Multi-Device Parallel MRI Reconstruction: Efficient Partitioning for Undersampled 5D Cardiac CINE.

Cardiac CINE, a form of dynamic cardiac MRI, is indispensable in the diagnosis and treatment of heart conditions, offering detailed visualization essential for the early detection of cardiac diseases. As the demand for higher-resolution images increases, so does the volume of data requiring processing, presenting significant computational challenges that can impede the efficiency of diagnostic imaging. Our research presents an approach that takes advantage of the computational power of multiple Graphics Processing Units (GPUs) to address these challenges. GPUs are devices capable of performing large volumes of computations in a short period, and have significantly improved the cardiac MRI reconstruction process, allowing images to be produced faster. The innovation of our work resides in utilizing a multi-device system capable of processing the substantial data volumes demanded by high-resolution, five-dimensional cardiac MRI. This system surpasses the memory capacity limitations of single GPUs by partitioning large datasets into smaller, manageable segments for parallel processing, thereby preserving image integrity and accelerating reconstruction times. Utilizing OpenCL technology, our system offers adaptability and cross-platform functionality, ensuring wider applicability. The proposed multi-device approach offers an advancement in medical imaging, accelerating the reconstruction process and facilitating faster and more effective cardiac health assessment.

Computing thickness of irregularly-shaped thin walls using a locally semi-implicit scheme with extrapolation to solve the Laplace equation: Application to the right ventricle.

Cardiac Magnetic Resonance (CMR) Imaging is currently considered the gold standard imaging modality in cardiology. However, it is accompanied by a tradeoff between spatial resolution and acquisition time. Providing accurate measures of thin walls relative to the image resolution may prove challenging. One such anatomical structure is the cardiac right ventricle. Methods for measuring thickness of wall-like anatomical structures often rely on the Laplace equation to provide point-to-point correspondences between both boundaries. This work presents limex, a novel method to solve the Laplace equation using ghost nodes and providing extrapolated values, which is tested on three different datasets: a mathematical phantom, a set of biventricular segmentations from CMR images of ten pigs and the database used at the RV Segmentation Challenge held at MICCAI'12. Thickness measurements using the proposed methodology are more accurate than state-of-the-art methods, especially with the coarsest image resolutions, yielding mean L1 norms of the error between 43.28% and 86.52% lower than the second-best methods on the different test datasets. It is also computationally affordable. Limex has outperformed other state-of-the-art methods in classifying RV myocardial segments by their thickness.

Insight into ADHD diagnosis with deep learning on Actimetry: Quantitative interpretation of occlusion maps in age and gender subgroups.

Attention Deficit/Hyperactivity Disorder (ADHD) is a prevalent neurodevelopmental disorder in childhood that often persists into adulthood. Objectively diagnosing ADHD can be challenging due to the reliance on subjective questionnaires in clinical assessment. Fortunately, recent advancements in artificial intelligence (AI) have shown promise in providing objective diagnoses through the analysis of medical images or activity recordings. These AI-based techniques have demonstrated accurate ADHD diagnosis; however, the growing complexity of deep learning models has introduced a lack of interpretability. These models often function as black boxes, unable to offer meaningful insights into the data patterns that characterize ADHD. OBJECTIVE: This paper proposes a methodology to interpret the output of an AI-based diagnosis system for combined ADHD in age and gender-stratified populations. METHODS: Our system is based on the analysis of 24 hour-long activity records using Convolutional Neural Networks (CNNs) to classify spectrograms of activity windows. These windows are interpreted using occlusion maps to highlight the time-frequency patterns explaining ADHD activity. RESULTS: Significant differences in the frequency patterns between ADHD and controls both in diurnal and nocturnal activity were found for all the populations. Temporal dispersion also presented differences in the male population. CONCLUSION: The proposed interpretation techniques for CNNs highlighted gender- and age-related differences between ADHD patients and controls. Leveraging these differences could potentially lead to improved diagnostic accuracy, especially if a larger and more balanced dataset is utilized. SIGNIFICANCE: Our findings pave the way for the development of an AI-based diagnosis system for ADHD that offers interpretability, thereby providing valuable insights into the underlying etiology of the disease.

Toward deep learning replacement of gadolinium in neuro-oncology: A review of contrast-enhanced synthetic MRI.

