Landmark-guided diffeomorphic demons algorithm and its application to automatic segmentation of the whole spine and pelvis in CT images

Purpose

A fully automatic multiatlas-based method for segmentation of the spine and pelvis in a torso CT volume is proposed. A novel landmark-guided diffeomorphic demons algorithm is used to register a given CT image to multiple atlas volumes. This algorithm can utilize both grayscale image information and given landmark coordinate information optimally.

Methods

The segmentation has four steps. Firstly, 170 bony landmarks are detected in the given volume. Using these landmark positions, an atlas selection procedure is performed to reduce the computational cost of the following registration. Then the chosen atlas volumes are registered to the given CT image. Finally, voxelwise label voting is performed to determine the final segmentation result.

Results

The proposed method was evaluated using 50 torso CT datasets as well as the public SpineWeb dataset. As a result, a mean distance error of \(0.59\pm 0.14\hbox { mm}\) and a mean Dice coefficient of \(0.90\pm 0.02\) were achieved for the whole spine and the pelvic bones, which are competitive with other state-of-the-art methods.

Conclusion

From the experimental results, the usefulness of the proposed segmentation method was validated.

     

A novel platform for electromagnetic navigated ultrasound bronchoscopy (EBUS)

Purpose

Endobronchial ultrasound transbronchial needle aspiration (EBUS-TBNA) of mediastinal lymph nodes is essential for lung cancer staging and distinction between curative and palliative treatment. Precise sampling is crucial. Navigation and multimodal imaging may improve the efficiency of EBUS-TBNA. We demonstrate a novel EBUS-TBNA navigation system in a dedicated airway phantom.

Methods

Using a convex probe EBUS bronchoscope (CP-EBUS) with an integrated sensor for electromagnetic (EM) position tracking, we performed navigated CP-EBUS in a phantom. Preoperative computed tomography (CT) and real-time ultrasound (US) images were integrated into a navigation platform for EM navigated bronchoscopy. The coordinates of targets in CT and US volumes were registered in the navigation system, and the position deviation was calculated.

Results

The system visualized all tumor models and displayed their fused CT and US images in correct positions in the navigation system. Navigating the EBUS bronchoscope was fast and easy. Mean error observed between US and CT positions for 11 target lesions (37 measurements) was \(2.8\pm 1.0\) mm, maximum error was 5.9 mm.

Conclusion

The feasibility of our novel navigated CP-EBUS system was successfully demonstrated. An EBUS navigation system is needed to meet future requirements of precise mediastinal lymph node mapping, and provides new opportunities for procedure documentation in EBUS-TBNA.

     

Automated muscle segmentation from CT images of the hip and thigh using a hierarchical multi-atlas method

Purpose

Patient-specific quantitative assessments of muscle mass and biomechanical musculoskeletal simulations require segmentation of the muscles from medical images. The objective of this work is to automate muscle segmentation from CT data of the hip and thigh.

Method

We propose a hierarchical multi-atlas method in which each hierarchy includes spatial normalization using simpler pre-segmented structures in order to reduce the inter-patient variability of more complex target structures.

Results

The proposed hierarchical method was evaluated with 19 muscles from 20 CT images of the hip and thigh using the manual segmentation by expert orthopedic surgeons as ground truth. The average symmetric surface distance was significantly reduced in the proposed method (1.53 mm) in comparison with the conventional method (2.65 mm).

Conclusion

We demonstrated that the proposed hierarchical multi-atlas method improved the accuracy of muscle segmentation from CT images, in which large inter-patient variability and insufficient contrast were involved.

     

Extraction of open-state mitral valve geometry from CT volumes

Purpose

The importance of mitral valve therapies is rising due to an aging population. Visualization and quantification of the valve anatomy from image acquisitions is an essential component of surgical and interventional planning. The segmentation of the mitral valve from computed tomography (CT) acquisitions is challenging due to high variation in appearance and visibility across subjects. We present a novel semi-automatic approach to segment the open-state valve in 3D CT volumes that combines user-defined landmarks to an initial valve model which is automatically adapted to the image information, even if the image data provide only partial visibility of the valve.

Methods

Context information and automatic view initialization are derived from segmentation of the left heart lumina, which incorporates topological, shape and regional information. The valve model is initialized with user-defined landmarks in views generated from the context segmentation and then adapted to the image data in an active surface approach guided by landmarks derived from sheetness analysis. The resulting model is refined by user landmarks.

