RESUMO
Noninvasive 3D reconstruction of a bone requires very accurate 2D navigated scans of bone. The use of brightness-mode ultrasound seems to be promising, if some 2D scans of bone are obtained in a fully automatic manner. This paper presents a rapid and fully automatic method for segmenting bone in a standard 2D ultrasound image (B-mode image). The algorithm focuses on segmenting bone in the B-mode image using RF data of the image. The article introduces the signal-processing scheme designed based on RF data to automatically segment bone in the B-mode image. The segmentation accuracy was assessed by performing various tests for this algorithm for various locations of the limbs of the human body. The algorithm was tested for 120 images taken at different locations of limbs of the human body. The sensitivity of these tests was calculated to be 0.99 and specificity was found to be 1. The suggested segmentation approach provides a reliable means of detecting bone in B-mode image.