Gadolinium-based contrast agents (GBCAs) have become a crucial part of MRI acquisitions in neuro-oncology for the detection, characterization and monitoring of brain tumors. However, contrast-enhanced (CE) acquisitions not only raise safety concerns, but also lead to patient discomfort, the need of more skilled manpower and cost increase. Recently, several proposed deep learning works intend to reduce, or even eliminate, the need of GBCAs. This study reviews the published works related to the synthesis of CE images from low-dose and/or their native -non CE- counterparts. The data, type of neural network, and number of input modalities for each method are summarized as well as the evaluation methods. Based on this analysis, we discuss the main issues that these methods need to overcome in order to become suitable for their clinical usage. We also hypothesize some future trends that research on this topic may follow.

KomaMRI.jl: An open-source framework for general MRI simulations with GPU acceleration.

PURPOSE: To develop an open-source, high-performance, easy-to-use, extensible, cross-platform, and general MRI simulation framework (Koma). METHODS: Koma was developed using the Julia programming language. Like other MRI simulators, it solves the Bloch equations with CPU and GPU parallelization. The inputs are the scanner parameters, the phantom, and the pulse sequence that is Pulseq-compatible. The raw data is stored in the ISMRMRD format. For the reconstruction, MRIReco.jl is used. A graphical user interface utilizing web technologies was also designed. Two types of experiments were performed: one to compare the quality of the results and the execution speed, and the second to compare its usability. Finally, the use of Koma in quantitative imaging was demonstrated by simulating Magnetic Resonance Fingerprinting (MRF) acquisitions. RESULTS: Koma was compared to two well-known open-source MRI simulators, JEMRIS and MRiLab. Highly accurate results (with mean absolute differences below 0.1% compared to JEMRIS) and better GPU performance than MRiLab were demonstrated. In an experiment with students, Koma was proved to be easy to use, eight times faster on personal computers than JEMRIS, and 65% of test subjects recommended it. The potential for designing acquisition and reconstruction techniques was also shown through the simulation of MRF acquisitions, with conclusions that agree with the literature. CONCLUSIONS: Koma's speed and flexibility have the potential to make simulations more accessible for education and research. Koma is expected to be used for designing and testing novel pulse sequences before implementing them in the scanner with Pulseq files, and for creating synthetic data to train machine learning models.

Diffusion sampling schemes: A generalized methodology with nongeometric criteria.

PURPOSE: The aim of this paper is to show that geometrical criteria for designing multishell q $$ q $$ -space sampling procedures do not necessarily translate into reconstruction matrices with high figures of merit commonly used in the compressed sensing theory. In addition, we show that a well-known method for visiting k-space in radial three-dimensional acquisitions, namely, the Spiral Phyllotaxis, is a competitive initialization for the optimization of our nonconvex objective function. THEORY AND METHODS: We propose the gradient design method WISH (WeIghting SHells) which uses an objective function that accounts for weighted distances between gradients within M-tuples of consecutive shells, with M $$ M $$ ranging between 1 and the maximum number of shells S $$ S $$ . All the M $$ M $$ -tuples share the same weight ω M $$ {\omega}_M $$ . The objective function is optimized for a sample of these weights, using Spiral Phyllotaxis as initialization. State-of-the-art General Electrostatic Energy Minimization (GEEM) and Spherical Codes (SC) were used for comparison. For the three methods, reconstruction matrices of the attenuation signal using MAP-MRI were tested using figures of merit borrowed from the Compressed Sensing theory (namely, Restricted Isometry Property -RIP- and Coherence); we also tested the gradient design using a geometric criterion based on Voronoi cells. RESULTS: For RIP and Coherence, WISH got better results in at least one combination of weights, whilst the criterion based on Voronoi cells showed an unrelated pattern. CONCLUSION: The versatility provided by WISH is supported by better results. Optimization in the weight parameter space is likely to provide additional improvements. For a practical design with an intermediate number of gradients, our results recommend to carry out the methodology here used to determine the appropriate gradient table.

Cerebral White Matter Connectivity in Adolescent Idiopathic Scoliosis: A Diffusion Magnetic Resonance Imaging Study.