Results

For evaluation, three clinicians segmented the open valve in 10 CT volumes of patients with mitral valve insufficiency. Despite notable differences in landmark definition, the resulting valve meshes were overall similar in appearance, with a mean surface distance of \(1.62 \pm 2.10\) mm. Each volume could be segmented in 5–22 min.

Conclusions

Our approach enables an expert user to easily segment the open mitral valve in CT data, even when image noise or low contrast limits the visibility of the valve.

     

Fully automatic segmentation of paraspinal muscles from 3D torso CT images via multi-scale iterative random forest classifications

Purpose

To develop and validate a fully automatic method for segmentation of paraspinal muscles from 3D torso CT images.

Methods

We propose a novel learning-based method to address this challenging problem. Multi-scale iterative random forest classifications with multi-source information are employed in this study to speed up the segmentation and to improve the accuracy. Here, multi-source images include the original torso CT images and later also the iteratively estimated and refined probability maps of the paraspinal muscles. We validated our method on 20 torso CT data with associated manual segmentation. We randomly partitioned the 20 CT data into two evenly distributed groups and took one group as the training data and the other group as the test data.

Results

The proposed method achieved a mean Dice coefficient of 93.0%. It took on average 46.5 s to segment a 3D torso CT image with the size ranging from \(512 \times 512 \times 802\) voxels to \(512 \times 512 \times 1031\) voxels.

Conclusions

Our fully automatic, learning-based method can accurately segment paraspinal muscles from 3D torso CT images. It generates segmentation results that are better than those achieved by the state-of-the-art methods.

     

Automated liver segmentation from a postmortem CT scan based on a statistical shape model

Purpose

Automated liver segmentation from a postmortem computed tomography (PMCT) volume is a challenging problem owing to the large deformation and intensity changes caused by severe pathology and/or postmortem changes. This paper addresses this problem by a novel segmentation algorithm using a statistical shape model (SSM) for a postmortem liver.

Methods

The location and shape parameters of a liver are directly estimated from a given volume by the proposed SSM-guided expectation–maximization (EM) algorithm without any spatial standardization that might fail owing to the large deformation and intensity changes. The estimated location and shape parameters are then used as a constraint of the subsequent fine segmentation process based on graph cuts. Algorithms with eight different SSMs were trained using 144 in vivo and 32 postmortem livers, and the segmentation algorithm was tested on 32 postmortem livers in a twofold cross validation manner. The segmentation performance is measured by the Jaccard index (JI) between the segmentation result and the true liver label.

Results

The average JI of the segmentation result with the best SSM was 0.8501, which was better compared with the results obtained using conventional SSMs and the results of the previous postmortem liver segmentation with statistically significant difference.

Conclusions

We proposed an algorithm for automated liver segmentation from a PMCT volume, in which an SSM-guided EM algorithm estimated the location and shape parameters of a liver in a given volume accurately. We demonstrated the effectiveness of the proposed algorithm using actual postmortem CT volumes.

     

Scale-adaptive supervoxel-based random forests for liver tumor segmentation in dynamic contrast-enhanced CT scans

Purpose

Toward an efficient clinical management of hepatocellular carcinoma (HCC), we propose a classification framework dedicated to tumor necrosis rate estimation from dynamic contrast-enhanced CT scans. Based on machine learning, it requires weak interaction efforts to segment healthy, active and necrotic liver tissues.

Methods

Our contributions are two-fold. First, we apply random forest (RF) on supervoxels using multi-phase supervoxel-based features that discriminate tissues based on their dynamic in response to contrast agent injection. Second, we extend this technique in a hierarchical multi-scale fashion to deal with multiple spatial extents and appearance heterogeneity. It translates in an adaptive data sampling scheme combining RF and hierarchical multi-scale tree resulting from recursive supervoxel decomposition. By concatenating multi-phase features across the hierarchical multi-scale tree to describe leaf supervoxels, we enable RF to automatically infer the most informative scales without defining any explicit rules on how to combine them.

Results

Assessment on clinical data confirms the benefits of multi-phase information embedded in a multi-scale supervoxel representation for HCC tumor segmentation.

