Fully automated deep learning-based localization and segmentation of the locus coeruleus in aging and Parkinson's disease using neuromelanin-sensitive MRI.

PURPOSE: Development and performance measurement of a fully automated pipeline that localizes and segments the locus coeruleus in so-called neuromelanin-sensitive magnetic resonance imaging data for the derivation of quantitative biomarkers of neurodegenerative diseases such as Alzheimer's disease and Parkinson's disease. METHODS: We propose a pipeline composed of several 3D-Unet-based convolutional neural networks for iterative multi-scale localization and multi-rater segmentation and non-deep learning-based components for automated biomarker extraction. We trained on the healthy aging cohort and did not carry out any adaption or fine-tuning prior to the application to Parkinson's disease subjects. RESULTS: The localization and segmentation pipeline demonstrated sufficient performance as measured by Euclidean distance (on average around 1.3mm on healthy aging subjects and 2.2mm in Parkinson's disease subjects) and Dice similarity coefficient (overall around [Formula: see text] on healthy aging subjects and [Formula: see text] for subjects with Parkinson's disease) as well as promising agreement with respect to contrast ratios in terms of intraclass correlation coefficient of [Formula: see text] for healthy aging subjects compared to a manual segmentation procedure. Lower values ([Formula: see text]) for Parkinson's disease subjects indicate the need for further investigation and tests before the application to clinical samples. CONCLUSION: These promising results suggest the usability of the proposed algorithm for data of healthy aging subjects and pave the way for further investigations using this approach on different clinical datasets to validate its practical usability more conclusively.

Combining convolutional neural networks and star convex cuts for fast whole spine vertebra segmentation in MRI.

BACKGROUND AND OBJECTIVE: We propose an automatic approach for fast vertebral body segmentation in three-dimensional magnetic resonance images of the whole spine. Previous works are limited to the lower thoracolumbar section and often take minutes to compute, which is problematic in clinical routine, for study data sets with numerous subjects or when the cervical or upper thoracic spine is to be analyzed. METHODS: We address these limitations by a novel graph cut formulation based on vertebra patches extracted along the spine. For each patch, our formulation incorporates appearance and shape information derived from a task-specific convolutional neural network as well as star-convexity constraints that ensure a topologically correct segmentation of each vertebra. When segmenting vertebrae individually, ambiguities will occur due to overlapping segmentations of adjacent vertebrae. We tackle this problem by novel non-overlap constraints between neighboring patches based on so-called encoding swaps. The latter allow us to obtain a globally optimal multi-label segmentation of all vertebrae in polynomial time. RESULTS: We validated our approach on two data sets. The first contains T1- and T2-weighted whole spine images of 64 subjects with varying health conditions. The second comprises 23 T2-weighted thoracolumbar images of young healthy adults and is publicly available. Our method yielded Dice coefficients of 93.8 â€¯±â€¯ 2.6% and 96.0 â€¯±â€¯ 1.0% for both data sets with a run time of 1.35 â€¯±â€¯ 0.08 s and 0.90 â€¯±â€¯ 0.03 s per vertebra on consumer hardware. A complete whole spine segmentation took 32.4 ±â€¯1.92 s on average. CONCLUSIONS: Our results are superior to those of previous works at a fraction of their run time, which illustrates the efficiency and effectiveness of our whole spine segmentation approach.

Multi-segmental spine image registration supporting image-guided interventions of spinal metastases.

BACKGROUND: Radiofrequency ablation was introduced recently to treat spinal metastases, which are among the most common metastases. These minimally-invasive interventions are most often image-guided by flat-panel CT scans, withholding soft tissue contrast like MR imaging. Image fusion of diagnostic MR and operative CT images could provide important and useful information during interventions. METHOD: Diagnostic MR and interventional flat-panel CT scans of 19 patients, who underwent radiofrequency ablations of spinal metastases were obtained. Our presented approach piecewise rigidly registers single vertebrae using normalized gradient fields and embeds them within a fused image. Registration accuracy was determined via Euclidean distances between corresponding landmark pairs of ground truth data. RESULTS: Our method resulted in an average registration error of 2.35mm. An optimal image fusion performed by landmark registrations achieved an average registration error of 1.70mm. Additionally, intra- and inter-reader variability was determined, resulting in mean distances of corresponding landmark pairs of 1.05mm (MRI) and 1.03mm (flat-panel CT) for the intra-reader variability and 1.36mm and 1.28mm for the inter-reader variability, respectively. CONCLUSIONS: Our multi-segmental approach with normalized gradient fields as image similarity measure can handle spine deformations due to patient positioning and avoid time-consuming manually performed registration. Thus, our method can provide practical and applicable intervention support without significantly delaying the clinical workflow or additional workload.

