Raw Magnetic Resonance Imaging Research Articles

Estimation and assessment of sagittal spinal curvature and thoracic muscle morphometry in different postures.

Spine models are typically developed from supine clinical imaging data, and hence clearly do not fully reflect postures that replicate subjects' clinical symptoms. Our objectives were to develop a method to: (i) estimate the subject-specific sagittal curvature of the whole spine in different postures from limited imaging data, (ii) obtain muscle lines-of-action in different postures and analyze the effect of posture on muscle fascicle length, and (iii) correct for cosine between the magnetic resonance imaging (MRI) scan plane and dominant fiber line-of-action for muscle parameters (cross-sectional area (CSA) and position). The thoracic spines of six healthy volunteers were scanned in four postures (supine, standing, flexion, and sitting) in an upright MRI. Geometry of the sagittal spine was approximated with a circular spline. A pipeline was developed to estimate spine geometry in different postures and was validated. The lines-of-action for two muscles, erector spinae (ES) and transversospinalis (TS) were obtained for every posture and hence muscle fascicle lengths were computed. A correction factor based on published literature was then computed and applied to the muscle parameters. The maximum registration error between the estimated spine geometry and MRI data was small (average RMSE∼1.2%). The muscle fascicle length increased (up to 20%) in flexion when compared to erect postures. The correction factor reduced muscle parameters (∼5% for ES and ∼25% for TS) when compared to raw MRI data. The proposed pipeline is a preliminary step in subject-specific modeling. Direction cosines of muscles could be used while improving the inputs of spine models.

A novel approach for computerized quantitative image analysis of proximal femur bone shape deformities based on the hip joint symmetry

As a result of most of the bone disorders seen in hip joints, shape deformities occur in the structural form of the hip joint components. Image-based quantitative analysis and assessment of these deformities in bone shapes are very important for the evaluation, treatment, and prognosis of the various hip joint bone disorders. In this article, a novel approach for the image-based computerized quantitative analysis of proximal femur shape deformities is presented. In the proposed approach, shape deformities of the pathological proximal femurs were quantified over the contralateral healthy proximal femur shape structure of the same patient in 2D by taking the hip joint symmetry property of human anatomy into consideration. It is based on the idea that if the right and left proximal femurs in bilateral hip joints are highly symmetrical and also if one of the proximal femurs is healthy and the contralateral one is pathological, the non-overlapping bone shape regions can represent the deformities in pathological proximal femurs when both proximal femurs are registered to overlap each other. In the methodological process of the proposed study, a set of image preprocessing operations was primarily performed on the raw magnetic resonance imaging (MRI) data. Then, the segmented proximal femurs in bilateral hip joint images were automatically aligned with the Iterative Closest Point (ICP) rigid registration method. Following the registration, a set of image postprocessing operations was performed on the images of proximal femurs aligned. In the quantification phase, the bone shape deformities in pathological proximal femurs were quantified simply in terms of the mismatching area in 2D by measuring a shape variation index representing the total bone shape deformity ratio. To evaluate the proposed quantitative shape analysis approach, bilateral hip joints in a total of 13 coronal MRI sections of 13 patients with Legg-Calve-Perthes disease (LCPD) were used. Experimental studies have shown that the proposed approach has quite promising results in the quantitative representation of the pathological proximal femur shape deformities. Furthermore, consistent results have been observed for the Waldenström classification stages of the disease. The shape deformity ratios in pathological proximal femurs were quantified as 9.44% (±1.40), 18.38% (±6.30), 24.73% (±12.42), and 27.66% (±10.41), respectively for the Initial, Fragmentation, Reossification, and Remodelling stages of LCPD with the quantification error rates of 0.29% (±0.16), 0.58% (±0.71), 1.12% (±0.82), and 0.80% (±0.98). Additionally, a mean error rate of 0.65% (±0.68) was observed for the quantified shape deformity ratios of all samples.

Brain Age Prediction With Morphological Features Using Deep Neural Networks: Results From Predictive Analytic Competition 2019.

