Search

Artificial intelligence (AI) holds much promise for enabling highly desired imaging diagnostics improvements. One of the most limiting bottlenecks for the development of useful clinical-grade AI models is the lack of training data. One aspect is the large amount of cases needed and another is the necessity of high-quality ground truth annotation. The aim of the project was to establish and describe the construction of a database with substantial amounts of detail-annotated oncology imaging data from pathology and radiology. A specific objective was to be proactive, that is, to support undefined subsequent AI training across a wide range of tasks, such as detection, quantification, segmentation, and classification, which puts particular focus on the quality and generality of the annotations. The main outcome of this project was the database as such, with a collection of labeled image data from breast, ovary, skin, colon, skeleton, and liver. In addition, this effort also served as an exploration of best practices for further scalability of high-quality image collections, and a main contribution of the study was generic lessons learned regarding how to successfully organize efforts to construct medical imaging databases for AI training, summarized as eight guiding principles covering team, process, and execution aspects., Funding Agencies|Linkoping University; Visual Sweden [VS1702]

Restricted access, please contact Mischa.Woisetschlager@regionostergotland.se, claes.lundstrom@liu.se or joel.hedlund@liu.se to request access.

Restricted access, Please contact Mischa.Woisetschlager@regionostergotland.se, claes.lundstrom@liu.se or joel.hedlund@liu.se to request access.

Four-dimensional (4D) flow magnetic resonance imaging (4D Flow MRI) enables acquisition of time-resolved three-directional velocity data in the entire heart and all major thoracic vessels. The segmentation of these tissues is typically performed using semi-automatic methods. Some of which primarily rely on the velocity data and result in a segmentation of the vessels only during the systolic phases. Other methods, mostly applied on the heart, rely on separately acquired balanced Steady State Free Precession (b-SSFP) MR images, after which the segmentations are superimposed on the 4D Flow MRI. While b-SSFP images typically cover the whole cardiac cycle and have good contrast, they suffer from a number of problems, such as large slice thickness, limited coverage of the cardiac anatomy, and being prone to displacement errors caused by respiratory motion. To address these limitations we propose a multi-atlas segmentation method, which relies only on 4D Flow MRI data, to automatically generate four-dimensional segmentations that include the entire thoracic cardiovascular system present in these datasets. The approach was evaluated on 4D Flow MR datasets from a cohort of 27 healthy volunteers and 83 patients with mildly impaired systolic left-ventricular function. Comparison of manual and automatic segmentations of the cardiac chambers at end-systolic and end-diastolic timeframes showed agreements comparable to those previously reported for automatic segmentation methods of b-SSFP MR images. Furthermore, automatic segmentation of the entire thoracic cardiovascular system improves visualization of 4D Flow MRI and facilitates computation of hemodynamic parameters., Funding details: 310612; Funding details: FP7, Seventh Framework Programme; Funding details: 621-2014-6191, VR, Vetenskapsrådet; Funding details: 223615; Funding details: 20140398; Funding text: This work was partially funded by the FP7-funded project DOPPLER-CIP [grant number 223615]; the European Union’s Seventh Framework Programme ( FP7/2007-2013 ) [grant number 310612 ]; the Swedish Research Council [grant number 621-2014-6191 ]; and the Swedish Heart and Lung Foundation [grant number 20140398 ].

Study Design Retrospective study. Objectives To investigate parameters of axial vertebral deformation in patients with scoliosis compared to a control group, and to determine whether these parameters correlated with the severity of spine curvature, measured as the Cobb angle. Summary of Background Data Adolescent idiopathic scoliosis (AIS) is the most common type of spinal deformity. Many studies have investigated vertebral deformation, in terms of wedging and pedicle deformations, but few studies have investigated actual structural changes within vertebrae. Methods This study included 20 patients with AIS (Lenke 1–3, mean age: 15.6 years, range: 11–20). We compared preoperative low-dose computed tomography(CT) examinations of patients with AIS to those of a control group matched for age and sex. The control individuals had no spinal deformity, but they were admitted to the emergency department for trauma CTs. We measured the Cobb angles and the axial vertebral rotation (AVR), axial vertebral bodyasymmetry (AVBA), and frontal vertebral body rotation (FVBR) for the superior end, inferior end, and apical vertebrae, with in-house–developed software. Correlations between entities were investigated with the Pearson correlation test. Results The average Cobb angles were 49.3° and 1.3° for the scoliotic and control groups, respectively. The patient and control groups showed significant differences in the AVRs of all three vertebra levels (p < .01), the AVBAs of the superior end and apical vertebrae (p < .008), and the FVBR of the apical vertebra (p = .011). Correlations were only found between the AVBA and FVBR in the superior end vertebra (r = 0.728, p < .001) and in the apical vertebra (r = 0.713, p < .001). Conclusions Compared with controls, patients with scoliosis showed clear morphologic differences in the midaxial plane vertebrae. Differences in AVR, AVBA, and FVBR were most pronounced at the apical vertebra. The FVBR provided valuable additional informa