Vertebral body segmentation in wide range clinical routine spine MRI data.

BACKGROUND AND OBJECTIVE: In this work we propose a 3D vertebral body segmentation approach for clinical magnetic resonance (MR) spine imaging. So far, vertebrae segmentation approaches in MR spine imaging are either limited to particular MR imaging sequences or require minutes to compute, which can be hindering in clinical routine. The major contribution of our work is a reasonably precise segmentation result, within seconds and with minimal user interaction, for spine MR imaging commonly used in clinical routine. Our focus lies on the applicability towards a large variety of clinical MR imaging sequences, dealing with low image quality, high anisotropy and spine pathologies. METHODS: Our method starts with a intensity correction step to deal with bias field artifacts and a minimal user-assisted initialization. Next, appearance-based vertebral body probability maps guide a subsequent hybrid level-set segmentation. RESULTS: We tested our method on different MR imaging sequences from 48 subjects. Overall, our evaluation set contains 63 datasets including 419 vertebral bodies, which differ in age, sex and presence of spine pathologies. This is the largest set of reference segmentations of clinical routine spine MR imaging so far. We achieved a Dice coefficient of 86.0%, a mean Euclidean surface distance error of 1.59 ±â€¯0.24 mm and a Hausdorff distance of 6.86 mm. CONCLUSIONS: These results illustrate the robustness of our segmentation approach towards the variety of MR image data, which is a pivotal aspect for clinical usefulness and reliable diagnosis.

Automatic gallbladder segmentation using combined 2D and 3D shape features to perform volumetric analysis in native and secretin-enhanced MRCP sequences.

OBJECTIVES: We aimed to develop the first fully automated 3D gallbladder segmentation approach to perform volumetric analysis in volume data of magnetic resonance (MR) cholangiopancreatography (MRCP) sequences. Volumetric gallbladder analysis is performed for non-contrast-enhanced and secretin-enhanced MRCP sequences. MATERIALS AND METHODS: Native and secretin-enhanced MRCP volume data were produced with a 1.5-T MR system. Images of coronal maximum intensity projections (MIP) are used to automatically compute 2D characteristic shape features of the gallbladder in the MIP images. A gallbladder shape space is generated to derive 3D gallbladder shape features, which are then combined with 2D gallbladder shape features in a support vector machine approach to detect gallbladder regions in MRCP volume data. A region-based level set approach is used for fine segmentation. Volumetric analysis is performed for both sequences to calculate gallbladder volume differences between both sequences. RESULTS: The approach presented achieves segmentation results with mean Dice coefficients of 0.917 in non-contrast-enhanced sequences and 0.904 in secretin-enhanced sequences. CONCLUSION: This is the first approach developed to detect and segment gallbladders in MR-based volume data automatically in both sequences. It can be used to perform gallbladder volume determination in epidemiological studies and to detect abnormal gallbladder volumes or shapes. The positive volume differences between both sequences may indicate the quantity of the pancreatobiliary reflux.

Joint deformable liver registration and bias field correction for MR-guided HDR brachytherapy.

PURPOSE: In interstitial high-dose rate brachytherapy, liver cancer is treated by internal radiation, requiring percutaneous placement of applicators within or close to the tumor. To maximize utility, the optimal applicator configuration is pre-planned on magnetic resonance images. The pre-planned configuration is then implemented via a magnetic resonance-guided intervention. Mapping the pre-planning information onto interventional data would reduce the radiologist's cognitive load during the intervention and could possibly minimize discrepancies between optimally pre-planned and actually placed applicators. METHODS: We propose a fast and robust two-step registration framework suitable for interventional settings: first, we utilize a multi-resolution rigid registration to correct for differences in patient positioning (rotation and translation). Second, we employ a novel iterative approach alternating between bias field correction and Markov random field deformable registration in a multi-resolution framework to compensate for non-rigid movements of the liver, the tumors and the organs at risk. In contrast to existing pre-correction methods, our multi-resolution scheme can recover bias field artifacts of different extents at marginal computational costs. RESULTS: We compared our approach to deformable registration via B-splines, demons and the SyN method on 22 registration tasks from eleven patients. Results showed that our approach is more accurate than the contenders for liver as well as for tumor tissues. We yield average liver volume overlaps of 94.0 ± 2.7% and average surface-to-surface distances of 2.02 ± 0.87 mm and 3.55 ± 2.19 mm for liver and tumor tissue, respectively. The reported distances are close to (or even below) the slice spacing (2.5 - 3.0 mm) of our data. Our approach is also the fastest, taking 35.8 ± 12.8 s per task. CONCLUSION: The presented approach is sufficiently accurate to map information available from brachytherapy pre-planning onto interventional data. It is also reasonably fast, providing a starting point for computer-aidance during intervention.