Morphological changes in the brain over the lifespan have been successfully described by using structural magnetic resonance imaging (MRI) in conjunction with machine learning (ML) algorithms. International challenges and scientific initiatives to share open access imaging datasets also contributed significantly to the advance in brain structure characterization and brain age prediction methods. In this work, we present the results of the predictive model based on deep neural networks (DNN) proposed during the Predictive Analytic Competition 2019 for brain age prediction of 2638 healthy individuals. We used FreeSurfer software to extract some morphological descriptors from the raw MRI scans of the subjects collected from 17 sites. We compared the proposed DNN architecture with other ML algorithms commonly used in the literature (RF, SVR, Lasso). Our results highlight that the DNN models achieved the best performance with MAE = 4.6 on the hold-out test, outperforming the other ML strategies. We also propose a complete ML framework to perform a robust statistical evaluation of feature importance for the clinical interpretability of the results.

Intensity non-uniformity correction in MR imaging using residual cycle generative adversarial network

Correcting or reducing the effects of voxel intensity non-uniformity (INU) within a given tissue type is a crucial issue for quantitative magnetic resonance (MR) image analysis in daily clinical practice. Although having no severe impact on visual diagnosis, the INU can highly degrade the performance of automatic quantitative analysis such as segmentation, registration, feature extraction and radiomics. In this study, we present an advanced deep learning based INU correction algorithm called residual cycle generative adversarial network (res-cycle GAN), which integrates the residual block concept into a cycle-consistent GAN (cycle-GAN). In cycle-GAN, an inverse transformation was implemented between the INU uncorrected and corrected magnetic resonance imaging (MRI) images to constrain the model through forcing the calculation of both an INU corrected MRI and a synthetic corrected MRI. A fully convolution neural network integrating residual blocks was applied in the generator of cycle-GAN to enhance end-to-end raw MRI to INU corrected MRI transformation. A cohort of 55 abdominal patients with T1-weighted MR INU images and their corrections with a clinically established and commonly used method, namely, N4ITK were used as a pair to evaluate the proposed res-cycle GAN based INU correction algorithm. Quantitatively comparisons of normalized mean absolute error (NMAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC) indices, and spatial non-uniformity (SNU) were made among the proposed method and other approaches. Our res-cycle GAN based method achieved an NMAE of 0.011 ± 0.002, a PSNR of 28.0 ± 1.9 dB, an NCC of 0.970 ± 0.017, and a SNU of 0.298 ± 0.085. Our proposed method has significant improvements (p < 0.05) in NMAE, PSNR, NCC and SNU over other algorithms including conventional GAN and U-net. Once the model is well trained, our approach can automatically generate the corrected MR images in a few minutes, eliminating the need for manual setting of parameters.

Repeatability of radiomic features in magnetic resonance imaging of glioblastoma: Test-retest and image registration analyses.

To assess the repeatability of radiomic features in magnetic resonance (MR) imaging of glioblastoma (GBM) tumors with respect to test-retest, different image registration approaches and inhomogeneity bias field correction. We analyzed MR images of 17 GBM patients including T1- and T2-weighted images (performed within the same imaging unit on two consecutive days). For image segmentation, we used a comprehensive segmentation approach including entire tumor, active area of tumor, necrotic regions in T1-weighted images, and edema regions in T2-weighted images (test studies only; registration to retest studies is discussed next). Analysis included N3, N4 as well as no bias correction performed on raw MR images. We evaluated 20 image registration approaches, generated by cross-combination of four transformation and five cost function methods. In total, 714 images (17 patients×2 images×((4 transformations×5 cost functions)+1 test image) and 2856 segmentations (714 images×4 segmentations) were prepared for feature extraction. Various radiomic features were extracted, including the use of preprocessing filters, specifically wavelet (WAV) and Laplacian of Gaussian (LOG), as well as discretization into fixed bin width and fixed bin count (16, 32, 64, 128, and 256), Exponential, Gradient, Logarithm, Square and Square Root scales. Intraclass correlation coefficients (ICC) were calculated to assess the repeatability of MRI radiomic features (high repeatability defined as ICC≥95%). In our ICC results, we observed high repeatability (ICC≥95%) with respect to image preprocessing, different image registration algorithms, and test-retest analysis, for example: RLNU and GLNU from GLRLM, GLNU and DNU from GLDM, Coarseness and Busyness from NGTDM, GLNU and ZP from GLSZM, and Energy and RMS from first order. Highest fraction (percent) of repeatable features was observed, among registration techniques, for the method Full Affine transformation with 12 degrees of freedom using Mutual Information cost function (mean 32.4%), and among image processing methods, for the method Laplacian of Gaussian (LOG) with Sigma (2.5-4.5mm) (mean 78.9%). The trends were relatively consistent for N4, N3, or no bias correction. Our results showed varying performances in repeatability of MR radiomic features for GBM tumors due to test-retest and image registration. The findings have implications for appropriate usage in diagnostic and predictive models.

