Raw Magnetic Resonance Imaging Research Articles

QRATER: a collaborative and centralized imaging quality control web-based application

Quality control (QC) is an important part of all scientific analyses, including neuroscience. With manual curation considered the gold standard, there remains a lack of available tools that make manual neuroimaging QC accessible, fast, and easy. In this article we present Qrater, a containerized web-based Python application that enables viewing and rating any type of image for QC purposes. Qrater functionalities allow collaboration between various raters on the same dataset which can facilitate completing large QC tasks. Qrater was used to evaluate QC rater performance on three different magnetic resonance (MR) image QC tasks by a group of raters having different amounts of experience. The tasks included QC of raw MR images (10,196 images), QC of linear registration to a standard template (10,196 images), and QC of skull segmentation (6,968 images). We measured the proportion of failed images, average rating time per image, intra- and inter-rater agreement, as well as the comparison against QC using a conventional method. The median time spent rating per image differed significantly between raters (depending on rater experience) in each of the three QC tasks. Evaluating raw MR images was slightly faster using Qrater than an image viewer (expert: 99 vs. 90 images in 63 min; trainee 99 vs 79 images in 98 min). Reviewing the linear registration using Qrater was twice faster for the expert (99 vs. 43 images in 36 min) and three times faster for the trainee (99 vs. 30 images in 37 min). The greatest difference in rating speed resulted from the skull segmentation task where the expert took a full minute to inspect the volume on a slice-by-slice basis compared to just 3 s using Qrater. Rating agreement also depended on the experience of the raters and the task at hand: trained raters’ inter-rater agreements with the expert’s gold standard were moderate for both raw images (Fleiss’ Kappa = 0.44) and linear registration (Fleiss’ Kappa = 0.56); the experts’ inter-rater agreement of the skull segmentation task was excellent (Cohen’s Kappa = 0.83). These results demonstrate that Qrater is a useful asset for QC tasks that rely on manual evaluation of QC images.

Application of Convolutional Neural Networks for Early Detection and Classification of Alzheimer's disease from MRI Images

This study investigates the application of convolutional neural networks (CNNs) and traditional machine learning algorithms for the early detection and classification of Alzheimer's disease (AD) using brain Magnetic Resonance Imaging (MRI) data. We compare the performance of CNNs with Support Vector Machines (SVM), Random Forest (RF), and Gradient Boosting Machines (GBM) on a dataset comprising MRI images from AD patients and healthy controls. Results show that CNNs achieved the highest accuracy (90.2%) and area under the receiver operating characteristic curve (AUC-ROC) of 0.95, outperforming SVM, RF, and GBM. The CNN model also exhibited high sensitivity (87.5%) and specificity (92.6%) in distinguishing between AD patients and healthy controls. These findings highlight the effectiveness of CNN-based approaches in leveraging raw MRI images for accurate and early detection of AD.

Comparative MRI analysis of the forebrain of three sauropsida models

The study of the brain by magnetic resonance imaging (MRI) allows to obtain detailed anatomical images, useful to describe specific encephalic structures and to analyze possible variabilities. It is widely used in clinical practice and is becoming increasingly used in veterinary medicine, even in exotic animals; however, despite its potential, its use in comparative neuroanatomy studies is still incipient. It is a technology that in recent years has significantly improved anatomical resolution, together with the fact that it is non-invasive and allows for systematic comparative analysis. All this makes it particularly interesting and useful in evolutionary neuroscience studies, since it allows for the analysis and comparison of brains of rare or otherwise inaccessible species. In the present study, we have analyzed the prosencephalon of three representative sauropsid species, the turtle Trachemys scripta (order Testudine), the lizard Pogona vitticeps (order Squamata) and the snake Python regius (order Squamata) by MRI. In addition, we used MRI sections to analyze the total brain volume and ventricular system of these species, employing volumetric and chemometric analyses together. The raw MRI data of the sauropsida models analyzed in the present study are available for viewing and downloading and have allowed us to produce an atlas of the forebrain of each of the species analyzed, with the main brain regions. In addition, our volumetric data showed that the three groups presented clear differences in terms of total and ventricular brain volumes, particularly the turtles, which in all cases presented distinctive characteristics compared to the lizards and snakes.