A multiple center milestone study of clinical vertebra segmentation is presented in this paper. Vertebra segmentation is a fundamental step for spinal image analysis and intervention. The first half of the study was conducted in the spine segmentation challenge in 2014 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) Workshop on Computational Spine Imaging (CSI 2014). The objective was to evaluate the performance of several state-of-the-art vertebra segmentation algorithms on computed tomography (CT) scans using ten training and five testing dataset, all healthy cases; the second half of the study was conducted after the challenge, where additional 5 abnormal cases are used for testing to evaluate the performance under abnormal cases. Dice coefficients and absolute surface distances were used as evaluation metrics. Segmentation of each vertebra as a single geometric unit, as well as separate segmentation of vertebra substructures, was evaluated. Five teams participated in the comparative study. The top performers in the study achieved Dice coefficient of 0.93 in the upper thoracic, 0.95 in the lower thoracic and 0.96 in the lumbar spine for healthy cases, and 0.88 in the upper thoracic, 0.89 in the lower thoracic and 0.92 in the lumbar spine for osteoporotic and fractured cases. The strengths and weaknesses of each method as well as future suggestion for improvement are discussed. This is the first multi-center comparative study for vertebra segmentation methods, which will provide an up-to-date performance milestone for the fast growing spinal image analysis and intervention. (C) 2016 Elsevier Ltd. All rights reserved.

To present a semi-automatic method with minimal user interaction for quantitative analysis of the patellofemoral motion pattern. 4D CT data capturing the patellofemoral motion pattern of a continuous flexion and extension were collected for five patients prone to patellar luxation both pre- and post-surgically. For the proposed method, an observer would place landmarks in a single 3D volume, which then are automatically propagated to the other volumes in a time sequence. From the landmarks in each volume, the measures patellar displacement, patellar tilt and angle between femur and tibia were computed. Evaluation of the observer variability showed the proposed semi-automatic method to be favorable over a fully manual counterpart, with an observer variability of approximately 1.5 for the angle between femur and tibia, 1.5 mm for the patellar displacement, and 4.0-5.0 for the patellar tilt. The proposed method showed that surgery reduced the patellar displacement and tilt at maximum extension with approximately 10-15 mm and 15-20 for three patients but with less evident differences for two of the patients. A semi-automatic method suitable for quantification of the patellofemoral motion pattern as captured by 4D CT data has been presented. Its observer variability is on par with that of other methods but with the distinct advantage to support continuous motions during the image acquisition.

This paper presents one of the participating methods to the intervertebral disc segmentation challenge organized in conjunction with the 3rd MICCAI Workshop amp; Challenge on Computational Methods and Clinical Applications for Spine Imaging - MICCAI-CSI2015. The presented method consist of three steps. In the first step, vertebral bodies are detected and labeled using integral channel features and a graphical parts model. The second step consists of image registration, where a set of image volumes with corresponding intervertebral disc atlases are registered to the target volume using the output from the first step as initialization. In the final step, the registered atlases are combined using label fusion to derive the final segmentation. The pipeline was evaluated using a set of 15 + 10 T2-weighted image volumes provided as training and test data respectively for the segmentation challenge. For the training data, a mean disc centroid distance of 0.86 mm and an average DICE score of 91% was achieved, and for the test data the corresponding results were 0.90 mm and 90%.

Radiotherapy planning and attenuation correction of PET images require simulation of radiation transport. The necessary physical properties are typically derived from computed tomography (CT) images, but in some cases, including stereotactic neurosurgery and combined PET/MR imaging, only magnetic resonance (MR) images are available. With these applications in mind, we describe how a realistic, patient-specific, pseudo-CT of the head can be derived from anatomical MR images. We refer to the method as atlas-based regression, because of its similarity to atlas-based segmentation. Given a target MR and an atlas database comprising MR and CT pairs, atlas-based regression works by registering each atlas MR to the target MR, applying the resulting displacement fields to the corresponding atlas CTs and, finally, fusing the deformed atlas CTs into a single pseudo-CT. We use a deformable registration algorithm known as the Morphon and augment it with a certainty mask that allows a tailoring of the influence certain regions are allowed to have on the registration. Moreover, we propose a novel method of fusion, wherein the collection of deformed CTs is iteratively registered to their joint mean and find that the resulting mean CT becomes more similar to the target CT. However, the voxelwise median provided even better results; at least as good as earlier work that required special MR imaging techniques. This makes atlas-based regression a good candidate for clinical use.

