1
|
Martín-González E, Moya-Sáez E, Menchón-Lara RM, Royuela-Del-Val J, Palencia-de-Lara C, Rodríguez-Cayetano M, Simmross-Wattenberg F, Alberola-López C. A clinically viable vendor-independent and device-agnostic solution for accelerated cardiac MRI reconstruction. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2021; 207:106143. [PMID: 34029830 DOI: 10.1016/j.cmpb.2021.106143] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 12/18/2020] [Accepted: 04/25/2021] [Indexed: 06/12/2023]
Abstract
BACKGROUND AND OBJECTIVE Recent research has reported methods that reconstruct cardiac MR images acquired with acceleration factors as high as 15 in Cartesian coordinates. However, the computational cost of these techniques is quite high, taking about 40 min of CPU time in a typical current machine. This delay between acquisition and final result can completely rule out the use of MRI in clinical environments in favor of other techniques, such as CT. In spite of this, reconstruction methods reported elsewhere can be parallelized to a high degree, a fact that makes them suitable for GPU-type computing devices. This paper contributes a vendor-independent, device-agnostic implementation of such a method to reconstruct 2D motion-compensated, compressed-sensing MRI sequences in clinically viable times. METHODS By leveraging our OpenCLIPER framework, the proposed system works in any computing device (CPU, GPU, DSP, FPGA, etc.), as long as an OpenCL implementation is available, and development is significantly simplified versus a pure OpenCL implementation. In OpenCLIPER, the problem is partitioned in independent black boxes which may be connected as needed, while device initialization and maintenance is handled automatically. Parallel implementations of both a groupwise FFD-based registration method, as well as a multicoil extension of the NESTA algorithm have been carried out as processes of OpenCLIPER. Our platform also includes significant development and debugging aids. HIP code and precompiled libraries can be integrated seamlessly as well since OpenCLIPER makes data objects shareable between OpenCL and HIP. This also opens an opportunity to include CUDA source code (via HIP) in prospective developments. RESULTS The proposed solution can reconstruct a whole 12-14 slice CINE volume acquired in 19-32 coils and 20 phases, with an acceleration factor of ranging 4-8, in a few seconds, with results comparable to another popular platform (BART). If motion compensation is included, reconstruction time is in the order of one minute. CONCLUSIONS We have obtained clinically-viable times in GPUs from different vendors, with delays in some platforms that do not have correspondence with its price in the market. We also contribute a parallel groupwise registration subsystem for motion estimation/compensation and a parallel multicoil NESTA subsystem for l1-l2-norm problem solving.
Collapse
Affiliation(s)
- Elena Martín-González
- Laboratorio de Procesado de Imagen (Image Processing Laboratory), Universidad de Valladolid, Valladolid 47011, Spain.
| | - Elisa Moya-Sáez
- Laboratorio de Procesado de Imagen (Image Processing Laboratory), Universidad de Valladolid, Valladolid 47011, Spain.
| | - Rosa-María Menchón-Lara
- Laboratorio de Procesado de Imagen (Image Processing Laboratory), Universidad de Valladolid, Valladolid 47011, Spain.
| | | | | | - Manuel Rodríguez-Cayetano
- Laboratorio de Procesado de Imagen (Image Processing Laboratory), Universidad de Valladolid, Valladolid 47011, Spain.
| | - Federico Simmross-Wattenberg
- Laboratorio de Procesado de Imagen (Image Processing Laboratory), Universidad de Valladolid, Valladolid 47011, Spain.
| | - Carlos Alberola-López
- Laboratorio de Procesado de Imagen (Image Processing Laboratory), Universidad de Valladolid, Valladolid 47011, Spain.