Conclusion

Dedicated but not limited only to HCC management, both contributions reach further steps toward more accurate multi-label tissue classification.

     

Shape-based acetabular cartilage segmentation: application to CT and MRI datasets

Purpose

A new method for acetabular cartilage segmentation in both computed tomography (CT) arthrography and magnetic resonance imaging (MRI) datasets with leg tension is developed and tested.

Methods

The new segmentation method is based on the combination of shape and intensity information. Shape information is acquired according to the predictable nonlinear relationship between the U-shaped acetabulum region and acetabular cartilage. Intensity information is obtained from the acetabular cartilage region automatically to complete the segmentation procedures. This method is evaluated using 54 CT arthrography datasets with two different radiation doses and 20 MRI datasets. Additionally, the performance of this method in identifying acetabular cartilage is compared with four other acetabular cartilage segmentation methods.

Results

This method performed better than the comparison methods. Indeed, this method maintained good accuracy level for 74 datasets independent of the cartilage modality and with minimum user interaction in the bone segmentation procedures. In addition, this method was efficient in noisy conditions and in detection of the damaged cartilages with zero thickness, which confirmed its potential clinical usefulness.

Conclusions

Our new method proposes acetabular cartilage segmentation in three different datasets based on the combination of the shape and intensity information. This method executes well in situations where there are clear boundaries between the acetabular and femoral cartilages. However, the acetabular cartilage and pelvic bone information should be obtained from one dataset such as CT arthrography or MRI datasets with leg traction.

     

Single STE-MR Acquisition in MR-Based Attenuation Correction of Brain PET Imaging Employing a Fully Automated and Reproducible Level-Set Segmentation Approach

Purpose

The aim of this study is to introduce a fully automatic and reproducible short echo-time (STE) magnetic resonance imaging (MRI) segmentation approach for MR-based attenuation correction of positron emission tomography (PET) data in head region.

Procedures

Single STE-MR imaging was followed by generating attenuation correction maps (μ-maps) through exploiting an automated clustering-based level-set segmentation approach to classify head images into three regions of cortical bone, air, and soft tissue. Quantitative assessment was performed by comparing the STE-derived region classes with the corresponding regions extracted from X-ray computed tomography (CT) images.

Results

The proposed segmentation method returned accuracy and specificity values of over 90 % for cortical bone, air, and soft tissue regions. The MR- and CT-derived μ-maps were compared by quantitative histogram analysis.

Conclusions

The results suggest that the proposed automated segmentation approach can reliably discriminate bony structures from the proximal air and soft tissue in single STE-MR images, which is suitable for generating MR-based μ-maps for attenuation correction of PET data.

     

Automatic 3D liver location and segmentation via convolutional neural network and graph cut

Purpose

Segmentation of the liver from abdominal computed tomography (CT) images is an essential step in some computer-assisted clinical interventions, such as surgery planning for living donor liver transplant, radiotherapy and volume measurement. In this work, we develop a deep learning algorithm with graph cut refinement to automatically segment the liver in CT scans.

Methods

The proposed method consists of two main steps: (i) simultaneously liver detection and probabilistic segmentation using 3D convolutional neural network; (ii) accuracy refinement of the initial segmentation with graph cut and the previously learned probability map.

Results

The proposed approach was validated on forty CT volumes taken from two public databases MICCAI-Sliver07 and 3Dircadb1. For the MICCAI-Sliver07 test dataset, the calculated mean ratios of volumetric overlap error (VOE), relative volume difference (RVD), average symmetric surface distance (ASD), root-mean-square symmetric surface distance (RMSD) and maximum symmetric surface distance (MSD) are 5.9, 2.7 %, 0.91, 1.88 and 18.94 mm, respectively. For the 3Dircadb1 dataset, the calculated mean ratios of VOE, RVD, ASD, RMSD and MSD are 9.36, 0.97 %, 1.89, 4.15 and 33.14 mm, respectively.

Conclusions

The proposed method is fully automatic without any user interaction. Quantitative results reveal that the proposed approach is efficient and accurate for hepatic volume estimation in a clinical setup. The high correlation between the automatic and manual references shows that the proposed method can be good enough to replace the time-consuming and nonreproducible manual segmentation method.

