In this paper, we propose a feature extraction and machine learning method for the classification of ultrasound images obtained from lumbar spine of pregnant patients in the transverse plane. A set of features, including matching values and positions, appearance of black pixels within predefined windows along the midline, are extracted from the ultrasound images using template matching and midline detection. Artificial neural network is utilized to classify the bone images and inter-spinous images. The neural network is trained with 1000 images from 25 pregnant subjects and tested on 720 images from a separate set of 18 pregnant patients. A high success rate (96.95% on training set, 95.75% on validation set and 94.12% on test set) is achieved with the proposed method. The trained neural network further tested on 43 videos collected from 43 pregnant subjects and successfully identified the proper needle insertion site (interspinous region) in all of the cases. Therefore, the proposed method is able to identify the ultrasound images of lumbar spine in an automatic manner, so as to facilitate the anesthetists' work to identify the needle insertion point precisely and effectively.
展开▼