Discriminative and convergent validity of strain elastography for detecting tendinopathy within the supraspinatus tendon: a cross-sectional study

BackgroundWe aimed to explore the discriminative validity of ultrasound strain elastography (SEL) between patients with painful supraspinatus tendinopathy and healthy control shoulders, as well as the associations between SEL and magnetic resonance imaging (MRI), conventional ultrasound (tendon thickness), and the Disabilities of the Arm, Shoulder and Hand questionnaire (DASH).MethodsThirty patients with shoulder pain and MRI-verified supraspinatus tendinopathy and 30 healthy control shoulders (no pain) were examined using SEL, MRI, and conventional ultrasound of the supraspinatus tendon. SEL variables included raw data, ratios between the deltoid muscle and supraspinatus tendon (deltoid ratio), color rating, and presence of red/yellow lesions (middle, worst part, and total tendon).ResultsStatistically significant increases in odds ratios for being symptomatic (increased softening) were seen for all raw data variables, corresponding to 3.978 (95% confidence interval [CI], 1.414-11.197) for middle, 4.602 (95% CI, 1.536-13.788) for worst, and 4.865 (95% CI, 1.406-16.836) for total tendon, and 1.260 (95% CI, 1.027-1.545) for the deltoid ratio (worst), adjusted for sex and body mass index (BMI). Tendon thickness was not associated with SEL; however, significantly positive associations were found between raw data variables and MRI (β ≥ 0.58, P < .01), and positive associations were found between raw data variables and the DASH score (β = 0.01, P ≤ .04), adjusted for sex and BMI.ConclusionsRaw data variables and the deltoid ratio (worst) discriminated between patients with painful supraspinatus tendinopathy and healthy control shoulders when adjusted for sex and BMI. Associations were statistically significant for raw data variables and MRI or DASH score when adjusted for sex and BMI. Further studies are needed to understand SEL and the role of sex and BMI, including the responsiveness of SEL.

Segmentation and Detection of Brain Tumor by using Machine Learning

The segmentation and detection of brain pathologies in medical images is an indispensible step. This helps the radiologist to diagnose a variety of brain deformity and helps in the set up for a suitable treatment. Magnetic Resonance Imaging (MRI) plays a significant character in the research area of neuroscience. The proposed work is a study and probing of different classification techniques used for automated detection and segmentation of brain tumor from MRI in the field of machine learning. This paper try to present the feature extraction from raw MRI and fed the same to four classifier named as, Support Vector Machine (SVM), Decision Tree (DT), k-Nearest Neighbors (KNN), and Artificial Neural Network (ANN). This mechanism was done in various stages for Computer Aided Detection System. In the preliminary stage the pre-processing and post-processing of MR image enhancement is done. This was done as the processed image is more likely suitable for the analysis. Then the k-means clustering is used to sectioning the MRI by applied mean gray level method. In the subsequent stage, statistical feature analysis were done, the features were computed using Haralick’s equation for feature based on the Gray Level Co-occurrence Matrix. Feature chosen dependent on tumor region, location, periphery, and color from the sectioned image is then classified by applying the classification techniques. In the third stage SVM, DT, ANN, and KNN classifiers were used for diagnoses. The performances of the classifiers are tested and evaluated successfully.

