Selective image similarity measure for bronchoscope tracking based on image registration

Deguchi, Daisuke; Mori, Kensaku; Feuerstein, Marco; Kitasaka, Takayuki; Maurer, Calvin R.; Suenaga, Yasuhito; Takabatake, Hirotsugu; Mori, Masaki; Natori, Hiroshi

doi:10.1016/j.media.2009.06.001

Cited by 71 publications

(59 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method, which can greatly improve the performance of bronchoscope tracking, consists of four steps: (a) division of the input RB image into subblocks, (b) feature value computation in each subblock, (c) selection of the most appropriate subblocks, and (d) image similarity computation in the selected subblocks. For more details refer to Deguchi et al (2009).…”

Section: Intensity-based Image Registrationmentioning

confidence: 99%

Development and comparison of new hybrid motion tracking for bronchoscopic navigation

Luo

Feuerstein

Deguchi

et al. 2012

Medical Image Analysis

Self Cite

View full text Add to dashboard Cite

This paper presents a new hybrid camera motion tracking method for bronchoscopic navigation combining SIFT, epipolar geometry analysis, Kalman filtering, and image registration. In a thorough evaluation, we compare it to state-of-the-art tracking methods. Our hybrid algorithm for predicting bronchoscope motion uses SIFT features and epipolar constraints to obtain an estimate for interframe pose displacements and Kalman filtering to find an estimate for the magnitude of the motion. We then execute bronchoscope tracking by performing image registration initialized by these estimates. This procedure registers the actual bronchoscopic video and the virtual camera images generated from 3D chest CT data taken prior to bronchoscopic examination for continuous bronchoscopic navigation. A comparative assessment of our new method and the state-of-the-art methods is performed on actual patient data and phantom data. Experimental results from both datasets demonstrate a significant performance boost of navigation using our new method. Our hybrid method is a promising means for bronchoscope tracking, and outperforms other methods based solely on Kalman filtering or image features and image registration.

show abstract

Section: Intensity-based Image Registrationmentioning

confidence: 99%

Development and comparison of new hybrid motion tracking for bronchoscopic navigation

Luo

Feuerstein

Deguchi

et al. 2012

Medical Image Analysis

Self Cite

View full text Add to dashboard Cite

show abstract

“…The choice of an image similarity measure depends on the modality of the images to be registered. Common examples of image similarity measures include cross-correlation [3], mutual information and sum of squared intensity differences. The second step is to solve the registration model.…”

Section: Introductionmentioning

confidence: 99%

A High Performance Point Based Image Alignment Approach Using an Artificial Immune System

Omer¹,

Abdel-Nasser²

2013

IJFCC

View full text Add to dashboard Cite

Abstract-In this paper, a high performance image alignment approach is presented. This approach is classified as a point based alignment approach. An artificial immune system (AIS) with a modified mutation formula is used to find the correspondence points between the reference and the input images. After the correspondence is found, the least mean squares technique (LMS) is used to determine the transformation which is used to align the two images. This approach doesn't require any additional refinement or features detector as some others approaches required. To demonstrate the effectiveness of proposed algorithm, it compared with two state-of-the-art algorithms for different data sets.

show abstract

“…However, a major drawback is that image registration techniques heavily depend on characteristic information of bronchial trees (e.g., bifurcations or folds), so they can fail easily to track the bronchoscope camera in the case of the shortage of such information [5]. Featurebased bronchoscope motion estimation is a promising means for dealing with this problem during bronchoscope tracking [10,4].…”

Section: Introductionmentioning

confidence: 99%

“…Intensity-based registration commonly defines a similarity measure and maximizes the similarities or minimizes the dissimilarities between an RB image I (i) R and a virtual bronchoscopic (VB) image I V . We here use a modified mean squared error (MoMSE ) [5] similarity measure. Let I V (Q (i) ) be a VB image generated from the predicted pose Q (i) = Q (i−1) ∆Q (i) of the current frame using volume rendering techniques, where Q (i−1) denotes the previous camera pose and ∆Q (i) the inter-frame motion information between successive frames.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Application Driven Comparison of Several Feature Extraction Algorithms in Bronchoscope Tracking During Navigated Bronchoscopy

Luo

Feuerstein

Reichl

et al. 2010

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. This paper compares Kanade-Lucas-Tomasi (KLT), speeded up robust feature (SURF), and scale invariant feature transformation (SIFT) features applied to bronchoscope tracking. In our study, we first use KLT, SURF, or SIFT features and epipolar constraints to obtaininterframe translation (up to scale) and orientation displacements and Kalman filtering to recover an estimate for the magnitude of the motion (scale factor determination), and then multiply inter-frame motion parameters onto the previous pose of the bronchoscope camera to achieve the predicted pose, which is used to initialize intensity-based image registration to refine the current pose of the bronchoscope camera. We evaluate the KLT-, SURF-, and SIFT-based bronchoscope camera motion tracking methods on patient datasets. According to experimental results, we may conclude that SIFT features are more robust than KLT and SURF features at predicting the bronchoscope motion, and all methods for predicting the bronchoscope camera motion show a significant performance boost compared to sole intensity-based image registration without an additional position sensor.

show abstract

Selective image similarity measure for bronchoscope tracking based on image registration

Cited by 71 publications

References 21 publications

Development and comparison of new hybrid motion tracking for bronchoscopic navigation

Development and comparison of new hybrid motion tracking for bronchoscopic navigation

A High Performance Point Based Image Alignment Approach Using an Artificial Immune System

An Application Driven Comparison of Several Feature Extraction Algorithms in Bronchoscope Tracking During Navigated Bronchoscopy

Contact Info

Product

Resources

About