Adolescent idiopathic scoliosis (AIS) is characterized by the radiographic presence of a frontal plane curve, with a magnitude greater than 10° (Cobb technique). Diffusion MRI can be employed to assess the cerebral white matter. The aim of this study was to analyze, by means of MRI, the presence of any alteration in the connectivity of cerebral white matter in AIS patients. In this study, 22 patients with AIS participated. The imaging protocol consisted in T1 and diffusion-weighted acquisitions. Based on the information from one of the diffusion acquisitions, a whole brain tractography was performed with the MRtrix tool. Tractography is a method to deduce the trajectory of fiber bundles through the white matter based on the diffusion MRI data. By combining cortical segmentation with tractography, a connectivity matrix of size 84 × 84 was constructed using FA (fractional anisotropy), and the number of streamlines as connectomics metrics. The results obtained support the hypothesis that alterations in cerebral white matter connectivity in patients with adolescent idiopathic scoliosis (AIS) exist. We consider that the application of diffusion MRI, together with transcranial magnetic stimulation neurophysiologically, is useful to search the etiology of AIS.

Synthetic MRI improves radiomics-based glioblastoma survival prediction.

Glioblastoma is an aggressive and fast-growing brain tumor with poor prognosis. Predicting the expected survival of patients with glioblastoma is a key task for efficient treatment and surgery planning. Survival predictions could be enhanced by means of a radiomic system. However, these systems demand high numbers of multicontrast images, the acquisitions of which are time consuming, giving rise to patient discomfort and low healthcare system efficiency. Synthetic MRI could favor deployment of radiomic systems in the clinic by allowing practitioners not only to reduce acquisition time, but also to retrospectively complete databases or to replace artifacted images. In this work we analyze the replacement of an actually acquired MR weighted image by a synthesized version to predict survival of glioblastoma patients with a radiomic system. Each synthesized version was realistically generated from two acquired images with a deep learning synthetic MRI approach based on a convolutional neural network. Specifically, two weighted images were considered for the replacement one at a time, a T2w and a FLAIR, which were synthesized from the pairs T1w and FLAIR, and T1w and T2w, respectively. Furthermore, a radiomic system for survival prediction, which can classify patients into two groups (survival >480 days and ≤ 480 days), was built. Results show that the radiomic system fed with the synthesized image achieves similar performance compared with using the acquired one, and better performance than a model that does not include this image. Hence, our results confirm that synthetic MRI does add to glioblastoma survival prediction within a radiomics-based approach.

Time-efficient three-dimensional transmural scar assessment provides relevant substrate characterization for ventricular tachycardia features and long-term recurrences in ischemic cardiomyopathy.

Delayed gadolinium-enhanced cardiac magnetic resonance (LGE-CMR) imaging requires novel and time-efficient approaches to characterize the myocardial substrate associated with ventricular arrhythmia in patients with ischemic cardiomyopathy. Using a translational approach in pigs and patients with established myocardial infarction, we tested and validated a novel 3D methodology to assess ventricular scar using custom transmural criteria and a semiautomatic approach to obtain transmural scar maps in ventricular models reconstructed from both 3D-acquired and 3D-upsampled-2D-acquired LGE-CMR images. The results showed that 3D-upsampled models from 2D LGE-CMR images provided a time-efficient alternative to 3D-acquired sequences to assess the myocardial substrate associated with ischemic cardiomyopathy. Scar assessment from 2D-LGE-CMR sequences using 3D-upsampled models was superior to conventional 2D assessment to identify scar sizes associated with the cycle length of spontaneous ventricular tachycardia episodes and long-term ventricular tachycardia recurrences after catheter ablation. This novel methodology may represent an efficient approach in clinical practice after manual or automatic segmentation of myocardial borders in a small number of conventional 2D LGE-CMR slices and automatic scar detection.

A deep learning approach for synthetic MRI based on two routine sequences and training with synthetic data.