Automatized spleen segmentation in non-contrast-enhanced MR volume data using subject-specific shape priors.

To develop the first fully automated 3D spleen segmentation framework derived from T1-weighted magnetic resonance (MR) imaging data and to verify its performance for spleen delineation and volumetry. This approach considers the issue of low contrast between spleen and adjacent tissue in non-contrast-enhanced MR images. Native T1-weighted MR volume data was performed on a 1.5 T MR system in an epidemiological study. We analyzed random subsamples of MR examinations without pathologies to develop and verify the spleen segmentation framework. The framework is modularized to include different kinds of prior knowledge into the segmentation pipeline. Classification by support vector machines differentiates between five different shape types in computed foreground probability maps and recognizes characteristic spleen regions in axial slices of MR volume data. A spleen-shape space generated by training produces subject-specific prior shape knowledge that is then incorporated into a final 3D level set segmentation method. Individually adapted shape-driven forces as well as image-driven forces resulting from refined foreground probability maps steer the level set successfully to the segment the spleen. The framework achieves promising segmentation results with mean Dice coefficients of nearly 0.91 and low volumetric mean errors of 6.3%. The presented spleen segmentation approach can delineate spleen tissue in native MR volume data. Several kinds of prior shape knowledge including subject-specific 3D prior shape knowledge can be used to guide segmentation processes achieving promising results.

On computerized methods for spine analysis in MRI: a systematic review.

PURPOSE: In the last decades, the increasing medical interest in magnetic resonance imaging (MRI) of the spine gave rise to a growing number of publications on computerized methods for spine analysis, covering goals such as localization and segmentation of vertebrae and intervertebral discs as well as the extraction and segmentation of the spinal canal and cord. We provide a critical systematic review to work in the field, putting focus on approaches that can be applied to different imaging sequences and settings. METHODS: Work is analysed on two levels. First, methods are reviewed in detail so that the reader understands justifications and constraints of particular work. Second, work is classified according to relevant attributes and tabulated to give an impression on recent trends. We discuss the general methodical and evaluational aspects of the work as well as challenges specific to MRI such as the lack of intensity standardization and partial volume effects. RESULTS: Methods can be condensed to a small number of optimization frameworks, e.g., graphical models, cost-minimal paths and deformable models. Works sharing the same framework mainly differentiate by the types of information, i.e., pose, geometry and appearance, that are used and by the implementation thereof. MRI-specific challenges are rarely addressed explicitly, calling into question the applicability of most methods to changing imaging sequences or settings. Most often, little attention is paid to evaluation, meaning that results lack comparability and reproducibility although publicly available data sets exist. CONCLUSION: The diversity of MRI sequences and settings still poses challenges to computerized spine analysis. Further research is necessary to implement methods that are actually applicable in practice, e.g., in clinical routine or for study purposes. Certainly, manual guidance will be necessary at some point, for instance to deal with changing subject positions. Therefore, future work should put attention to the appropriate integration of manual interaction.

Fully automatized renal parenchyma volumetry using a support vector machine based recognition system for subject-specific probability map generation in native MR volume data.