MRI-based brain tumor detection and types of classification using CEHJB-RI and JMST

The Brain Tumor (BT), which forms in the brain cells and spreads to the whole brain, may lead to death. Hence, early diagnosis of BT is significant. Still, the detection of BT between the skull and brain region is not concentrated, which results in misclassification outcomes. Thus, this article proposes Magnetic Resonance Imaging (MRI)-based BT detection and types’ classification utilizing Carlitz Exponential Hamilton Jacobi Bellman-based Reinforcement Learning (CEHJB-RL) and JenSorensen similarity-based Minimum Spanning Tree (JMST). Primarily, raw MRI images are taken and then pre-processed. Then, with skull and without skull regions are extracted from the pre-processed image and are subjected to the graph construction. Conversely, the edges are detected from the pre-processed image that can be patch-extracted and subjected to graph construction. By utilizing JMST and Morphological Operations (MO), the graphs are constructed. Thereafter, the features are extracted and fed to the classifier. Then, the type of BT is classified by the classifier using CEHJB-RL. Concerning the performance metrics, the outcomes illustrated that the proposed technique attained a higher accuracy (99.27%), which is better than other existing techniques.

Multi-class point cloud completion networks for 3D cardiac anatomy reconstruction from cine magnetic resonance images.

Cine magnetic resonance imaging (MRI) is the current gold standard for the assessment of cardiac anatomy and function. However, it typically only acquires a set of two-dimensional (2D) slices of the underlying three-dimensional (3D) anatomy of the heart, thus limiting the understanding and analysis of both healthy and pathological cardiac morphology and physiology. In this paper, we propose a novel fully automatic surface reconstruction pipeline capable of reconstructing multi-class 3D cardiac anatomy meshes from raw cine MRI acquisitions. Its key component is a multi-class point cloud completion network (PCCN) capable of correcting both the sparsity and misalignment issues of the 3D reconstruction task in a unified model. We first evaluate the PCCN on a large synthetic dataset of biventricular anatomies and observe Chamfer distances between reconstructed and gold standard anatomies below or similar to the underlying image resolution for multiple levels of slice misalignment. Furthermore, we find a reduction in reconstruction error compared to a benchmark 3D U-Net by 32% and 24% in terms of Hausdorff distance and mean surface distance, respectively. We then apply the PCCN as part of our automated reconstruction pipeline to 1000 subjects from the UK Biobank study in a cross-domain transfer setting and demonstrate its ability to reconstruct accurate and topologically plausible biventricular heart meshes with clinical metrics comparable to the previous literature. Finally, we investigate the robustness of our proposed approach and observe its capacity to successfully handle multiple common outlier conditions.

Brain tumour MRI detection and classification based on the convolutional neural network

Magnetic Resonance Imaging (MRI) has emerged as a widely used diagnostic technique for brain tumour detection. However, the diagnosis of brain tumours poses significant challenges due to their occurrence in diverse locations and various types. Furthermore, MRI generates images that require manual analysis by physicians, which can be laborious and prone to errors. To enhance the efficacy and accuracy of brain tumour detection, recent advances in artificial intelligence have led to the development of machine learning algorithms. In this study, a convolutional neural network (CNN) based method was proposed for brain tumour detection and classification through the preprocessing of raw MRI images. The customized CNN model achieves an accuracy of 98% on a dataset consisting of four types of MRI images, including three types of brain tumours and healthy brain images, with preprocessing applied to all images. The CNN model demonstrates an accuracy of 95% in classifying raw MRI images from the dataset. The CNN model's performance is further improved by training the model with preprocessed images that have been transformed into the same colour space and object area zoomed in. These findings provide a promising avenue for the development of automated and efficient brain tumour detection systems using CNN and MRI.