This paper describes the concept of eigenspine, a concept applicable for determining the correlation between pair-wise combinationsof measures useful for describing the three-dimensional spinal deformities associated with adolescent idiopathic scoliosis. Theproposed data analysis scheme is based upon the use of principal component analysis (PCA) and canonical correlation analysis(CCA). PCA is employed to reduce the dimensionality of the data space, thereby providing a regularization of the measurements,and CCA is employed to determine the linear dependence between pair-wise combinations of different measures. The usefulness ofthe eigenspine concept is demonstrated by analyzing the position and the rotation of all lumbar and thoracic vertebrae as obtainedfrom 46 patients suffering from adolescent idiopathic scoliosis. The analysis showed that the strongest linear relationship is foundbetween the lateral displacement and the coronal rotation of the vertebrae, and that a somewhat weaker but still strong correlationis found between the coronal rotation and the axial rotation of the vertebrae. These results are well in-line with the generalunderstanding of idiopathic scoliosis. Noteworthy though is that the correlation between the anterior-posterior displacement and thesagittal rotation was not as strong as expected and that the obtained results further indicate the need for including the axial vertebralrotation as a measure when characterizing different types of idiopathic scoliosis. Apart from analyzing pair-wise correlationsbetween different measures, the method is believed to be suitable for finding a maximally descriptive low-dimensional combinationof measures describing spinal deformities in idiopathic scoliosis.

In this paper, we propose the concept of eigenspine, a data analysis schemeuseful for quantifying the linear correlation between different measures relevant fordescribing spinal deformities associated with spinal diseases, such as idiopathic scoliosis.The proposed concept builds upon the use of principal component analysis(PCA) and canonical correlation analysis (CCA), where PCA is used to reduce thenumber of dimensions in the measurement space, thereby providing a regularizationof the measurements, and where CCA is used to determine the linear dependence betweenpair-wise combinations of the different measures. To demonstrate the usefulnessof the eigenspine concept, the measures describing position and rotation of thelumbar and the thoracic vertebrae of 22 patients suffering from idiopathic scoliosiswere analyzed. The analysis showed that the strongest linear relationship is foundbetween the anterior-posterior displacement and the sagittal rotation of the vertebrae,and that a somewhat weaker but still strong correlation is found between thelateral displacement and the frontal rotation of the vertebrae. These results are wellin-line with the general understanding of idiopathic scoliosis. Noteworthy though isthat the obtained results from the analysis further proposes axial vertebral rotationas a differentiating measure when characterizing idiopathic scoliosis. Apart fromanalyzing pair-wise linear correlations between different measures, the method isbelieved to be suitable for finding a maximally descriptive low-dimensional combinationof measures describing spinal deformities in idiopathic scoliosis.

The purpose of this work is to investigate, and improve, the feasibility of advanced Region Of Interest (ROI) selection schemes in clinical volume rendering. In particular, this work implements and evaluates an Automated Anatomical ROI (AA-ROI) approach based on the combination of automatic image registration (AIR) and Distance-Based Transfer Functions (DBTFs), designed for automatic selection of complex anatomical shapes without relying on prohibitive amounts of interaction. Domain knowledge and clinical experience has been included in the project through participation of practicing radiologists in all phases of the project. This has resulted in a set of requirements that are critical for Direct Volume Rendering applications to be utilized in clinical practice and a prototype AA-ROI implementation that was developed to addresses critical points in existing solutions. The feasibility of the developed approach was assessed through a study where five radiologists investigated three medical data sets with complex ROIs, using both traditional tools and the developed prototype software. Our analysis indicate that advanced, registration based ROI schemes could increase clinical efficiency in time-critical settings for cases with complex ROIs, but also that their clinical feasibility is conditional with respect to the radiologists trust in the registration process and its application to the data.