In epidemiological studies as well as in clinical practice the amount of produced medical image data strongly increased in the last decade. In this context organ segmentation in MR volume data gained increasing attention for medical applications. Especially in large-scale population-based studies organ volumetry is highly relevant requiring exact organ segmentation. Since manual segmentation is time-consuming and prone to reader variability, large-scale studies need automatized methods to perform organ segmentation. Fully automatic organ segmentation in native MR image data has proven to be a very challenging task. Imaging artifacts as well as inter- and intrasubject MR-intensity differences complicate the application of supervised learning strategies. Thus, we propose a modularized framework of a two-stepped probabilistic approach that generates subject-specific probability maps for renal parenchyma tissue, which are refined subsequently by using several, extended segmentation strategies. We present a three class-based support vector machine recognition system that incorporates Fourier descriptors as shape features to recognize and segment characteristic parenchyma parts. Probabilistic methods use the segmented characteristic parenchyma parts to generate high quality subject-specific parenchyma probability maps. Several refinement strategies including a final shape-based 3D level set segmentation technique are used in subsequent processing modules to segment renal parenchyma. Furthermore, our framework recognizes and excludes renal cysts from parenchymal volume, which is important to analyze renal functions. Volume errors and Dice coefficients show that our presented framework outperforms existing approaches.

Fast Three-Dimensional Single-Particle Tracking in Natural Brain Tissue.

Observation of molecular dynamics is often biased by the optical very heterogeneous environment of cells and complex tissue. Here, we have designed an algorithm that facilitates molecular dynamic analyses within brain slices. We adjust fast astigmatism-based three-dimensional single-particle tracking techniques to depth-dependent optical aberrations induced by the refractive index mismatch so that they are applicable to complex samples. In contrast to existing techniques, our online calibration method determines the aberration directly from the acquired two-dimensional image stream by exploiting the inherent particle movement and the redundancy introduced by the astigmatism. The method improves the positioning by reducing the systematic errors introduced by the aberrations, and allows correct derivation of the cellular morphology and molecular diffusion parameters in three dimensions independently of the imaging depth. No additional experimental effort for the user is required. Our method will be useful for many imaging configurations, which allow imaging in deep cellular structures.

Fully Automated Renal Tissue Volumetry in MR Volume Data Using Prior-Shape-Based Segmentation in Subject-Specific Probability Maps.

Organ segmentation in magnetic resonance (MR) volume data is of increasing interest in epidemiological studies and clinical practice. Especially in large-scale population-based studies, organ volumetry is highly relevant requiring exact organ segmentation. Since manual segmentation is time consuming and prone to reader variability, large-scale studies need automatic methods to perform organ segmentation. In this paper, we present an automated framework for renal tissue segmentation that computes renal parenchyma, cortex, and medulla volumetry in native MR volume data without any user interaction. We introduce a novel strategy of subject-specific probability map computation for renal tissue types, which takes inter- and intra-MR-intensity variability into account. Several kinds of tissue-related 2-D and 3-D prior-shape knowledge are incorporated in modularized framework parts to segment renal parenchyma in a final level set segmentation strategy. Subject-specific probabilities for medulla and cortex tissue are applied in a fuzzy clustering technique to delineate cortex and medulla tissue inside segmented parenchyma regions. The novel subject-specific computation approach provides clearly improved tissue probability map quality than existing methods. Comparing to existing methods, the framework provides improved results for parenchyma segmentation. Furthermore, cortex and medulla segmentation qualities are very promising but cannot be compared to existing methods since state-of-the art methods for automated cortex and medulla segmentation in native MR volume data are still missing.

Ultrasound texture-based CAD system for detecting neuromuscular diseases.

PURPOSE: Diagnosis of neuromuscular diseases in ultrasonography is a challenging task since experts are often unable to discriminate between healthy and pathological cases. A computer-aided diagnosis (CAD) system for skeletal muscle ultrasonography was developed and tested for myositis detection in ultrasound images of biceps brachii. METHODS: Several types of features were extracted from rectangular and polygonal image regions-of-interest (ROIs), including first-order statistics, wavelet-based features, and Haralick's features. Features were chosen that are sensitive to the change in contrast and structure for pathological ultrasound images of neuromuscular diseases. The number of features was reduced by applying different sequential feature selection strategies followed by a supervised principal component analysis. For classification, two linear approaches were investigated: Fisher's classifier and the linear support vector machine (SVM) as well as the nonlinear [Formula: see text]-nearest neighbor approach. The CAD system was benchmarked on datasets of 18 subjects, seven of which were healthy, while 11 were affected by myositis. Three expert radiologists provided pre-classification and testing interpretations. RESULTS: Leave-one-out cross-validation on the training data revealed that the linear SVM was best suited for discriminating healthy and pathological muscle tissue, achieving 85/87 % accuracy, 90 % sensitivity, and 83/85 % specificity, depending on the radiologist. CONCLUSION: A muscle ultrasonography CAD system was developed, allowing a classification of an ultrasound image by one-click positioning of rectangular ROIs with minimal user effort. The applicability of the system was demonstrated with the challenging example of myositis detection, showing highly accurate results that were robust to imprecise user input.

