Browsing by Author "Kaya, Mert"

Now showing 1 - 6 of 6

Metadata only
Gabor filter based localization of needles in ultrasound guided robotic interventions
(IEEE, 2014) Kaya, Mert; Bebek, Özkan; Mechanical Engineering; BEBEK, Özkan; Kaya, Mert
This paper presents an entropy based parameter tuning method for needle segmentation, and a probability map based needle tip estimation method using Gabor-based line filter. The proposed automatic parameter tuning method optimizes the threshold value that is used by the Otsu's thresholding technique to binarize the ultrasound image. A probability map is created to estimate the needle tip location using the Gabor filtered image and the binarized image. The pixel with the maximum probability represents the needle tip location. Finally, an enhancement method to improve needle visibility is proposed. The proposed methods are experimentally tested in four different phantoms and distilled water. The image processing time is reduced by 24% using the proposed tuning method, and the needle tip location can be successfully estimated using the probability map.
Embargo
Localization of biopsy needles in 2D ultrasound images using gabor filter
(2015-06) Kaya, Mert; Bebek, Özkan; Bebek, Özkan; Uğurdağ, Hasan Fatih; Ünay, D.; Department of Electrical and Electronics Engineering; Kaya, Mert
Needle Biopsies are common medical procedures to remove tissue samples from the body to be examined for diseases. During these procedures, the needle and its tip are tracked using medical imaging devices to avoid damage to the tissues and to get samples from the targeted site. Ultrasound (US) is a common imaging method to localize the needles because it does not have any known side effects and provides a larger workspace. However, excessive artifacts and low resolution of the US images make it difficult to localize the needles in the biopsies. A possible solution to this problem is to detect the needle and its tip using an image processing algorithm. In this thesis, a novel needle localization method is proposed for 2D US images based on the Gabor filter. This method enhances the needle outline to localize the needle, while suppressing other structures in the US images. The needle tip is localized in real-time and the estimation noise is reduced using the Kalman filter. Using the bin packing method, the processing time is reduced by 56%, without a GPU. Also, a simulation study to understand the accuracy of the system within different mediums is conducted. The proposed needle localization scheme was tested using various phantoms. An external sensor was used to evaluate the accuracy of the needle insertion angle estimation and the needle tip localization methods. The Root mean square (RMS) error of the needle insertion angle was 2.29° ; and the RMS error of the needle tip position was 1.17 mm. The results showed that this localization scheme can be used in the US image guided percutaneous needle procedures.
Metadata only
Needle localization using gabor filtering in 2D ultrasound images
(IEEE, 2014) Kaya, Mert; Bebek, Özkan; Mechanical Engineering; BEBEK, Özkan; Kaya, Mert
In the percutaneous needle procedures using ultrasound (US) imaging, the needle should be detected precisely to avoid damage to the tissue and to get the samples from the appropriate site. Excessive artifacts and low resolution of the US images make it difficult to detect the needle and its tip. It is possible to enhance the needle image using image processing; and this work proposes a novel needle detection method in 2D US images based on the Gabor filter. This method enhances the needle outline while suppressing the other structures in the image. First, the needle insertion angle is estimated and then the needle trajectory is found with the RANSAC line estimator. The experiments with three different phantoms showed that the algorithm is robust and could work in percutaneous needle procedures using US images.
Metadata only
Real-time needle tip localization in 2D ultrasound images for robotic biopsies
(IEEE, 2015) Kaya, Mert; Şenel, Enes; Ahmad, Mirza Awais; Orhan, Orçun; Bebek, Özkan; Mechanical Engineering; BEBEK, Özkan; Kaya, Mert; Şenel, Enes; Ahmad, Mirza Awais; Orhan, Orçun
In this paper, real-time needle tip tracking method using 2D ultrasound (US) images for robotic biopsies is presented. In this method, the needle tip is estimated with the Gabor filter based image processing algorithm, and the estimation noise is reduced with the Kalman filter. This paper also presents the needle tip tracking simulation to test accuracy of the Kalman filter under position misalignments and tissue deformations. In order to execute proposed method in real-time, the bin packing method is used and the processing time is reduced by 56%, without a GPU. The proposed method was tested in four different phantoms and water medium. The accuracy of the needle tip estimation was measured with optical tracking system, and root mean square error (RMS) of the tip position is found to be 1.17 mm. The experiments showed that the algorithm could track the needle tip in real-time.
Metadata only
Visual tracking of biopsy needles in 2D ultrasound images
(2016) Kaya, Mert; Şenel, Enes; Ahmad, Mirza Awais; Bebek, Özkan; Mechanical Engineering; BEBEK, Özkan; Kaya, Mert; Şenel, Enes; Ahmad, Mirza Awais
Ultrasound (US) is one the most commonly used medical imaging techniques in percutaneous needle procedures. However, US images are inherently noisy and contain excessive number of artifacts. Hence, it is not easy to track the needle tip in the US images during the needle insertions. At this point, image based visual tracking techniques can be used for needle tip tracking. This paper presents a method for visual tracking of biopsy needles in 2D US images using sum of squared differences and sum of conditional variances. Second order Gauss-Newton optimization is used to decrease processing time and make the tracking more robust. The needle template images used in the method are updated with a strategy to prevent needle loss and detection failures during tracking. The paper also explains how to identify needle losses during tracking and how to recover the needle position without using a needle localization algorithm. We demonstrate the precision of the visual needle tip tracking method with experiments under challenging tracking conditions.
Metadata only
Visual tracking of multiple moving targets in 2D ultrasound guided robotic percutaneous interventions
(IEEE, 2017) Kaya, Mert; Şenel, Enes; Ahmad, Mirza Awais; Bebek, Özkan; Mechanical Engineering; BEBEK, Özkan; Kaya, Mert; Şenel, Enes; Ahmad, Mirza Awais
Percutaneous needle procedures are mostly carried out with the guidance of 2D ultrasound (US) imaging. US images are inherently noisy and their resolutions are low. Hence, target tracking can be challenging. Image based tracking methods can be used to track the needle and the target. This paper proposes visual tracking of multiple moving points, such as biopsy needles and targets, in 2D US images using normalized cross correlation and mutual information similarity functions. Both moving and deformable targets can be tracked. An affine motion model is used for small and moving target tracking and a thin plate spline motion model is used for deformable target tracking. During the tracking, needle and target template images are updated with a template update strategy. Also, tracking outputs of normalized cross correlation and mutual information are fused using the Kalman filter to reduce the tracking error. During the experiments, needle is inserted using a needle insertion robot. 2D US probe is attached to a robotic arm's end effector to servo the probe along the needle insertion path. Proposed needle and target tracking methods were tested with phantoms. Accuracies of the needle tip and moving target tracking methods were measured using an optical tracking system. Experimental results showed that the proposed tracking method could be used to simultaneously track the needle tip and the targets in real-time in 2D US guided percutaneous needle procedures.