| |
Collapse
|
2
|
Menchón-Lara RM, Royuela-Del-Val J, Simmross-Wattenberg F, Casaseca-de-la-Higuera P, Martín-Fernández M, Alberola-López C. Fast 4D elastic group-wise image registration. Convolutional interpolation revisited. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2021; 200:105812. [PMID: 33160691 DOI: 10.1016/j.cmpb.2020.105812] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/30/2019] [Accepted: 10/15/2020] [Indexed: 06/11/2023]
Abstract
BACKGROUND AND OBJECTIVE This paper proposes a new and highly efficient implementation of 3D+t groupwise registration based on the free-form deformation paradigm. METHODS Deformation is posed as a cascade of 1D convolutions, achieving great reduction in execution time for evaluation of transformations and gradients. RESULTS The proposed method has been applied to 4D cardiac MRI and 4D thoracic CT monomodal datasets. Results show an average runtime reduction above 90%, both in CPU and GPU executions, compared with the classical tensor product formulation. CONCLUSIONS Our implementation, although fully developed for the metric sum of squared differences, can be extended to other metrics and its adaptation to multiresolution strategies is straightforward. Therefore, it can be extremely useful to speed up image registration procedures in different applications where high dimensional data are involved.
Collapse
Affiliation(s)
- Rosa-María Menchón-Lara
- Laboratorio de Procesado de Imagen. ETSI de Telecomunicación, Universidad de Valladolid, Valladolid, Spain.
| | | | | | | | - Marcos Martín-Fernández
- Laboratorio de Procesado de Imagen. ETSI de Telecomunicación, Universidad de Valladolid, Valladolid, Spain
| | - Carlos Alberola-López
- Laboratorio de Procesado de Imagen. ETSI de Telecomunicación, Universidad de Valladolid, Valladolid, Spain
| |
Collapse
|
3
|
Groupwise Non-Rigid Registration with Deep Learning: An Affordable Solution Applied to 2D Cardiac Cine MRI Reconstruction. ENTROPY 2020; 22:e22060687. [PMID: 33286459 PMCID: PMC7517224 DOI: 10.3390/e22060687] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 05/15/2020] [Revised: 06/15/2020] [Accepted: 06/16/2020] [Indexed: 12/22/2022]
Abstract
Groupwise image (GW) registration is customarily used for subsequent processing in medical imaging. However, it is computationally expensive due to repeated calculation of transformations and gradients. In this paper, we propose a deep learning (DL) architecture that achieves GW elastic registration of a 2D dynamic sequence on an affordable average GPU. Our solution, referred to as dGW, is a simplified version of the well-known U-net. In our GW solution, the image that the other images are registered to, referred to in the paper as template image, is iteratively obtained together with the registered images. Design and evaluation have been carried out using 2D cine cardiac MR slices from 2 databases respectively consisting of 89 and 41 subjects. The first database was used for training and validation with 66.6–33.3% split. The second one was used for validation (50%) and testing (50%). Additional network hyperparameters, which are—in essence—those that control the transformation smoothness degree, are obtained by means of a forward selection procedure. Our results show a 9-fold runtime reduction with respect to an optimization-based implementation; in addition, making use of the well-known structural similarity (SSIM) index we have obtained significative differences with dGW with respect to an alternative DL solution based on Voxelmorph.
Collapse
|
4
|
Sanz-Estébanez S, Cordero-Grande L, Sevilla T, Revilla-Orodea A, de Luis-García R, Martín-Fernández M, Alberola-López C. Vortical features for myocardial rotation assessment in hypertrophic cardiomyopathy using cardiac tagged magnetic resonance. Med Image Anal 2018; 47:191-202. [PMID: 29753999 DOI: 10.1016/j.media.2018.03.005] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/06/2017] [Revised: 01/10/2018] [Accepted: 03/14/2018] [Indexed: 11/16/2022]
Abstract
Left ventricular rotational motion is a feature of normal and diseased cardiac function. However, classical torsion and twist measures rely on the definition of a rotational axis which may not exist. This paper reviews global and local rotation descriptors of myocardial motion and introduces new curl-based (vortical) features built from tensorial magnitudes, intended to provide better comprehension about fibrotic tissue characteristics mechanical properties. Fifty-six cardiomyopathy patients and twenty-two healthy volunteers have been studied using tagged magnetic resonance by means of harmonic phase analysis. Rotation descriptors are built, with no assumption about a regular geometrical model, from different approaches. The extracted vortical features have been tested by means of a sequential cardiomyopathy classification procedure; they have proven useful for the regional characterization of the left ventricular function by showing great separability not only between pathologic and healthy patients but also, and specifically, between heterogeneous phenotypes within cardiomyopathies.
