and osteopenia from x-ray radiographs of femur and tibia bones. The proposed model has satisfactory performance on all metrics namely average accuracy 90%, average recall 90%, average F1 score 90%. From these performance results, alternative detection methods using CNN can be considered by medical parties or parties who can utilize the first diagnosis of osteopenia to osteoporosis bone disease handling compared to conventional methods.]]>
Copyrights © 2025