A New Deep Learning Model based on Neuroimaging for Predicting Alzheimer's Disease

Background: The psychological aspects of the brain in Alzheimer's disease (AD) are significantly affected. These alterations in brain anatomy take place due to a variety of reasons, including the shrinking of grey and white matter in the brain. Magnetic resonance imaging (MRI) scans can be used to measure it, and these scans offer a chance for early identification of AD utilizing classification methods, like convolutional neural network (CNN). The majority of AD-related tests are now constrained by the test measures. It is, thus, crucial to find an affordable method for image categorization using minimal information. Because of developments in machine learning and medical imaging, the field of computerized health care has evolved rapidly. Recent developments in deep learning, in particular, herald a new era of clinical decision-making that is heavily reliant on multimedia systems. Methods: In the proposed work, we have investigated various CNN-based transfer-learning strategies for predicting AD using MRI scans of the brain's structural organization. According to an analysis of the data, the suggested model makes use of a number of sites related to Alzheimer's disease. In order to interpret structural brain pictures in both 2D and 3D, the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset includes straightforward CNN designs based on 2D and 3D convolutions. Results: According to these results, deep neural networks may be able to automatically learn which imaging biomarkers are indicative of Alzheimer's disease and exploit them for precise early disease detection. The proposed techniques have been found to achieve an accuracy of 93.24%. Conclusion: This research aimed to classify Alzheimer's disease (AD) using transfer learning. We have used strict pre-processing steps on raw MRI data from the ADNI dataset and used the AlexNet, i.e., Alzheimer's disease has been categorized using pre-processed data and the CNN classifier.

Biometric magnetic resonance imaging analysis of fetal brain development in Down syndrome.

To assess brain development in living fetuses with Down syndrome (DS) by biometric measurements on fetal brain magnetic resonance images (MRI). We scanned 10 MRIs of fetuses with confirmed trisomy 21 at birth and 12 control fetal MRIs without any detected anomalies. Fetal brain MRIs were analyzed using 14 fetal brain and skull biometric parameters. We compared measures between DS and controls in both raw MRIs and motion-corrected and anterior-posterior commissure-aligned images. In the reconstructed images, the measured values of the height of the cerebellar vermis (HV) and anteroposterior diameter of the cerebellar vermis (APDV) were significantly smaller, and the anteroposterior diameter of the fourth ventricle (APDF) was significantly larger in fetuses with DS than controls. In the raw MRIs, the measured values of the right lateral ventricle were significantly larger in fetuses with DS than in controls. Logistic regression analyses revealed that a new parameter, the cerebellar-to-fourth-ventricle ratio (i.e., (APDV * Height of the vermis)/APDF), was significantly smaller in fetuses with DS than controls and was the most predictive to distinguish between fetuses with DS and controls. The study revealed that fetuses with DS have smaller cerebellums and larger fourth ventricles compared to the controls.

Denoising of Nifti (MRI) Images with a Regularized Neighborhood Pixel Similarity Wavelet Algorithm.

The recovery of semantics from corrupted images is a significant challenge in image processing. Noise can obscure features, interfere with accurate analysis, and bias results. To address this issue, the Regularized Neighborhood Pixel Similarity Wavelet algorithm (PixSimWave) was developed for denoising Nifti (magnetic resonance imaging (MRI)). The PixSimWave algorithm uses regularized pixel similarity detection to improve the accuracy of noise reduction by creating patches to analyze the intensity of pixels and locate matching pixels, as well as adaptive neighborhood filtering to estimate noisy pixel values by allocating each pixel a weight based on its similarity. The wavelet transform breaks down the image into scales and orientations, allowing a sparse image representation to allocate a soft threshold on its similarity to the original pixels. The proposed method was evaluated on simulated and raw T1w MRIs, outperforming other methods in terms of an SSIM value of 0.9908 for a low Rician noise level of 3% and 0.9881 for a high noise level of 17%. The addition of Gaussian noise improved PSNR and SSIM, with the results indicating that the proposed method outperformed other models while preserving edges and textures. In summary, the PixSimWave algorithm is a viable noise-elimination approach that employs both sparse wavelet coefficients and regularized similarity with decreased computation time, improving the accuracy of noise reduction in images.

Artificial Intelligence to Preoperatively Predict Proximal Junction Kyphosis Following Adult Spinal Deformity Surgery: Soft Tissue Imaging May Be Necessary for Accurate Models.

