Search

Purpose Minimally invasive procedures, such as microwave ablation, are becoming first-line treatment options for early-stage liver cancer due to lower complication rates and shorter recovery times than conventional surgical techniques. Although these procedures are promising, one reason preventing widespread adoption is inadequate local tumor ablation leading to observations of higher local cancer recurrence compared to conventional procedures. Poor ablation coverage has been associated with two-dimensional (2D) ultrasound (US) guidance of the therapy needle applicators and has stimulated investigation into the use of three-dimensional (3D) US imaging for these procedures. We have developed a supervised 3D US needle applicator segmentation algorithm using a single user input to augment the addition of 3D US to the current focal liver tumor ablation workflow with the goals of identifying and improving needle applicator localization efficiency. Methods The algorithm is initialized by creating a spherical search space of line segments around a manually chosen seed point that is selected by a user on the needle applicator visualized in a 3D US image. The most probable trajectory is chosen by maximizing the count and intensity of threshold voxels along a line segment and is filtered using the Otsu method to determine the tip location. Homogeneous tissue mimicking phantom images containing needle applicators were used to optimize the parameters of the algorithm prior to a four-user investigation on retrospective 3D US images of patients who underwent microwave ablation for liver cancer. Trajectory, axis localization, and tip errors were computed based on comparisons to manual segmentations in 3D US images. Results Segmentation of needle applicators in ten phantom 3D US images was optimized to median (Q1, Q3) trajectory, axis, and tip errors of 2.1 (1.1, 3.6)°, 1.3 (0.8, 2.1) mm, and 1.3 (0.7, 2.5) mm, respectively, with a mean ± SD segmentation computation time of 0.246 ± 0.007 s. Use of the segmentation method with a 16 in vivo 3D US patient dataset resulted in median (Q1, Q3) trajectory, axis, and tip errors of 4.5 (2.4, 5.2)°, 1.9 (1.7, 2.1) mm, and 5.1 (2.2, 5.9) mm based on all users. Conclusions Segmentation of needle applicators in 3D US images during minimally invasive liver cancer therapeutic procedures could provide a utility that enables enhanced needle applicator guidance, placement verification, and improved clinical workflow. A semi-automated 3D US needle applicator segmentation algorithm used in vivo demonstrated localization of the visualized trajectory and tip with less than 5° and 5.2 mm errors, respectively, in less than 0.31 s. This offers the ability to assess and adjust needle applicator placements intraoperatively to potentially decrease the observed liver cancer recurrence rates associated with current ablation procedures. Although optimized for deep and oblique angle needle applicator insertions, this proposed workflow has the potential to be altered for a variety of image-guided minimally invasive procedures to improve localization and verification of therapy needle applicators intraoperatively.

Purpose: Manual segmentation of lung tumors is observer dependent and time-consuming but an important component of radiology and radiation oncology workflow. The objective of this study was to generate an automated lung tumor measurement tool for segmentation of pulmonary metastatic tumors from x-ray computed tomography (CT) images to improve reproducibility and decrease the time required to segment tumor boundaries. Methods: The authors developed an automated lung tumor segmentation algorithm for volumetric image analysis of chest CT images using shape constrained Otsu multithresholding (SCOMT) and sparse field active surface (SFAS) algorithms. The observer was required to select the tumor center and the SCOMT algorithm subsequently created an initial surface that was deformed using level set SFAS to minimize the total energy consisting of mean separation, edge, partial volume, rolling, distribution, background, shape, volume, smoothness, and curvature energies. Results: The proposed segmentation algorithm was compared to manual segmentation whereby 21 tumors were evaluated using one-dimensional (1D) response evaluation criteria in solid tumors (RECIST), two-dimensional (2D) World Health Organization (WHO), and 3D volume measurements. Linear regression goodness-of-fit measures (r2 = 0.63, p