Collapse
Affiliation(s)
- Santiago Sanz-Estébanez
- Laboratorio de Procesado de Imagen, Department of Teoría de la Señal y Comunicaciones e Ingeniería Telemática, ETSIT, Universidad de Valladolid, Campus Miguel Delibes s.n., Valladolid 40011, Spain. http://www.lpi.tel.uva.es/ssanest
| | - Lucilio Cordero-Grande
- Centre for the Developing Brain and Department of Biomedical Engineering, Division of Imaging Science and Biomedical Engineering, King's College London, St Thomas' Hospital, London SE1 7EH, U.K.
| | - Teresa Sevilla
- Unidad de Imagen Cardiaca, Hospital Clínico Universitario de Valladolid, CIBER de enfermedades cardiovasculares (CIBERCV), Valladolid 47005, Spain
| | - Ana Revilla-Orodea
- Unidad de Imagen Cardiaca, Hospital Clínico Universitario de Valladolid, CIBER de enfermedades cardiovasculares (CIBERCV), Valladolid 47005, Spain
| | - Rodrigo de Luis-García
- Laboratorio de Procesado de Imagen, Department of Teoría de la Señal y Comunicaciones e Ingeniería Telemática, ETSIT, Universidad de Valladolid, Campus Miguel Delibes s.n., Valladolid 40011, Spain.
| | - Marcos Martín-Fernández
- Laboratorio de Procesado de Imagen, Department of Teoría de la Señal y Comunicaciones e Ingeniería Telemática, ETSIT, Universidad de Valladolid, Campus Miguel Delibes s.n., Valladolid 40011, Spain.
| | - Carlos Alberola-López
- Laboratorio de Procesado de Imagen, Department of Teoría de la Señal y Comunicaciones e Ingeniería Telemática, ETSIT, Universidad de Valladolid, Campus Miguel Delibes s.n., Valladolid 40011, Spain.
| |
Collapse
|
5
|
Pontre B, Cowan BR, DiBella E, Kulaseharan S, Likhite D, Noorman N, Tautz L, Tustison N, Wollny G, Young AA, Suinesiaputra A. An Open Benchmark Challenge for Motion Correction of Myocardial Perfusion MRI. IEEE J Biomed Health Inform 2017; 21:1315-1326. [PMID: 28880152 PMCID: PMC5658235 DOI: 10.1109/jbhi.2016.2597145] [Citation(s) in RCA: 15] [Impact Index Per Article: 2.1] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/10/2022]
Abstract
Cardiac magnetic resonance perfusion examinations enable noninvasive quantification of myocardial blood flow. However, motion between frames due to breathing must be corrected for quantitative analysis. Although several methods have been proposed, there is a lack of widely available benchmarks to compare different algorithms. We sought to compare many algorithms from several groups in an open benchmark challenge. Nine clinical studies from two different centers comprising normal and diseased myocardium at both rest and stress were made available for this study. The primary validation measure was regional myocardial blood flow based on the transfer coefficient (Ktrans), which was computed using a compartment model and the myocardial perfusion reserve (MPR) index. The ground truth was calculated using contours drawn manually on all frames by a single observer, and visually inspected by a second observer. Six groups participated and 19 different motion correction algorithms were compared. Each method used one of three different motion models: rigid, global affine, or local deformation. The similarity metric also varied with methods employing either sum-of-squared differences, mutual information, or cross correlation. There were no significant differences in Ktrans or MPR compared across different motion models or similarity metrics. Compared with the ground truth, only Ktrans for the sum-of-squared differences metric, and for local deformation motion models, had significant bias. In conclusion, the open benchmark enabled evaluation of clinical perfusion indices over a wide range of methods. In particular, there was no benefit of nonrigid registration techniques over the other methods evaluated in this study. The benchmark data and results are available from the Cardiac Atlas Project ( www.cardiacatlas.org).