Retrospective cohort. In a cohort of patients undergoing adult spinal deformity (ASD) surgery, we used artificial intelligence to compare three models of preoperatively predicting radiographic proximal junction kyphosis (PJK) using: (1) traditional demographics and radiographic measurements, (2) raw preoperative scoliosis radiographs, and (3) raw preoperative thoracic magnetic resonance imaging (MRI). Despite many proposed risk factors, PJK following ASD surgery remains difficult to predict. A single-institution, retrospective cohort study was undertaken for patients undergoing ASD surgery from 2009 to 2021. PJK was defined as a sagittal Cobb angle of upper-instrumented vertebra (UIV) and UIV+2>10° and a postoperative change in UIV/UIV+2>10°. For model 1, a support vector machine was used to predict PJK within 2 years postoperatively using clinical and traditional sagittal/coronal radiographic variables and intended levels of instrumentation. Next, for model 2, a convolutional neural network (CNN) was trained on raw preoperative lateral and posterior-anterior scoliosis radiographs. Finally, for model 3, a CNN was trained on raw preoperative thoracic T1 MRIs. A total of 191 patients underwent ASD surgery with at least 2-year follow-up and 89 (46.6%) developed radiographic PJK within 2 years. Model 1: Using clinical variables and traditional radiographic measurements, the model achieved a sensitivity: 57.2% and a specificity: 56.3%. Model 2: a CNN with raw scoliosis x-rays predicted PJK with a sensitivity: 68.2% and specificity: 58.3%. Model 3: a CNN with raw thoracic MRIs predicted PJK with average sensitivity: 73.1% and specificity: 79.5%. Finally, an attention map outlined the imaging features used by model 3 elucidated that soft tissue features predominated all true positive PJK predictions. The use of raw MRIs in an artificial intelligence model improved the accuracy of PJK prediction compared with raw scoliosis radiographs and traditional clinical/radiographic measurements. The improved predictive accuracy using MRI may indicate that PJK is best predicted by soft tissue degeneration and muscle atrophy.

Deep learning-based automated lesion segmentation on mouse stroke magnetic resonance images

Magnetic resonance imaging (MRI) is widely used for ischemic stroke lesion detection in mice. A challenge is that lesion segmentation often relies on manual tracing by trained experts, which is labor-intensive, time-consuming, and prone to inter- and intra-rater variability. Here, we present a fully automated ischemic stroke lesion segmentation method for mouse T2-weighted MRI data. As an end-to-end deep learning approach, the automated lesion segmentation requires very little preprocessing and works directly on the raw MRI scans. We randomly split a large dataset of 382 MRI scans into a subset (n = 293) to train the automated lesion segmentation and a subset (n = 89) to evaluate its performance. We compared Dice coefficients and accuracy of lesion volume against manual segmentation, as well as its performance on an independent dataset from an open repository with different imaging characteristics. The automated lesion segmentation produced segmentation masks with a smooth, compact, and realistic appearance that are in high agreement with manual segmentation. We report dice scores higher than the agreement between two human raters reported in previous studies, highlighting the ability to remove individual human bias and standardize the process across research studies and centers.

Brain Tumor Class Detection in Flair/T2 Modality MRI Slices Using Elephant-Herd Algorithm Optimized Features.

Several advances in computing facilities were made due to the advancement of science and technology, including the implementation of automation in multi-specialty hospitals. This research aims to develop an efficient deep-learning-based brain-tumor (BT) detection scheme to detect the tumor in FLAIR- and T2-modality magnetic-resonance-imaging (MRI) slices. MRI slices of the axial-plane brain are used to test and verify the scheme. The reliability of the developed scheme is also verified through clinically collected MRI slices. In the proposed scheme, the following stages are involved: (i) pre-processing the raw MRI image, (ii) deep-feature extraction using pretrained schemes, (iii) watershed-algorithm-based BT segmentation and mining the shape features, (iv) feature optimization using the elephant-herding algorithm (EHA), and (v) binary classification and verification using three-fold cross-validation. Using (a) individual features, (b) dual deep features, and (c) integrated features, the BT-classification task is accomplished in this study. Each experiment is conducted separately on the chosen BRATS and TCIA benchmark MRI slices. This research indicates that the integrated feature-based scheme helps to achieve a classification accuracy of 99.6667% when a support-vector-machine (SVM) classifier is considered. Further, the performance of this scheme is verified using noise-attacked MRI slices, and better classification results are achieved.