BACKGROUND AND OBJECTIVE: Synthetic magnetic resonance imaging (MRI) is a low cost procedure that serves as a bridge between qualitative and quantitative MRI. However, the proposed methods require very specific sequences or private protocols which have scarcely found integration in clinical scanners. We propose a learning-based approach to compute T1, T2, and PD parametric maps from only a pair of T1- and T2-weighted images customarily acquired in the clinical routine. METHODS: Our approach is based on a convolutional neural network (CNN) trained with synthetic data; specifically, a synthetic dataset with 120 volumes was constructed from the anatomical brain model of the BrainWeb tool and served as the training set. The CNN learns an end-to-end mapping function to transform the input T1- and T2-weighted images to their underlying T1, T2, and PD parametric maps. Then, conventional weighted images unseen by the network are analytically synthesized from the parametric maps. The network can be fine tuned with a small database of actual weighted images and maps for better performance. RESULTS: This approach is able to accurately compute parametric maps from synthetic brain data achieving normalized squared error values predominantly below 1%. It also yields realistic parametric maps from actual MR brain acquisitions with T1, T2, and PD values in the range of the literature and with correlation values above 0.95 compared to the T1 and T2 maps obtained from relaxometry sequences. Further, the synthesized weighted images are visually realistic; the mean square error values are always below 9% and the structural similarity index is usually above 0.90. Network fine tuning with actual maps improves performance, while training exclusively with a small database of actual maps shows a performance degradation. CONCLUSIONS: These results show that our approach is able to provide realistic parametric maps and weighted images out of a CNN that (a) is trained with a synthetic dataset and (b) needs only two inputs, which are in turn obtained from a common full-brain acquisition that takes less than 8 min of scan time. Although a fine tuning with actual maps improves performance, synthetic data is crucial to reach acceptable performance levels. Hence, we show the utility of our approach for both quantitative MRI in clinical viable times and for the synthesis of additional weighted images to those actually acquired.

Magnetic Resonance Simulation in Education: Quantitative Evaluation of an Actual Classroom Experience.

Magnetic resonance is an imaging modality that implies a high complexity for radiographers. Despite some simulators having been developed for training purposes, we are not aware of any attempt to quantitatively measure their educational performance. The present study gives an answer to the question: Does an MRI simulator built on specific functional and non-functional requirements help radiographers learn MRI theoretical and practical concepts better than traditional educational method based on lectures? Our study was carried out in a single day by a total of 60 students of a main hospital in Madrid, Spain. The experiment followed a randomized pre-test post-test design with a control group that used a traditional educational method, and an experimental group that used our simulator. Knowledge level was assessed by means of an instrument with evidence of validity in its format and content, while its reliability was analyzed after the experiment. Statistical differences between both groups were measured. Significant statistical differences were found in favor of the participants who used the simulator for both the post-test score and the gain (difference between post-test and pre-test scores). The effect size turned out to be significant as well. In this work we evaluated a magnetic resonance simulation paradigm as a tool to help in the training of radiographers. The study shows that a simulator built on specific design requirements is a valuable complement to traditional education procedures, backed up with significant quantitative results.

Elastic AlignedSENSE for Dynamic MR Reconstruction: A Proof of Concept in Cardiac Cine.

Numerous methods in the extensive literature on magnetic resonance imaging (MRI) reconstruction exploit temporal redundancy to accelerate cardiac cine. Some of them include motion compensation, which involves high computational costs and long runtimes. In this work, we proposed a method-elastic alignedSENSE (EAS)-for the direct reconstruction of a motion-free image plus a set of nonrigid deformations to reconstruct a 2D cardiac sequence. The feasibility of the proposed approach was tested in 2D Cartesian and golden radial multi-coil breath-hold cardiac cine acquisitions. The proposed approach was compared against parallel imaging compressed sense (sPICS) and group-wise motion corrected compressed sense (GWCS) reconstructions. EAS provides better results on objective measures with considerable less runtime when an acceleration factor is higher than 10×. Subjective assessment of an expert, however, invited proposing the combination of EAS and GWCS as a preferable alternative to GWCS or EAS in isolation.

A clinically viable vendor-independent and device-agnostic solution for accelerated cardiac MRI reconstruction.

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.

Hospitalization trends and chronobiology for mental disorders in Spain from 2005 to 2015.