     

Early nontumorous CT findings after irreversible electroporation of locally advanced pancreatic cancer

Purpose

The use of irreversible electroporation (IRE) has been a relatively recent development in the palliative treatment of locally advanced pancreatic cancer. With CT as a key modality in patient follow-up, recognition of nontumorous imaging findings is paramount after IRE.

Methods

A retrospective review of patients having undergone IRE for locally advanced pancreatic adenocarcinoma was performed. A total of 36 patients met inclusion criteria and their imaging studies were reviewed by two radiologists. Nontumorous abnormalities identified in the peri-electroporation bed on Computed Tomography (CT) during the early postoperative period (within 30 days) were characterized and classified into categories.

Results

Our results indicate that the most common nontumorous findings in the peri-electroporation bed were vascular, followed by changes involving the gastrointestinal tract, peritoneal cavity, and, infrequently, the biliary tree.

Conclusions

Interpretation of CT imaging of the postoperative peri-electroporation bed is challenging. This review of CT findings allows the radiologist to recognize and anticipate significant nontumorous findings in the peri-electroporation bed during early follow-up after IRE.

     

Automatic abdominal multi-organ segmentation using deep convolutional neural network and time-implicit level sets

Purpose

Multi-organ segmentation from CT images is an essential step for computer-aided diagnosis and surgery planning. However, manual delineation of the organs by radiologists is tedious, time-consuming and poorly reproducible. Therefore, we propose a fully automatic method for the segmentation of multiple organs from three-dimensional abdominal CT images.

Methods

The proposed method employs deep fully convolutional neural networks (CNNs) for organ detection and segmentation, which is further refined by a time-implicit multi-phase evolution method. Firstly, a 3D CNN is trained to automatically localize and delineate the organs of interest with a probability prediction map. The learned probability map provides both subject-specific spatial priors and initialization for subsequent fine segmentation. Then, for the refinement of the multi-organ segmentation, image intensity models, probability priors as well as a disjoint region constraint are incorporated into an unified energy functional. Finally, a novel time-implicit multi-phase level-set algorithm is utilized to efficiently optimize the proposed energy functional model.

Results

Our method has been evaluated on 140 abdominal CT scans for the segmentation of four organs (liver, spleen and both kidneys). With respect to the ground truth, average Dice overlap ratios for the liver, spleen and both kidneys are 96.0, 94.2 and 95.4%, respectively, and average symmetric surface distance is less than 1.3 mm for all the segmented organs. The computation time for a CT volume is 125 s in average. The achieved accuracy compares well to state-of-the-art methods with much higher efficiency.

Conclusion

A fully automatic method for multi-organ segmentation from abdominal CT images was developed and evaluated. The results demonstrated its potential in clinical usage with high effectiveness, robustness and efficiency.

     

Cardiac CT: Technological Advances in Hardware,Software, and Machine Learning Applications

Purpose of Review

Multidetector row computed tomography (CT) allows noninvasive imaging of the heart and coronary arteries. The purpose of this review is to briefly summarize recent advances in CT hardware and software technology, and machine learning applications for cardiovascular imaging.

Recent Findings

In the last decades, there have been significant improvements in CT hardware focusing on faster gantry rotation resulting in improved temporal resolution. Concurrent hardware improvements include improved spatial resolution and higher coverage of the patient, enabling faster acquisition. Advances in cardiac CT software include methods for measurement of noninvasive FFR, coronary plaque characterization, and adipose tissue characteristics around the heart. Machine learning approaches using cardiac CT have been shown to improve both risk of prognosis and lesion-specific ischemia.

Summary

Recent advances in CT hardware and software have expanded the clinical utility of CT for cardiovascular imaging. In the next decades, continued advances can be anticipated in these areas, and in machine learning applications in cardiac CT, as they are incorporated into clinical routine for image acquisition, image analysis, and prediction of patient outcomes.

     

Validation of a method for retroperitoneal tumor segmentation

Purpose

In 2005, an application for surgical planning called AYRA\({\textregistered }\) was designed and validated by different surgeons and engineers at the Virgen del Rocío University Hospital, Seville (Spain). However, the segmentation methods included in AYRA and in other surgical planning applications are not able to segment accurately tumors that appear in soft tissue. The aims of this paper are to offer an exhaustive validation of an accurate semiautomatic segmentation tool to delimitate retroperitoneal tumors from CT images and to aid physicians in planning both radiotherapy doses and surgery.