Convolution neural network–based Alzheimer's disease classification using hybrid enhanced independent component analysis based segmented gray matter of T2 weighted magnetic resonance imaging with clinical valuation

In recent times, accurate and early diagnosis of Alzheimer's disease (AD) plays a vital role in patient care and further treatment. Predicting AD from mild cognitive impairment (MCI) and cognitive normal (CN) has become popular. Neuroimaging and computer-aided diagnosis techniques are used for classification of AD by physicians in the early stage. Most of the previous machine learning techniques work on handpicked features. In the recent days, deep learning has been applied for many medical image applications. Existing deep learning systems work on raw magnetic resonance imaging (MRI) images and cortical surface as an input to the convolution neural network (CNN) to perform classification of AD. AD affects the brain volume and changes the gray matter texture. In our work, we used 1820 T2-weighted brain magnetic resonance volumes including 635 AD MRIs, 548 MCI MRIs, and 637 CN MRIs, sliced into 18,017 voxels. We proposed an approach to extract the gray matter from brain voxels and perform the classification using the CNN. A Gaussian filter is used to enhance the voxels, and skull stripping algorithm is used to remove the irrelevant tissues from enhanced voxels. Then, those voxels are segmented by hybrid enhanced independent component analysis. Segmented gray matter is used as an input to the CNN. We performed clinical valuation using our proposed approach and achieved 90.47% accuracy, 86.66% of recall, and 92.59% precision.

Classification of Alzheimer Disease based on Normalized Hu Moment Invariants and Multiclassifier

There is a great benefit of Alzheimer disease (AD) classification for health care application. AD is the most common form of dementia. This paper presents a new methodology of invariant interest point descriptor for Alzheimer disease classification. The descriptor depends on the normalized Hu Moment Invariants (NHMI). The proposed approach deals with raw Magnetic Resonance Imaging (MRI) of Alzheimer disease. Seven Hu moments are computed for extracting images’ features. These moments are then normalized giving new more powerful features that highly improve the classification system performance. The moments are invariant which is the robustness point of Hu moments algorithm to extract features. The classification process is implemented using two different classifiers, K-Nearest Neighbors algorithm (KNN) and Linear Support Vector Machines (SVM). A comparison among their performances is investigated. The results are evaluated on Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. The best classification accuracy is 91.4% for KNN classifier and 100% for SVM classifier.

A new experimental carotid siphon aneurysm model in canine based on the MR angiography and rapid prototyping technology

Objective The aim of the experiment is to make an intracranial aneurysm model in canine.Methods A digital tube was made based on raw magnetic resonance images of the human intracraaial carotid artery.Then 6 tubes were made in the 3 D rapid prototyping machine and coated with silicone.Finally we isolated the common carotid arteries of 6 canines and made them go through the tubes and anastomosed them end-to-side to get the aneurysm model.Six stents were implanted after one week.Results Six aneurysm models were successfully made in canines.The parent artery had similar geometrv of the human carotid siphon.All the aneurysms and parent arteries were patent in one week's follow-up.One canine died of excessive anesthesia after stentingr Two vaseular models kept patent in one month without stenosis.The other 3 had some stenosis on the bends of the vessel.Conclusions The aneurysm model in tIle experiment has high flexibility and reliability.The model provides an effective tool for research and testing neurovascular devices.It's also a useful device to train the neuroradiologists and interventional physicians. Key words: Magnetic resonance angiography; Carotid artery,internal; Intracranial aneurysm

