Fast and robust 3D ultrasound registration – Block and game theoretic matching

Banerjee, Jyotirmoy; Klink, Camiel; Peters, Edward D.; Niessen, Wiro J.; Moelker, Adriaan; Walsum, Theo van

doi:10.1016/j.media.2014.11.004

Cited by 33 publications

(34 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yet quantitative evaluation of tracking the human liver under free breathing was reported only by Banerjee et al (2015), Harris et al (2010), Lediju Bell et al (2012) and Vijayan et al (2014) for 3D sequences and by Cifor et al (2012), (2013), De Luca et al (2012) and De Luca et al (2013) for 2D sequences. Intensity-based and hybrid approaches achieved good accuracy (∼1.4 mm mean tracking error (Harris et al 2010) and ∼90% mean overlap ratio (Cifor et al 2013)).…”

Section: Introductionmentioning

confidence: 99%

“…The non-rigid registration method of Vijayan et al (2014) estimated liver motion with an error of 1 mm (75% percentile of a root-mean squared metric over all datasets), which was lower than the inter-observer variability of 1.4 mm. More recently, a fast 3D affine block-matching algorithm with an outlier rejection strategy achieved a mean tracking error of 1.8 mm (Banerjee et al 2015). However, these methods were only tested off-line on short sequences (<1 min).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The 2014 liver ultrasound tracking benchmark

Luca¹,

Benz²,

Kondo³

et al. 2015

Phys. Med. Biol.

View full text Add to dashboard Cite

The Challenge on Liver Ultrasound Tracking (CLUST) was held in conjunction with the MICCAI 2014 conference to enable direct comparison of tracking methods for this application. This paper reports the outcome of this challenge, including setup, methods, results and experiences. The database included 54 2D and 3D sequences of the liver of healthy volunteers and tumor patients under free breathing. Participants had to provide the tracking results of 90% of the data (test set) for pre-defined point-landmarks (healthy volunteers) or for tumor segmentations (patient data). In this paper we compare the best six methods which participated in the challenge. Quantitative evaluation was performed by the organizers with respect to manual annotations. Results of all methods showed a mean tracking error ranging between 1.4 mm and 2.1 mm for 2D points, and between 2.6 mm and 4.6 mm for 3D points. Fusing all automatic results by considering the median tracking results, improved the mean error to 1.2 mm (2D) and 2.5 mm (3D). For all methods, the performance is still not comparable to human inter-rater variability, with a mean tracking error of 0.5–0.6 mm (2D) and 1.2–1.8 mm (3D). The segmentation task was fulfilled only by one participant, resulting in a Dice coefficient ranging from 76.7% to 92.3%. The CLUST database continues to be available and the online leader-board will be updated as an ongoing challenge.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The 2014 liver ultrasound tracking benchmark

Luca¹,

Benz²,

Kondo³

et al. 2015

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

“…In the global 4D tracking step, the whole liver volume is tracked by combining registrations to the previous and reference frame using the register‐to‐reference‐by‐tracking strategy . This is followed by the local 3D registration step, where the tracking result from the previous step is refined by performing registration on the neighborhood region close to the anatomical landmark j , using the register‐to‐reference strategy . Both steps use block‐matching, with normalized cross correlation as similarity metric, followed by an outlier rejection scheme.…”

Section: Methodsmentioning

confidence: 99%

“…Both steps use block‐matching, with normalized cross correlation as similarity metric, followed by an outlier rejection scheme. Finally the rigid transformation is estimated from the trusted block‐matching results …”

Section: Methodsmentioning

confidence: 99%

Evaluation of 2D and 3D ultrasound tracking algorithms and impact on ultrasound‐guided liver radiotherapy margins

et al. 2018

Self Cite

View full text Add to dashboard Cite

Liver motion estimation and prediction during free-breathing from 2D ultrasound images can substantially reduce the in-plane motion uncertainty and hence treatment margins. Employing an accurate tracking method while avoiding non-linear temporal prediction would be favorable. This approach has the potential to shorten treatment time compared to breath-hold and gated approaches, and increase treatment efficiency and safety.

show abstract

“…Often there is a trade‐off between accuracy and speed, especially if more advanced techniques require additional computational complexity. There have been several investigations into ultrasound tracking methods including those based upon the principles of slow feature analysis, normalized gradient fields, logDemons, Bayesian methods, and block matching based methods . A wide range of tracking accuracies have been reported because it is often difficult to compare methods due to the variation in the data sets used for analysis.…”

Section: Introductionmentioning

confidence: 99%

A block matching based approach with multiple simultaneous templates for the real‐time 2D ultrasound tracking of liver vessels

et al. 2017

View full text Add to dashboard Cite

Purpose The implementation of motion management techniques in radiation therapy can aid in mitigating uncertainties and reducing margins. For motion management to be effective, it is necessary to track key structures both accurately and at a real-time speed. Therefore, the focus of this work was to develop a 2D algorithm for the real-time tracking of ultrasound features to aid in radiation therapy motion management. Materials and Methods The developed algorithm utilized a similarity measure-based block matching algorithm incorporating training methods and multiple simultaneous templates. The algorithm is broken down into three primary components, all of which use normalized cross-correlation (NCC) as a similarity metric. First, a global feature shift to account for gross displacements from the previous frame is determined using large block sizes which encompass the entirety of the feature. Second, the most similar reference frame is chosen from a series of training images that are accumulated during the first K frames of tracking to aid in contour consistency and provide a starting point for the localized template initialization. Finally, localized block matching is performed through the simultaneous use of both a training frame and the previous frame. The localized block matching utilizes a series of templates positioned at the boundary points of the training and previous contours. The weighted final boundary points from both the previous and the training frame are ultimately combined and used to determine an affine transformation from the previous frame to the current frame. Results A mean tracking error of 0.72 ± 1.25 mm was observed for 85 point-landmarks across 39 ultrasound sequences relative to manual ground truth annotations. The image processing speed per landmark with the GPU implementation was between 41 and 165 frames per second (fps) during the training set accumulation, and between 73 and 234 fps after training set accumulation. Relative to a comparable multithreaded CPU approach using OpenMP, the GPU implementation resulted in speedups between −30% and 355% during training set accumulation, and between −37% and 639% post-accumulation. Conclusions Initial implementations indicated an accuracy that was comparable to or exceeding those achieved by alternative 2D tracking methods, with a computational speed that is more than sufficient for real-time applications in a radiation therapy environment. While the overall performance reached levels suitable for implementation in radiation therapy, the observed increase in failures for smaller features, as well as the algorithm’s inability to be applied to non-convex features warrants additional investigation to address the shortcomings observed.

show abstract

Fast and robust 3D ultrasound registration – Block and game theoretic matching

Cited by 33 publications

References 37 publications

The 2014 liver ultrasound tracking benchmark

The 2014 liver ultrasound tracking benchmark

Evaluation of 2D and 3D ultrasound tracking algorithms and impact on ultrasound‐guided liver radiotherapy margins

A block matching based approach with multiple simultaneous templates for the real‐time 2D ultrasound tracking of liver vessels

Contact Info

Product

Resources

About