This paper presents results from an evaluation of three previously presented methods for segmentation of cell nuclei in lung cytology samples scanned by whole-slide scanners. Whole-slide images from seven cases of endobronchial ultrasound-guided transbronchial needle aspiration samples were used for extracting a number of regions of interest, in which approximately 2700 cell nuclei were manually segmented to form the ground truth. The segmented cells included benign bronchial epithelium, lymphocytes, granulocytes, histiocytes and malignant epithelial cells. The best results were obtained with a method based upon adaptive thresholding and an added step of clustering for distinguishing between cytoplasm and cell nuclei. This method achieved a mean DICE-score of 0.81 and a sensitivity and specificity of 0.88 and 0.81 respectively. In addition, this method was by far the fastest method, with a mean processing time of 7.8 s per image (2048 x 2048 pixels per image). By further improvements, such as lowering the false positive rate and using parallel computing hardware, this method has the potential to form the first building block in a system for computerized screening of whole-slide images in lung cytology.

Medical imaging plays an increasingly important role in modern healthcare. In medical imaging, it is often relevant to relate different images to each other, something which can prove challenging, since there rarely exists a pre-defined mapping between the pixels in different images. Hence, there is a need to find such a mapping/transformation, a procedure known as image registration. Over the years, image registration has been proved useful in a number of clinical situations. Despite this, current use of image registration in clinical practice is rather limited, typically only used for image fusion. The limited use is, to a large extent, caused by excessive computation times, lack of established validation methods/metrics and a general skepticism toward the trustworthiness of the estimated transformations in deformable image registration. This thesis aims to overcome some of the issues limiting the use of image registration, by proposing a set of technical contributions and two clinical applications targeted at improved clinical efficiency. The contributions are made in the context of a generic framework for non-parametric image registration and using an image registration method known as the Morphon. In image registration, regularization of the estimated transformation forms an integral part in controlling the registration process, and in this thesis, two regularizers are proposed and their applicability demonstrated. Although the regularizers are similar in that they rely on local structure analysis, they differ in regard to implementation, where one is implemented as applying a set of filter kernels, and where the other is implemented as solving a global optimization problem. Furthermore, it is proposed to use a set of quadrature filters with parallel scales when estimating the phase-difference, driving the registration. A proposal that brings both accuracy and robustness to the registration process, as shown on a set of challenging image sequences. Computation

Graphics processing units (GPUs) are used today in a wide range of applications, mainly because they can dramatically accelerate parallel computing, are affordable and energy efficient. In the field of medical imaging, GPUs are in some cases crucial for enabling practical use of computationally demanding algorithms. This review presents the past and present work on GPU accelerated medical image processing, and is meant to serve as an overview and introduction to existing GPU implementations. The review covers GPU acceleration of basic image processing operations (filtering, interpolation, histogram estimation and distance transforms), the most commonly used algorithms in medical imaging (image registration, image segmentation and image denoising) and algorithms that are specific to individual modalities (CT, PET, SPECT, MRI, fMRI, DTI, ultrasound, optical imaging and microscopy). The review ends by highlighting some future possibilities and challenges.

The use of polynomial expansion in image registration has previously been shown to be beneficial due to fast convergence and high accuracy. However, earlier work has only briefly out-lined how non-rigid image registration is handled, e.g. not discussing issues like regularization of the displacement field or how to accumulate the displacement field. In this work, it is shown how non-rigid image registration based upon polynomial expansion can be integrated into a generic framework for non-rigid image registration achieving diffeomorphic displacement fields. The proposed non-rigid image registration algorithm using diffeomorphic field accumulation is evaluated on both synthetically deformed data and real image data and compared to additive field accumulation. The results clearly demonstrate the power of the diffeomorphic field accumulation., The original publication is available at www.springerlink.com: Daniel Forsberg, Mats Andersson and Hans Knutsson, Non-rigid Diffeomorphic Image Registration of Medical Images Using Polynomial Expansion, 2012, Lecture Notes in Computer Science, (7325), 304-312.http://dx.doi.org/10.1007/978-3-642-31298-4_36. Copyright: Springer-verlag http://www.springerlink.com/.

This book constitutes the refereed proceedings of the 5th International Workshop on Biomedical Image Registration, WBIR 2012, held in Nashville, Tennessee, USA, in July 2012. The 20 full papers and 11 poster papers included in this volume were carefully reviewed and selected from 44 submitted papers. They full papers are organized in the following topical sections: multiple image sets; brain; non-rigid anatomy; and frameworks and similarity measures., The original publication is available at www.springerlink.com: Daniel Forsberg, Gunnar Farnebäck, Hans Knutsson and Carl-Fredrik Westin, Multi-Modal Image Registration Using Polynomial Expansion and Mutual Information, 2012, Lecture Notes in Computer Science, (7359), 40-49.http://dx.doi.org/10.1007/978-3-642-31340-0_5. Copyright: Springer-verlag http://www.springerlink.com/.

