Ultrasound guided automatic localization of needle insertion site for epidural anesthesia

Abstract

In this paper, ultrasound imaging is utilized to detect the anatomical structure of the lumbar spine based on which an image processing algorithm will search for key features to identify the optimal needle insertion site. The key challenge lies in the nature of ultrasound images which are obscure and have low spatial resolution, induced by contamination from random speckle noises. In order to improve the interpretability of ultrasound images, a modified version of local normalization using the Difference of Gaussian algorithm is first used for pre-processing to filter the speckle noise and extract the main anatomical structure in the raw images obtained. Meanwhile, local means induced by non-uniform wave reflection rate is also successfully removed by the proposed pre-processing algorithm, thus a potential element that may degrade the image recognition accuracy is excluded. In the second stage, a template matching algorithm, augmented with a position correlation function, automatically identifies the key features of interest and thus the insertion site. The approach has been tested on more than 200 ultrasound images with a 100% success rate. The proposed system allows the anesthetist to use the approach efficiently without the burden of interpreting real time ultrasound images.

Segmentation of the spinous process and its acoustic shadow in vertebral ultrasound images

“…Recently, a method for automatic detection of the bone and inter-spinous regions in vertebral ultrasound images acquired in the transverse plane was developed by Yu et al (2013). The method first involves pre-processing the image using a difference of Gaussians filter and local intensity normalization.…”

Obstetric Anesthesiologists as Perioperative Physicians: Improving Peripartum Care and Patient Safety

“…The ultrasound images are video streamed, and algorithms are then used to process the images and to automatically recognize structures (articular process, vertebral body, and epidural space). Thus, the optimal needle insertion site can be identified [57,58]. This program may be useful in correctly identifying interspaces from ultrasound images in which anatomy is difficult to recognize visually.…”

Lumbar Ultrasound Image Feature Extraction and Classification with Support Vector Machine

“…In our previous research, we proposed decomposing the flying bat shape into three subfeatures: the bat ear (articular process), epidural space and vertebra body. The decomposed subfeatures matched the articular process and vertebra body with high accuracy on images obtained from volunteers (Yu et al 2013). …”

“…Before feature extraction, raw ultrasound images were pre-processed with the difference of Gaussian enhanced local normalization, so as to remove speckle noise and extract anatomic structures, with kernel parameters set as 20 3 20, s 5 4, for first kernel, and 20 3 20, s 5 10, for second kernel (Yu et al 2013(Yu et al , 2014a. After pre-processing, local intensity variance induced by ultrasound wave attenuation was also eliminated.…”