An end‐to‐end infant brain parcellation pipeline

ObjectiveAccurate infant brain parcellation is crucial for understanding early brain development; however, it is challenging due to the inherent low tissue contrast, high noise, and severe partial volume effects in infant magnetic resonance images (MRIs). The aim of this study was to develop an end-to-end pipeline that enabled accurate parcellation of infant brain MRIs. MethodsWe proposed an end-to-end pipeline that employs a two-stage global-to-local approach for accurate parcellation of infant brain MRIs. Specifically, in the global regions of interest (ROIs) localization stage, a combination of transformer and convolution operations was employed to capture both global spatial features and fine texture features, enabling an approximate localization of the ROIs across the whole brain. In the local ROIs refinement stage, leveraging the position priors from the first stage along with the raw MRIs, the boundaries of the ROIs are refined for a more accurate parcellation. ResultsWe utilized the Dice ratio to evaluate the accuracy of parcellation results. Results on 263 subjects from National Database for Autism Research (NDAR), Baby Connectome Project (BCP) and Cross-site datasets demonstrated the better accuracy and robustness of our method than other competing methods. ConclusionOur end-to-end pipeline may be capable of accurately parcellating 6-month-old infant brain MRIs.

MyI-Net: Fully Automatic Detection and Quantification of Myocardial Infarction from Cardiovascular MRI Images.

Myocardial infarction (MI) occurs when an artery supplying blood to the heart is abruptly occluded. The "gold standard" method for imaging MI is cardiovascular magnetic resonance imaging (MRI) with intravenously administered gadolinium-based contrast (with damaged areas apparent as late gadolinium enhancement [LGE]). However, no "gold standard" fully automated method for the quantification of MI exists. In this work, we propose an end-to-end fully automatic system (MyI-Net) for the detection and quantification of MI in MRI images. It has the potential to reduce uncertainty due to technical variability across labs and the inherent problems of data and labels. Our system consists of four processing stages designed to maintain the flow of information across scales. First, features from raw MRI images are generated using feature extractors built on ResNet and MoblieNet architectures. This is followed by atrous spatial pyramid pooling (ASPP) to produce spatial information at different scales to preserve more image context. High-level features from ASPP and initial low-level features are concatenated at the third stage and then passed to the fourth stage where spatial information is recovered via up-sampling to produce final image segmentation output into: (i) background, (ii) heart muscle, (iii) blood and (iv) LGE areas. Our experiments show that the model named MI-ResNet50-AC provides the best global accuracy (97.38%), mean accuracy (86.01%), weighted intersection over union (IoU) of 96.47%, and bfscore of 64.46% for the global segmentation. However, in detecting only LGE tissue, a smaller model, MI-ResNet18-AC, exhibited higher accuracy (74.41%) than MI-ResNet50-AC (64.29%). New models were compared with state-of-the-art models and manual quantification. Our models demonstrated favorable performance in global segmentation and LGE detection relative to the state-of-the-art, including a four-fold better performance in matching LGE pixels to contours produced by clinicians.

Reducing variations in multi-center Alzheimer's disease classification with convolutional adversarial autoencoder.

Based on brain magnetic resonance imaging (MRI), multiple variations ranging from MRI scanners to center-specific parameter settings, imaging protocols, and brain region-of-interest (ROI) definitions pose a big challenge for multi-center Alzheimer's disease characterization and classification. Existing approaches to reduce such variations require intricate multi-step, often manual preprocessing pipelines, including skull stripping, segmentation, registration, cortical reconstruction, and ROI outlining. Such procedures are time-consuming, and more importantly, tend to be user biased. Contrasting costly and biased preprocessing pipelines, the question arises whether we can design a deep learning model to automatically reduce these variations from multiple centers for Alzheimer's disease classification? In this study, we used T1 and T2-weighted structural MRI from Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset based on three groups with 375 subjects, respectively: patients with Alzheimer's disease (AD) dementia, with mild cognitive impairment (MCI), and healthy controls (HC); to test our approach, we defined AD classification as classifying an individual's structural image to one of the three group labels. We first introduced a convolutional adversarial autoencoder (CAAE) to reduce the variations existing in multi-center raw MRI scans by automatically registering them into a common aligned space. Afterward, a convolutional residual soft attention network (CRAT) was further proposed for AD classification. Canonical classification procedures demonstrated that our model achieved classification accuracies of 91.8%, 90.05%, and 88.10% for the 2-way classification tasks using the RAW aligned MRI scans, including AD vs. HC, AD vs. MCI, and MCI vs. HC, respectively. Thus, our automated approach achieves comparable or even better classification performance by comparing it with many baselines with dedicated conventional preprocessing pipelines. Furthermore, the uncovered brain hotpots, i.e., hippocampus, amygdala, and temporal pole, are consistent with previous studies.

