Non-uniform Rotational Distortion Research Articles

Su2037 Detecting Bowel Wall Fibrosis With Photoacoustic Imaging in an Animal Model of Crohn's Disease

Background/Aims: Optical coherence tomography (OCT) enables three-dimensional (3D) examination of the tissue architecture with micrometer resolution. In previous studies, endoscopic OCT was used to provide structural information correlated with histology for gastrointestinal pathologies. However, the utility of endoscopic OCT for investigating the pathology of the terminal ileum (TI) in patients with symptoms of inflammatory bowel disease (IBD), for example, has been very limited. Endoscopically, reaching the TI entails advancing the colonoscope through multiple sharp bends of the colon, which prevents smooth rotation of previous generations of proximally actuated probes. Recently, our group has developed ultrahigh speed endoscopic OCT technology with a micromotor imaging probe, which enables precise volumetric imaging by avoiding the non-uniform rotational distortion inherent in proximally actuated probes. In this study, we reported structural and vascular changes of TI with chronic inflammation using endoscopic OCT and OCT angiography (OCTA). Methods: This prototype endoscopic OCT system achieves an imaging speed of 600,000 A-scans per second and a frame rate of 400 frames per second. The diameter of the imaging probe was small enough to pass through the accessory port of the colonoscope, enabling coregistered biopsies. OCTA visualized microvasculature by measuring decorrelation between sequential OCT images to generate motion contrast from blood flow. The study was performed at the Veterans Affairs Boston Healthcare System (VABHS) with the approval from the institutional review boards from VABHS, Harvard Medical School, and MIT. Results: Figure 1 shows OCT images of a normal TI (Figs. 1(A, B)) and TI with chronic inflammation (Figs. 1(C, D)) as well as the coregistered en face OCTA (Fig. 1(E)) at the same depth as Fig. 1(C). Typical villous structure of normal TI can be observed in Fig. 1(A) at 100 μm depth, which was consistent with the colonoscopy findings (inset, Fig. 1(A)). The slight tilted orientation of the villi was caused by the gentle pressing of the imaging probe, which was observed along the cross-sectional OCT image as well (Fig. 1(B)). On the contrary, a loss of villous structure and nodular mucosal appearance were identified in Fig. 1(C) of TI with chronic inflammation, which was confirmed by the coregistered biopsy. In addition, a decreased imaging depth was observed in the cross-sectional OCT image (Fig. 1(D)). Fig 1(E) shows the coregistered en faceOCTA as Fig. 1(C) where increased vasculature with abnormal vascular branching was observed. Conclusions: This study demonstrates that ultrahigh speed endoscopic OCT technology and OCTA allows the visualization structural and vascular features of the TI with chronic inflammation. Acknowledgment: NIH R01CA075289-17, R44-CA101067-07, AFOSR FA9550-10-1-0551 and FA9550-12-1-0499.

Correction of rotational distortion for catheter-based en face OCT and OCT angiography.

We demonstrate a computationally efficient method for correcting the nonuniform rotational distortion (NURD) in catheter-based imaging systems to improve endoscopic en face optical coherence tomography (OCT) and OCT angiography. The method performs nonrigid registration using fiducial markers on the catheter to correct rotational speed variations. Algorithm performance is investigated with an ultrahigh-speed endoscopic OCT system and micromotor catheter. Scan nonuniformity is quantitatively characterized, and artifacts from rotational speed variations are significantly reduced. Furthermore, we present endoscopic en face OCT and OCT angiography images of human gastrointestinal tract in vivo to demonstrate the image quality improvement using the correction algorithm.

In vivo feasibility of endovascular Doppler optical coherence tomography

Feasibility of detecting intravascular flow using a catheter based endovascular optical coherence tomography (OCT) system is demonstrated in a porcine carotid model in vivo. The effects of A-line density, radial distance, signal-to-noise ratio, non-uniform rotational distortion (NURD), phase stability of the swept wavelength laser and interferometer system on Doppler shift detection limit were investigated in stationary and flow phantoms. Techniques for NURD induced phase shift artifact removal were developed by tracking the catheter sheath. Detection of high flow velocity (~51 cm/s) present in the porcine carotid artery was obtained by phase unwrapping techniques and compared to numerical simulation, taking into consideration flow profile distortion by the eccentrically positioned imaging catheter. Using diluted blood in saline mixture as clearing agent, simultaneous Doppler OCT imaging of intravascular flow and structural OCT imaging of the carotid artery wall was feasible. To our knowledge, this is the first in vivo demonstration of Doppler imaging and absolute measurement of intravascular flow using a rotating fiber catheter in carotid artery.