Methods

A panel of 6 experts manually segmented 11 cases of tumors, and the segmentation results were compared exhaustively with: the results provided by a surgical planning tool (AYRA), the segmentations obtained using a radiotherapy treatment planning system (Pinnacle\(^{\textregistered }\)), the segmentation results obtained by a group of experts in the delimitation of retroperitoneal tumors and the segmentation results using the algorithm under validation.

Results

11 cases of retroperitoneal tumors were tested. The proposed algorithm provided accurate results regarding the segmentation of the tumor. Moreover, the algorithm requires minimal computational time—an average of 90.5% less than that required when manually contouring the same tumor.

Conclusion

A method developed for the semiautomatic selection of retroperitoneal tumor has been validated in depth. AYRA, as well as other surgical and radiotherapy planning tools, could be greatly improved by including this algorithm.

     

Automatic bladder segmentation from CT images using deep CNN and 3D fully connected CRF-RNN

Purpose

Automatic approach for bladder segmentation from computed tomography (CT) images is highly desirable in clinical practice. It is a challenging task since the bladder usually suffers large variations of appearance and low soft-tissue contrast in CT images. In this study, we present a deep learning-based approach which involves a convolutional neural network (CNN) and a 3D fully connected conditional random fields recurrent neural network (CRF-RNN) to perform accurate bladder segmentation. We also propose a novel preprocessing method, called dual-channel preprocessing, to further advance the segmentation performance of our approach.

Methods

The presented approach works as following: first, we apply our proposed preprocessing method on the input CT image and obtain a dual-channel image which consists of the CT image and an enhanced bladder density map. Second, we exploit a CNN to predict a coarse voxel-wise bladder score map on this dual-channel image. Finally, a 3D fully connected CRF-RNN refines the coarse bladder score map and produce final fine-localized segmentation result.

Results

We compare our approach to the state-of-the-art V-net on a clinical dataset. Results show that our approach achieves superior segmentation accuracy, outperforming the V-net by a significant margin. The Dice Similarity Coefficient of our approach (92.24%) is 8.12% higher than that of the V-net. Moreover, the bladder probability maps performed by our approach present sharper boundaries and more accurate localizations compared with that of the V-net.

Conclusion

Our approach achieves higher segmentation accuracy than the state-of-the-art method on clinical data. Both the dual-channel processing and the 3D fully connected CRF-RNN contribute to this improvement. The united deep network composed of the CNN and 3D CRF-RNN also outperforms a system where the CRF model acts as a post-processing method disconnected from the CNN.

     

Fast and automatic bone segmentation and registration of 3D ultrasound to CT for the full pelvic anatomy: a comparative study

Purpose

Ultrasound (US) is a safer alternative to X-rays for bone imaging, and its popularity for orthopedic surgical navigation is growing. Routine use of intraoperative US for navigation requires fast, accurate and automatic alignment of tracked US to preoperative computed tomography (CT) patient models. Our group previously investigated image segmentation and registration to align untracked US to CT of only the partial pelvic anatomy. In this paper, we extend this to study the performance of these previously published techniques over the full pelvis in a tracked framework, to characterize their suitability in more realistic scenarios, along with an additional simplified segmentation method and similarity metric for registration.

Method

We evaluated phase symmetry segmentation, and Gaussian mixture model (GMM) and coherent point drift (CPD) registration methods on a pelvic phantom augmented with human soft tissue images. Additionally, we proposed and evaluated a simplified 3D bone segmentation algorithm we call Shadow–Peak (SP), which uses acoustic shadowing and peak intensities to detect bone surfaces. We paired this with a registration pipeline that optimizes the normalized cross-correlation (NCC) between distance maps of the segmented US–CT images.

Results

SP segmentation combined with the proposed NCC registration successfully aligned tracked US volumes to the preoperative CT model in all trials, in contrast to the other techniques. SP with NCC achieved a median target registration error (TRE) of 2.44 mm (maximum 4.06 mm), when imaging all three anterior pelvic structures, and a mean runtime of 27.3 s. SP segmentation with CPD registration was the next most accurate combination: median TRE of 3.19 mm (maximum 6.07 mm), though a much faster runtime of 4.2 s.