Deep learning and machine learning‐based early survival predictions of glioblastoma patients using pre‐operative three‐dimensional brain magnetic resonance imaging modalities

AbstractTo propose and implement an automated machine learning (ML) based methodology to predict the overall survival of glioblastoma multiforme (GBM) patients. In the proposed methodology, we used deep learning (DL) based 3D U‐shaped Convolutional Neural Network inspired encoder‐decoder architecture to segment the brain tumor. Further, feature extraction was performed on these segmented and raw magnetic resonance imaging (MRI) scans using a pre‐trained 2D residual neural network. The dimension‐reduced principal components were integrated with clinical data and the handcrafted features of tumor subregions to compare the performance of regression‐based automated ML techniques. Through the proposed methodology, we achieved the mean squared error (MSE) of 87 067.328, median squared error of 30 915.66, and a SpearmanR correlation of 0.326 for survival prediction (SP) with the validation set of Multimodal Brain Tumor Segmentation 2020 dataset. These results made the MSE far better than the existing automated techniques for the same patients. Automated SP of GBM patients is a crucial topic with its relevance in clinical use. The results proved that DL‐based feature extraction using 2D pre‐trained networks is better than many heavily trained 3D and 2D prediction models from scratch. The ensembled approach has produced better results than single models. The most crucial feature affecting GBM patients' survival is the patient's age, as per the feature importance plots presented in this work. The most critical MRI modality for SP of GBM patients is the T2 fluid attenuated inversion recovery, as evident from the feature importance plots.

Multimodal MRI-Based Whole-Brain Assessment in Patients In Anoxoischemic Coma by Using 3D Convolutional Neural Networks

BackgroundThere is an unfulfilled need to find the best way to automatically capture, analyze, organize, and merge structural and functional brain magnetic resonance imaging (MRI) data to ultimately extract relevant signals that can assist the medical decision process at the bedside of patients in postanoxic coma. We aimed to develop and validate a deep learning model to leverage multimodal 3D MRI whole-brain times series for an early evaluation of brain damages related to anoxoischemic coma.MethodsThis proof-of-concept, prospective, cohort study was undertaken at the intensive care unit affiliated with the University Hospital (Toulouse, France), between March 2018 and May 2020. All patients were scanned in coma state at least 2 days (4 ± 2 days) after cardiac arrest. Over the same period, age-matched healthy volunteers were recruited and included. Brain MRI quantification encompassed both “functional data” from regions of interest (precuneus and posterior cingulate cortex) with whole-brain functional connectivity analysis and “structural data” (gray matter volume, T1-weighted, fractional anisotropy, and mean diffusivity). A specifically designed 3D convolutional neuronal network (CNN) was created to allow conscious state discrimination (coma vs. controls) by using raw MRI indices as the input. A voxel-wise visualization method based on the study of convolutional filters was applied to support CNN outcome. The Ethics Committee of the University Teaching Hospital of Toulouse, France (2018-A31) approved the study and informed consent was obtained from all participants.ResultsThe final cohort consisted of 29 patients in postanoxic coma and 34 healthy volunteers. Coma patients were successfully discerned from controls by using 3D CNN in combination with different MR indices. The best accuracy was achieved by functional MRI data, in particular with resting-state functional MRI of the posterior cingulate cortex, with an accuracy of 0.96 (range 0.94–0.98) on the test set from 10-time repeated tenfold cross-validation. Even more satisfactory performances were achieved through the majority voting strategy, which was able to compensate for mistakes from single MR indices. Visualization maps allowed us to identify the most relevant regions for each MRI index, notably regions previously described as possibly being involved in consciousness emergence. Interestingly, a posteriori analysis of misclassified patients indicated that they may present some common functional MRI traits with controls, which suggests further favorable outcomes.ConclusionsA fully automated identification of clinically relevant signals from complex multimodal neuroimaging data is a major research topic that may bring a radical paradigm shift in the neuroprognostication of patients with severe brain injury. We report for the first time a successful discrimination between patients in postanoxic coma patients from people serving as controls by using 3D CNN whole-brain structural and functional MRI data.Clinical Trial Numberhttp://ClinicalTrials.gov (No. NCT03482115).