Semi-automatic motion compensation of contrast-enhanced ultrasound images from abdominal organs for perfusion analysis.

This paper presents a system for correcting motion influences in time-dependent 2D contrast-enhanced ultrasound (CEUS) images to assess tissue perfusion characteristics. The system consists of a semi-automatic frame selection method to find images with out-of-plane motion as well as a method for automatic motion compensation. Translational and non-rigid motion compensation is applied by introducing a temporal continuity assumption. A study consisting of 40 clinical datasets was conducted to compare the perfusion with simulated perfusion using pharmacokinetic modeling. Overall, the proposed approach decreased the mean average difference between the measured perfusion and the pharmacokinetic model estimation. It was non-inferior for three out of four patient cohorts to a manual approach and reduced the analysis time by 41% compared to manual processing.

A system to detect cerebral aneurysms in multimodality angiographic data sets.

PURPOSE: The early detection of cerebral aneurysms plays a major role in preventing subarachnoid hemorrhage. The authors present a system to automatically detect cerebral aneurysms in multimodal 3D angiographic data sets. The authors' system is parametrizable for contrast-enhanced magnetic resonance angiography (CE-MRA), time-of-flight magnetic resonance angiography (TOF-MRA), and computed tomography angiography (CTA). METHODS: Initial volumes of interest are found by applying a multiscale sphere-enhancing filter. Several features are combined in a linear discriminant function (LDF) to distinguish between true aneurysms and false positives. The features include shape information, spatial information, and probability information. The LDF can either be parametrized by domain experts or automatically by training. Vessel segmentation is avoided as it could heavily influence the detection algorithm. RESULTS: The authors tested their method with 151 clinical angiographic data sets containing 112 aneurysms. The authors reach a sensitivity of 95% with CE-MRA data sets at an average false positive rate per data set (FPDS) of 8.2. For TOF-MRA, we achieve 95% sensitivity at 11.3 FPDS. For CTA, we reach a sensitivity of 95% at 22.8 FPDS. For all modalities, the expert parametrization led to similar or better results than the trained parametrization eliminating the need for training. 93% of aneurysms that were smaller than 5 mm were found. The authors also showed that their algorithm is capable of detecting aneurysms that were previously overlooked by radiologists. CONCLUSIONS: The authors present an automatic system to detect cerebral aneurysms in multimodal angiographic data sets. The system proved as a suitable computer-aided detection tool to help radiologists find cerebral aneurysms.

Detection of spontaneous vesicle release at individual synapses using multiple wavelets in a CWT-based algorithm.

In this paper we present an algorithm for the detection of spontaneous activity at individual synapses in microscopy images. By employing the optical marker pHluorin, we are able to visualize synaptic vesicle release with a spatial resolution in the nm range in a non-invasive manner. We compute individual synaptic signals from automatically segmented regions of interest and detect peaks that represent synaptic activity using a continuous wavelet transform based algorithm. As opposed to standard peak detection algorithms, we employ multiple wavelets to match all relevant features of the peak. We evaluate our multiple wavelet algorithm (MWA) on real data and assess the performance on synthetic data over a wide range of signal-to-noise ratios.

Automatic segmentation of the left ventricle in 3D SPECT data by registration with a dynamic anatomic model.

We present a fully automatic 3D segmentation method for the left ventricle (LV) in human myocardial perfusion SPECT data. This model-based approach consists of 3 phases: 1. finding the LV in the dataset, 2. extracting its approximate shape and 3. segmenting its exact contour. Finding of the LV is done by flexible pattern matching, whereas segmentation is achieved by registering an anatomical model to the functional data. This model is a new kind of stable 3D mass spring model using direction-weighted 3D contour sensors. Our approach is much faster than manual segmention, which is standard in this application up to now. By testing it on 41 LV SPECT datasets of mostly pathological data, we could show, that it is very robust and its results are comparable with those made by human experts.