Purpose: To quantitatively evaluate local carotid arterial statin effects in 3D USimages using multiclassifier image texture analysis tools. Methods: Texture analysis tools were used to evaluate the effect of 80 mg atorvastatin administered daily to patients with carotid stenosis compared to those treated with placebo. Using three-dimensional carotid ultrasoundimages, 270 texture features from seven texture techniques were extracted from manually segmented carotid arteries based on the intima-media boundary [vessel wall (VW)]. Individual texture features were compared to the previously determined changes in VW volume (VWV) using the distance between classes, the Wilcoxon rank sum test, and accuracy of the classifiers. Texture features that resulted in maximal classification accuracy from each texture technique were selected using Pudil’s sequential floating forward selection (SFFS) as a method of ranking each technique. Finally, SFFS-selected texture features from all texture techniques were used in combination with 24 classifier fusion techniques to improve classification accuracy. Results: Using the measurement of change in VWV, the distance between classes (DBC), Wilcoxon rank sum (WRS) p -value, and median accuracy measures (ACC) were 0.3798, 0.076, and 54.50%, respectively. Texture features improved the detection of statin-related changes using DBC to 0.5199, using WRS to 0.002, and ACC to 63.87%, respectively. The texture techniques that most differentiated between atorvastatin and placebo classes were Fourier power spectrum and Laws texture energy measures. The average classification accuracy between atorvastatin and placebo classes was improved from 57.22 ± 12.11 % using VWV to 97.87 ± 3.93 % using specific texture features. Furthermore, the use of specific texture features resulted in the average area under the receiver-operator characteristic curve (AUC) a value of 0.9988 ± 0.0069 compared to 0.617 ± 0.15 using carotid VWV. Conclusions: Based on DBC, WRS, ACC, and AUC texture features derived from 3D carotid ultrasound were observed to be more sensitive in detecting statin-related changes in carotid atherosclerosis than VWV suggesting that texture classifiers can be used to detect changes in carotid atherosclerosis after therapy.

We present a semi‐automated algorithm for segmenting lungtumours on chest computed tomography(CT)images to be utilized for monitoring tumour response or progression. Seven lungtumours were evaluated; each tumour was radially sliced into 10 slices for a total of 70 radial slices, and manually segmented. The manual segmentations were used as the ground truth for evaluating the proposed algorithm. The tumourimage on each radial slice was classified into two categories: (1) well‐defined boundaries, located centrally in the lung parenchyma without significant vasculature; and (2) vascularized or juxtapleural (VJ). To segment the well‐defined boundaries, a shape constrained multi‐thresholding technique with one user‐defined seed point on the tumour is applied. For vascularized and juxtapleural tumours, this multi‐thresholding technique provided an initial contour that was deformed by a level set sparse field active model to produce the final segmentation for these tumours. The dice index (DI) measure was adopted to evaluate the segmentation results. The average and standard deviation values of the DI for the well‐defined and VJ images were 96.32 ± 1.12% and 95.19 ± 1.58%, respectively. The DI overall average and standard deviation of the 70 slices was 95.50 ± 1.55%. The preliminary results show that using the proposed algorithm produced accurate results.

Three-dimensional (3D) carotid ultrasound (US) phenotypes are increasingly being investigated for quantifying carotid arthrosclerosis for monitoring and assessment of patients who are at a greater risk of stroke. Vessel wall volume (VWV), which is the 3D measurement of vessel wall thickness plus plaque within the carotid artery, provides a high measurement sensitivity and reproducibility for carotid arthrosclerosis. VWV computation requires lumen and media-adventitia boundaries of the carotid arteries to be outlined. Its use in current clinical practice is limited by the requirement for manual segmentation which is extremely time consuming and operator-dependent. Therefore the objective of this study is to develop and validate a semiautomatic segmentation algorithm for delineating the media-adventitia and lumen boundaries of carotid arteries for patients with asymptomatic carotid stenosis. Due to the presence of plaque, poor definition of the vessel boundaries, intensity heterogeneity, image speckle, and shadowing, the carotid arteries are extremely challenging to segment using image information alone. Therefore, we combine various image cues with domain knowledge of the vessel geometry and some user interaction into the segmentation framework. We adopted an energy minimization approach based on the level sets formulation to segment the vessel wall and lumen using edge-based and region-based objective functions respectively. The proposed segmentation method was evaluated with respect to manually outlined boundaries using several similarity measures on 60 2D US slices from ten patients. The realization of semi-automated methods will accelerate the translation of 3DUS to real-time clinical research and clinical care.