Rhythm research has had a long tradition in psychiatry, especially in affective disorders. The study of trends in incidence plays a central role in epidemiology and public health. The aims of this research were to describe the socio-demographic and clinical characteristics of persons admitted for psychiatric hospitalization and their trends and periodicity in cases (global and by groups) in Spain over the 11 year study span. We conducted a cross-sectional study of the hospital discharge database of Castilla y León from 2005 to 2015, selecting hospitalizations for psychiatric reasons. Trends in the rates of hospitalization were studied by joinpoint regression analysis. Time series analysis for periodicities was done by spectral analysis, fast Fourier transform, and cosinor analysis. Some 49501 hospitalizations due to psychiatric disorders, out of 2717192 hospital admissions, took place during the study span. Hospitalizations for psychosis were frequent (15949, 32.2%), while such for eating disorders were infrequent, but showed the highest average stay (28 days) and DRG relative weight (2.41). The general trend was a statistically significant 2% annual increase in psychiatric hospitalizations over the 11 year span; substance abuse was the only exception to this trend. The whole population and the subgroups of psychosis and bipolar disorders showed significant circannual (one-year) variation in admissions. The rhythm percentage of the global group was 11.4%, while the rhythm percentages of the psychosis, bipolar, and eating disorders were 17.1%, 17.5%, and 9.6%, respectively (p < .05).

Fast 4D elastic group-wise image registration. Convolutional interpolation revisited.

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.

Groupwise Non-Rigid Registration with Deep Learning: An Affordable Solution Applied to 2D Cardiac Cine MRI Reconstruction.

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.

Integration of an intelligent tutoring system in a magnetic resonance simulator for education: Technical feasibility and user experience.

BACKGROUND AND OBJECTIVE: In this paper we propose to include an intelligent tutoring system (ITS) within a magnetic resonance (MR) simulator that has been developed in house. With this, we intend to measure the impact, in terms of user experience, of including an ITS in our simulator. METHODS: We thoroughly describe the integration procedure and we have tested the benefits of this integration by means of two actual educational experiences, with one of them using the simulator as a standalone tool, and the other with the joint use of simulator+ITS. The experiences have consisted of two online courses with a number of students around 180 in both of them, where measurements of usability, perceived utility and likelihood to recommend were collected. RESULTS: We have observed that the three measurements improved noticeably in the second course with respect to the first one; specifically, overall usability improved by 22.3%, perceived utility by an average of 55.1% and likelihood to recommend by 13.7%. In addition, quantitative measurements are complemented with comments in free text format directly provided by the students. Results show evidence on the benefits of integrating an ITS in terms of quantitative user experience, as well as qualitative comparative comments directly by students of both courses. CONCLUSIONS: This is the first time that an ITS is used within the scope of MR simulation for training purposes. Benefits of integrating an ITS within an MR simulator have been evaluated in terms of user experience, with satisfactory comparative results.

Objective ADHD Diagnosis Using Convolutional Neural Networks Over Daily-Life Activity Records.

Attention Deficit/Hyperactivity Disorder (ADHD) is the most common neurobehavioral disorder in children and adolescents. However, its etiology is still unknown, and this hinders the existence of reliable, fast and inexpensive standard diagnostic methods. OBJECTIVE: This paper proposes an end-to-end methodology for automatic diagnosis of the combined type of ADHD. METHODS: Diagnosis is based on the analysis of 24 hour-long activity records using Convolutional Neural Networks to classify spectrograms of activity windows. RESULTS: We achieve up to [Formula: see text] average sensitivity, [Formula: see text] specificity and AUC values over [Formula: see text]. Overall, our figures overcome those obtained by actigraphy-based methods reported in the literature as well as others based on more expensive (and not so convenient) acquisition methods. CONCLUSION: These results reinforce the idea that combining deep learning techniques together with actimetry can lead to a robust and efficient system for objective ADHD diagnosis. SIGNIFICANCE: Reliance on simple activity measurements leads to an inexpensive and non-invasive objective diagn-ostic method, which can be easily implemented with daily devices.

A Web-Based Educational Magnetic Resonance Simulator: Design, Implementation and Testing.

A new web-based education-oriented magnetic resonance (MR) simulator is presented. We have identified the main requirements that this simulator should comply with, so that trainees can face useful practical tasks such as setting the exact slice position and its properties, selecting the correct protocol or fitting the parameters to acquire an image. The tool follows the client-server model. The client contains the interface that mimics the console of a real machine and several of its features. The server stores anatomical models and executes the bulk of the simulation. This cross-platform simulator has been used in two real educational scenarios. The acceptance of the tool has been measured using two criteria, namely, the System Usability Scale and the Likelihood to Recommend, both with satisfactory results. Therefore, we conclude that given the potential of the tool, it may play a relevant role for the training of MRI operators and other involved personnel.