Collapse
|
6
|
Royuela-del-Val J, Cordero-Grande L, Simmross-Wattenberg F, Martín-Fernández M, Alberola-López C. Jacobian weighted temporal total variation for motion compensated compressed sensing reconstruction of dynamic MRI. Magn Reson Med 2016; 77:1208-1215. [DOI: 10.1002/mrm.26198] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/21/2015] [Revised: 02/10/2016] [Accepted: 02/12/2016] [Indexed: 11/11/2022]
Affiliation(s)
| | - Lucilio Cordero-Grande
- Division of Imaging Sciences and Biomedical Engineering, Centre for the Developing Brain; King's College London; London UK
| | | | | | | |
Collapse
|
7
|
Royuela-del-Val J, Cordero-Grande L, Simmross-Wattenberg F, Martín-Fernández M, Alberola-López C. Nonrigid groupwise registration for motion estimation and compensation in compressed sensing reconstruction of breath-hold cardiac cine MRI. Magn Reson Med 2015; 75:1525-36. [DOI: 10.1002/mrm.25733] [Citation(s) in RCA: 31] [Impact Index Per Article: 3.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/25/2014] [Revised: 03/19/2015] [Accepted: 03/22/2015] [Indexed: 11/06/2022]
Affiliation(s)
| | - Lucilio Cordero-Grande
- Centre for the Developing Brain; Division of Imaging Sciences and Biomedical Engineering; King's College London; London UK
| | | | | | | |
Collapse
|
8
|
Chiusano G, Staglianò A, Basso C, Verri A. Unsupervised tissue segmentation from dynamic contrast-enhanced magnetic resonance imaging. Artif Intell Med 2014; 61:53-61. [PMID: 24661609 DOI: 10.1016/j.artmed.2014.02.001] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/05/2013] [Revised: 02/06/2014] [Accepted: 02/20/2014] [Indexed: 11/28/2022]
Abstract
OBJECTIVE Design, implement, and validate an unsupervised method for tissue segmentation from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). METHODS For each DCE-MRI acquisition, after a spatial registration phase, the time-varying intensity of each voxel is represented as a sparse linear combination of adaptive basis signals. Both the basis signals and the sparse coefficients are learned by minimizing a functional consisting of a data fidelity term and a sparsity inducing penalty. Tissue segmentation is then obtained by applying a standard clustering algorithm to the computed representation. RESULTS Quantitative estimates on two real data sets are presented. In the first case, the overlap with expert annotation measured with the DICE metric is nearly 90% and thus 5% more accurate than state-of-the-art techniques. In the second case, assessment of the correlation between quantitative scores, obtained by the proposed method against imagery manually annotated by two experts, achieved a Pearson coefficient of 0.83 and 0.87, and a Spearman coefficient of 0.83 and 0.71, respectively. CONCLUSIONS The sparse representation of DCE MRI signals obtained by means of adaptive dictionary learning techniques appears to be well-suited for unsupervised tissue segmentation and applicable to different clinical contexts with little effort.
Collapse
Affiliation(s)
- Gabriele Chiusano
- Dipartimento di Informatica, Bioingegneria, Robotica e Ingegneria dei Sistemi, Università degli Studi di Genova, Via Dodecaneso 35, 16146 Genova, Italy.
| | - Alessandra Staglianò
- Dipartimento di Informatica, Bioingegneria, Robotica e Ingegneria dei Sistemi, Università degli Studi di Genova, Via Dodecaneso 35, 16146 Genova, Italy.
| | - Curzio Basso
- CAMELOT Srl, Via Greto di Cornigliano 6R, 16152 Genova, Italy.
| | - Alessandro Verri
- Dipartimento di Informatica, Bioingegneria, Robotica e Ingegneria dei Sistemi, Università degli Studi di Genova, Via Dodecaneso 35, 16146 Genova, Italy.
| |
Collapse
|