Conclusion

We demonstrate an accurate, automatic image processing pipeline for intraoperative alignment of US–CT over the full pelvis and compare its performance with the state-of-the-art methods. The proposed methods are amenable to clinical implementation due to their high accuracy on realistic data and acceptably low runtimes.

     

A geometric method for the detection and correction of segmentation leaks of anatomical structures in volumetric medical images

Purpose

Patient-specific models of anatomical structures and pathologies generated from volumetric medical images play an increasingly central role in many aspects of patient care. A key task in generating these models is the segmentation of anatomical structures and pathologies of interest. Although numerous segmentation methods are available, they often produce erroneous delineations that require time-consuming modifications.

Methods

   We present a new geometry-based algorithm for the reliable detection and correction of segmentation errors in volumetric medical images. The method is applicable to anatomical structures consisting of a few 3D star-shaped components. First, it detects segmentation errors by casting rays from the initial segmentation interior to its outer surface. It then classifies the segmentation surface into correct and erroneous regions by minimizing an energy functional that incorporates first- and second-order properties of the rays lengths. Finally, it corrects the segmentation errors by computing new locations for the erroneous surface points by Laplace deformation so that the new surface has maximum smoothness with respect to the rays-length gradient magnitude.

Results

   Our evaluation on initial segmentations of 16 abdominal aortic aneurysm and 12 lung tumors in CT scans obtained by both adaptive region-growing and active contours level-set segmentation improved the volumetric overlap error by 66 and 70.5 % respectively, with respect to the ground-truth.

Conclusions

   The advantages of our method are that it is independent of the initial segmentation algorithm that covers a variety of anatomical structures and pathologies, that it does not require a shape prior, and that it requires minimal user interaction.

     

Fully automatic detection of renal cysts in abdominal CT scans

Purpose

Simple renal cysts are a common benign finding in abdominal CT scans. However, since they may evolve in time, simple cysts need to be reported. With an ever-growing number of slices per CT scan, cysts are easily overlooked by the overloaded radiologist. In this paper, we address the detection of simple renal cysts as an incidental finding in a real clinical setting.

Methods

We propose a fully automatic framework for renal cyst detection, supported by a robust segmentation of the kidneys performed by a fully convolutional neural network. A combined 3D distance map of the kidneys and surrounding fluids provides initial candidates for cysts. Eventually, a second convolutional neural network classifies the candidates as cysts or non-cyst objects.

Results

Performance was evaluated on 52 abdominal CT scans selected at random in a real radiological workflow and containing over 70 cysts annotated by an experienced radiologist. Setting the minimal cyst diameter to 10 mm, the algorithm detected 59/70 cysts (true-positive rate = 84.3%) while producing an average of 1.6 false-positive per case.

Conclusions

The obtained results suggest the proposed framework is a promising approach for the automatic detection of renal cysts as incidental findings of abdominal CT scans.

     

Cancer and fertility preservation in Puerto Rico: a qualitative study of healthcare provider perceptions

Purpose

This study aims to examine healthcare provider perceptions of cancer-related infertility and fertility preservation (FP) in an underserved population and to highlight cognitive and structural barriers to use.

Methods

In-depth, semi-structured interviews were conducted with a sample of 16 healthcare providers participating in a larger ethnographic study on cancer survivorship and cancer-related infertility in Puerto Rico, an unincorporated US territory. Interviews were conducted in-person, audio-recorded, transcribed verbatim, and coded using the constant comparative method.

Results

Providers identified several barriers to FP in Puerto Rico: high cost in relation to income levels, lack of insurance coverage, gaps in provider knowledge of fertility clinics and financial assistance, lower prioritization of quality-of-life needs leading to inconsistent physician disclosure of fertility risks, geographical location of fertility clinics, and logistical challenges to maintaining FP offerings. Two factors act as facilitators: a high value placed on patient-provider communication and relationships and the formation of local alliances between the oncology and reproductive medicine fields, potentially leading to increased cross-specialty communication and referral.

Conclusions

Infertility is a continuing source of distress for cancer patients and survivors, and barriers to FP vary cross-culturally. In Puerto Rico, context-specific factors indicate potential areas of intervention. Greater awareness of fertility risks and options can be fostered through physician training in conjunction with organizational measures targeting cost barriers.