In this paper, we propose a novel algorithm for regularizing displacement fields in image registration. The method uses the local structure tensor and gradients of the displacement field to impose a local metric, which is then used optimizing a global cost function. The method allows for linear operators, such as tensors and differential operators modeling the underlying physical anatomy of the human body in medical images. The algorithm is tested using output from the Morphon image registration algorithm on MRI data as well as synthetic test data and the result is compared to the initial displacement field. The results clearly demonstrate the power of the method and the unique features brought forth through the global optimization approach., Dynamic Context Atlases for Image Denoising and Patient Safety

The visualization of images with a large dynamic range is a difficult task and this is especially the case for gray-level images. In radiology departments, this will force radiologists to review medical images several times, since the images need to be visualized with several different contrast windows (transfer functions) in order for the full information content of each image to be seen. Previously suggested methods for handling this situation include various approaches using histogram equalization and other methods for processing the image data. However, none of these utilize the underlying human anatomy in the images to control the visualization and the fact that different transfer functions are often only relevant for disjoint anatomical regions. In this paper, we propose a method for using model-based local transfer functions. It allows the reviewing radiologist to apply multiple transfer functions simultaneously to a medical image volume. This provides the radiologist with a tool for making the review process more efficient, by allowing him/her to review more of the information in a medical image volume with a single visualization. The transfer functions are automatically assigned to different anatomically relevant regions, based upon a model registered to the volume to be visualized. The transfer functions can be either pre-defined or interactively changed by the radiologist during the review process. All of this is achieved without adding any unfamiliar aspects to the radiologist’s normal work-flow, when reviewing medical image volumes.

The need of image registration is increasing, especially in the medical image domain. The simplest kind of image registration is to match two images that have similar intensity. More advanced cases include the problem of registering images of different intensity, for which phase based algorithms have proven to be superior. In some cases the phase based registration will fail as well, for instance when the images to be registered do not only differ in intensity but also in local phase. This is the case if a dark circle in the reference image is a bright circle in the source image. While rigid registration algorithms can use other parts of the image to calculate the global transformation, this problem is harder to solve for non-rigid registration. The solution that we propose in this work is to use the local phase of the magnitude of the local structure tensor, instead of the local phase of the image intensity. By doing this, we achieve invariance both to the image intensity and to the local phase and thereby only use the structural information, i.e. the shapes of the objects, for registration., The original publication is available at www.springerlink.com: Anders Eklund, Daniel Forsberg, Mats Andersson and Hans Knutsson, Using the Local Phase of the Magnitude of the Local Structure Tensor for Image Registration, 2011, Lecture Notes in Computer Science, (6688), 414-432. http://dx.doi.org/10.1007/978-3-642-21227-7_39 Copyright: Springer-verlag http://www.springerlink.com

The number of available imaging modalities increases both in clinical practice and in clinical studies. Even though data from multiple modalities might be available, image registration is typically only performed using data from a single modality. In this paper, we propose using certainty maps together with multi-channel diffeomorphic demons in order to improve both accuracy and robustness when performing image registration. The proposed method is evaluated using DTI data, multiple region overlap measures and a fiber bundle similarity metric.

Image registration is a well-known concept within the medical image domain and has been shown to be useful in a number of dierent tasks. However, due to sometimes long processing times, image registration is not fully utilized in clinical workows, where time is an important factor. During the last couple of years, a number of signicant projects have been introduced to make the computational power of GPUs available to a wider audience, where the most well known is CUDA. In this paper we present, with the aid of CUDA, a speedup in the range of 38-44x (from 29 minutes to 40 seconds) when implementing a phasebased non-rigid image registration algorithm, known as the Morphon, on a single GPU. The achieved speedup is in the same magnitude as the speedups reported from other non-rigid registration algorithms fully ported to the GPU. Given the impressive speedups, both reported in this paper and other papers, we therefore consider that it is now feasible to eectively integrate image registration into various clinical workows, where time is a critical factor.

Image registration is frequently used within the medical image domain and where methods with high performance are required. The need for high accuracy coupled with high speed is especially important for applications such as adaptive radiation therapy and image-guided surgery. During the last years, a number of significant projects have been introduced to make the computational power of GPUs available to a wider audience. The most well known project is the introduction of CUDA (Compute Unified Device Architecture). In this paper, we present a CUDA based GPU implementation of a non-rigid image registration algorithm, known as the Morphon, and compare it with a CPU implementation of the Morphon. The achieved speedup, in the range of 51-54x, is also compared with speedups reported from other non-rigid registration methods mplemented on the GPU. These include the Demons algorithm and a mutual information based algorithm.