SYSTEM AND METHOD FOR REDUCING ANGULAR GEOMETRIC DISTORTION IN AN IMAGING DEVICE

A system and method are provided for significantly reducing or substantially eliminating angular geometric distortions in devices designed for imaging and/or inspection of an interior portion or surface of a cavity. A series of processing steps or methods may be employed to eliminate Non-Uniform Rotational Distortion (NURD) in such devices, for example, unidirectional and bi-directional intravascular ultrasonic (IVUS) imaging systems. The system may include a processor and an electronic module which control operation of a transducer assembly provided at a distal end of a catheter assembly. The system invokes a first processing step or method to collect and store raw angle and line data, as well as one or more of second and third processing steps or methods which adjust for NURD experienced during backlash of a bi-directional imaging system and a fourth processing step or method which performs a line-to-line correlation function.

Azimuthal Registration of Image Sequences Affected by Nonuniform Rotation Distortion

Imaging modalities that use a mechanically rotated endoscopic probe to scan a tubular volume, such as an artery, often suffer from image degradation due to nonuniform rotation distortion (NURD). In this paper, we present a new method to align individual lines in a sequence of images. It is based on dynamic time warping, finding a continuous path through a cost matrix that measures the similarity between regions of two frames being aligned. The path represents the angular mismatch corresponding to the NURD. The prime advantage of this novel approach compared to earlier work is the line-to-line continuity, which accurately captures slow intraframe variations in rotational velocity of the probe. The algorithm is optimized using data from a clinically available intravascular optical coherence tomography (OCT) instrument in a realistic vessel phantom. Its efficacy is demonstrated on an in vivo recording, and compared with conventional global rotation block matching. Intravascular OCT is a particularly challenging modality for motion correction because, in clinical situations, the image is generally undersampled, and correlation between the speckle in different lines or frames is absent. The algorithm can be adapted to ingest data frame-by-frame, and can be implemented to work in real time.

Comparison of nonuniform rotational distortion between mechanical IVUS and OCT using a phantom model

Optical coherence tomography (OCT) is an optical analog of mechanical intravascular ultrasound (M-IVUS) with much higher spatial resolution. However, no data exist regarding the nonuniform rotational distortion (NURD) with OCT. The aim of the study was to investigate whether OCT generates less NURD relative to M-IVUS. A coronary artery phantom model was constructed with a rubber ring (3.68 mm in diameter), located at the distal end of the phantom. This model was also composed of eight equally spaced steel wires and an additional marker-wire. Two types of vascular phantoms were used, mild curve (90°) and acute curve (near 180°). Subsequent M-IVUS ( n = 6) and OCT ( n = 6) imaging was performed. Eight angles between eight wires, except the marker-wire, were measured from each image. These angles, measured with M-IVUS and OCT, were compared with those of high-resolution optical photography as a gold standard. The average in angle differences was significantly smaller in OCT compared with M-IVUS in the mild curve model (3.2 ± 1.0° vs, 6.9 ± 2.1°, p < 0.01). Compared with the latter model, the average in angle differences was exaggerated in the acute curve model with M-IVUS (9.1 ± 0.9° vs. 6.9 ± 2.1°, p < 0.05) but not with OCT (3.5 ± 0.8° vs. 3.2 ± 1.0°, p= not significant). OCT generates significantly less NURD compared with M-IVUS, especially in tortuous situation. (E-mail: ijang@partners.org)

Strain imaging with intravascular ultrasound: An in vivo study

The evaluation of mechanical properties of coronary plaques is of high interest for the assessment of coronary diseases. Intravascular ultrasound (IVUS) can be used to visualize strain in coronary tissue. In this study, strain imaging is performed using an IVUS system with a 40-MHz rotating single-element transducer. Radio frequency (rf) data are acquired during in vivo examinations and sampled at 100 MHz. Image frames are stored consecutively during 3 s at a frame rate of 30/s. Data are recorded at different levels of tissue compression. The required pressure difference is caused by natural pulsatile blood flow. The strain imaging algorithm estimates radial strain from rf data based on frame-to-frame correlation. Rotating transducers often show nonuniform rotational distortion (NURD), which leads to misaligned echo lines in consecutive frames. This results in lateral motion artifacts and causes decorrelation. This effect is reduced by lateral motion correction based on block-matching algorithms. Results show that strain imaging can successfully be performed in vivo with data acquired predominantly in diastole. Different coronary tissue regions can be identified by local strain variations. If NURD is present, strain image quality is degraded. In some cases NURD is reduced by repositioning the transducer.