Magnetic resonance imaging study of a soft actuator element during operation

The physical structure and the actuation mechanism of a Nafion-based soft actuator – a Pt-containing ionic polymer–metal composite (IPMC) – were investigated using in situ magnetic resonance imaging (MRI) during application of four different electrical regimes. Importantly, the raw MRI data were used to generate spatial maps of both proton density (PD) and proton spin–spin relaxation time (T2) across the sample. These were successfully employed to study changes in the distribution and chemical environment of water molecules absorbed within the operating actuator device. The IPMC sample was mapped in this way during the application of a small d.c. potential across its thickness. Three main phenomena were observed in the results: initial rapid increase in T2 at both electrodes, without an observed change in PD; slower formation of a region of high T2 and high PD at the IPMC cathode; and contraction of the polymer along the anode and its expansion along the cathode, giving rise to bending actuation. Reversing the polarity of the applied potential resulted in the reversal of the direction of the bending deformation of the IPMC sample and of the distribution of PD and T2 within it. These phenomena were explained in terms of the unusual structure of Nafion and its interaction with host ions and the electric field. Up to 20% of the total water content of the IPMC was found to be involved in long-range electro-diffusion.

The mixing state of fine powders measured by magnetic resonance imaging

The usefulness of magnetic resonance imaging (MRI) for the spatially resolved, quantitative characterization of mixtures of fine powders in the size range of about 10 μm is investigated. The sources of errors inherent in the signal translation process between raw MRI data and local concentrations are analyzed in detail. Possibilities for their correction are pointed out in deriving a global characteristic such as the standard deviation. The analysis of the mixing state of powders by MRI and the benefits of error correction are then illustrated by simple examples covering a wide range of compositions and mixing states. These experiments were made by mixing spherical, oil-filled (hence “visible” to MRI) melamine microcapsules in the size range of about 1–20 μm with solid melamine spheres of similar size and density. The mixture volume of typically 1 cm 3 was imaged with an isotropic resolution of 235 μm, providing over 60,000 contiguous data points per measurement. The benefit of error correction is shown to be significant, provided the mixtures are homogeneous to a large extent.

Factor Analysis of Medical Image Sequences Improves Evaluation of First-Pass MR Imaging Acquisitions for Myocardial Perfusion

Rationale and Objectives Factor analysis of medical image sequences (FAMIS) applied to gadolinium chelate–enhanced subsecond magnetic resonance (MR) imaging was evaluated as a postprocessing method for assessing myocardial perfusion in coronary artery disease (CAD). Materials and Methods To assess the accuracy of motion correction, five normal volunteers underwent MR imaging at rest. Thirteen patients with well-documented CAD and no myocardial infarction underwent MR imaging at rest and after dipyridamole administration. After motion correction, a single myocardial tissue factor (FAMISt) image was obtained with FAMIS for each raw MR imaging series acquisition. To evaluate how FAMIS could improve the analysis of these acquisitions, five readers visually assessed myocardial perfusion with FAMISt and raw MR images, and a multicase, multi–reader receiver operating characteristic analysis was performed. Results FAMISt images significantly improved detection of the perfusion defects when compared with raw MR images ( P = .002). Areas under the receiver operating characteristic curves ranged from 0.84 to 0.93 with FAMISt images and from 0.48 to 0.85 with raw MR images. Conclusion FAMIS applied to first-pass MR imaging series provided myocardial perfusion images that improve the objective assessment of myocardial perfusion in patients with CAD.

A new virtual reality approach for planning of cardiac interventions

A novel approach to three-dimensional (3D) visualization of high quality, respiratory compensated cardiac magnetic resonance (MR) data is presented with the purpose of assisting the cardiovascular surgeon and the invasive cardiologist in the pre-operative planning. Developments included: (1) optimization of 3D, MR scan protocols; (2) dedicated segmentation software; (3) optimization of model generation algorithms; (4) interactive, virtual reality visualization. The approach is based on a tool for interactive, real-time visualization of 3D cardiac MR datasets in the form of 3D heart models displayed on virtual reality equipment. This allows the cardiac surgeon and the cardiologist to examine the model as if they were actually holding it in their hands. To secure relevant examination of all details related to cardiac morphology, the model can be re-scaled and the viewpoint can be set to any point inside the heart. Finally, the original, raw MR images can be examined on line as textures in cut-planes through the heart models.