Enhancing the REMBRANDT MRI collection with expert segmentation labels and quantitative radiomic features

Malignancy of the brain and CNS is unfortunately a common diagnosis. A large subset of these lesions tends to be high grade tumors which portend poor prognoses and low survival rates, and are estimated to be the tenth leading cause of death worldwide. The complex nature of the brain tissue environment in which these lesions arise offers a rich opportunity for translational research. Magnetic Resonance Imaging (MRI) can provide a comprehensive view of the abnormal regions in the brain, therefore, its applications in the translational brain cancer research is considered essential for the diagnosis and monitoring of disease. Recent years has seen rapid growth in the field of radiogenomics, especially in cancer, and scientists have been able to successfully integrate the quantitative data extracted from medical images (also known as radiomics) with genomics to answer new and clinically relevant questions. In this paper, we took raw MRI scans from the REMBRANDT data collection from public domain, and performed volumetric segmentation to identify subregions of the brain. Radiomic features were then extracted to represent the MRIs in a quantitative yet summarized format. This resulting dataset now enables further biomedical and integrative data analysis, and is being made public via the NeuroImaging Tools & Resources Collaboratory (NITRC) repository (https://www.nitrc.org/projects/rembrandt_brain/).

FastMRI+, Clinical pathology annotations for knee and brain fully sampled magnetic resonance imaging data

Improving speed and image quality of Magnetic Resonance Imaging (MRI) using deep learning reconstruction is an active area of research. The fastMRI dataset contains large volumes of raw MRI data, which has enabled significant advances in this field. While the impact of the fastMRI dataset is unquestioned, the dataset currently lacks clinical expert pathology annotations, critical to addressing clinically relevant reconstruction frameworks and exploring important questions regarding rendering of specific pathology using such novel approaches. This work introduces fastMRI+, which consists of 16154 subspecialist expert bounding box annotations and 13 study-level labels for 22 different pathology categories on the fastMRI knee dataset, and 7570 subspecialist expert bounding box annotations and 643 study-level labels for 30 different pathology categories for the fastMRI brain dataset. The fastMRI+ dataset is open access and aims to support further research and advancement of medical imaging in MRI reconstruction and beyond.

Fast Image-Level MRI Harmonization via Spectrum Analysis.

Pooling structural magnetic resonance imaging (MRI) data from different imaging sites helps increase sample size to facilitate machine learning based neuroimage analysis, but usually suffers from significant cross-site and/or cross-scanner data heterogeneity. Existing studies often focus on reducing cross-site and/or cross-scanner heterogeneity at handcrafted feature level targeting specific tasks (e.g., classification or segmentation), limiting their adaptability in clinical practice. Research on image-level MRI harmonization targeting a broad range of applications is very limited. In this paper, we develop a spectrum swapping based image-level MRI harmonization (SSIMH) framework. Different from previous work, our method focuses on alleviating cross-scanner heterogeneity at raw image level. We first construct spectrum analysis to explore the influences of different frequency components on MRI harmonization. We then utilize a spectrum swapping method for the harmonization of raw MRIs acquired by different scanners. Our method does not rely on complex model training, and can be directly applied to fast real-time MRI harmonization. Experimental results on T1- and T2-weighted MRIs of phantom subjects acquired by using different scanners from the public ABCD dataset suggest the effectiveness of our method in structural MRI harmonization at the image level.