Spin-lattice Relaxation Research Articles

Metabolite T2 relaxation times decrease across the adult lifespan in a large multi-site cohort.

Relaxation correction is crucial for accurately estimating metabolite concentrations measured using in vivo MRS. However, the majority of MRS quantification routines assume that relaxation values remain constant across the lifespan, despite prior evidence of T2 changes with aging for multiple of the major metabolites. Here, we comprehensively investigate correlations between T2 and age in a large, multi-site cohort. We recruited approximately 10 male and 10 female participants from each decade of life: 18-29, 30-39, 40-49, 50-59, and 60+ y old (n = 101 total). We collected PRESS data at eight TEs (30, 50, 74, 101, 135, 179, 241, and 350 ms) from voxels placed in white-matter-rich centrum semiovale (CSO) and gray-matter-rich posterior cingulate cortex (PCC). We quantified metabolite amplitudes using Osprey and fit exponential decay curves to estimate T2. Older age was correlated with shorter T2 for tNAA2.0, tCr3.0, tCr3.9, tCho, and tissue water (CSO and PCC), as well as mI and Glx (PCC only); rs = -0.22 to -0.63, all p < 0.05, false discovery rate (FDR)-corrected. These associations largely remained statistically significant when controlling for cortical atrophy. By region, T2 values were longer in the CSO for tNAA2.0, tCr3.9, Glx, and tissue water and longer in the PCC for tCho and mI. T2 did not differ by region for tCr3.0. These findings underscore the importance of considering metabolite T2 differences with aging in MRS quantification. We suggest that future 3T work utilize the equations presented here to estimate age-specific T2 values instead of relying on uniform default values.

Dimensionality Reduction and Nearest Neighbors for Improving Out-of-Distribution Detection in Medical Image Segmentation

Clinically deployed deep learning-based segmentation models are known to fail on data outside of their training distributions. While clinicians review the segmentations, these models tend to perform well in most instances, which could exacerbate automation bias. Therefore, detecting out-of-distribution images at inference is critical to warn the clinicians that the model likely failed. This work applied the Mahalanobis distance (MD) post hoc to the bottleneck features of four Swin UNETR and nnU-net models that segmented the liver on T1-weighted magnetic resonance imaging and computed tomography. By reducing the dimensions of the bottleneck features with either principal component analysis or uniform manifold approximation and projection, images the models failed on were detected with high performance and minimal computational load. In addition, this work explored a non-parametric alternative to the MD, a k-th nearest neighbors distance (KNN). KNN drastically improved scalability and performance over MD when both were applied to raw and average-pooled bottleneck features. Our code is available at &lt;a href='https://github.com/mckellwoodland/dimen_reduce_mahal'&gt;https://github.com/mckellwoodland/dimen_reduce_mahal&lt;/a&gt;

Differentiating MYCN-amplified RB1 wild-type retinoblastoma from biallelic RB1 mutant retinoblastoma using MR-based radiomics: a retrospective multicenter case–control study

MYCN-amplified RB1 wild-type (MYCNampRB1+/+) retinoblastoma is a rare and aggressive subtype, often resistant to standard therapies. Identifying unique MRI features is crucial for diagnosing this subtype, as biopsy is not recommended. This study aimed to differentiate MYCNampRB1+/+ from the most prevalent RB1-/- retinoblastoma using pretreatment MRI and radiomics. Ninety-eight unilateral retinoblastoma patients (19 MYCN cases and 79 matched controls) were included. Tumors on T2-weighted MR images were manually delineated and validated by experienced radiologists. Radiomics analysis extracted 120 features per tumor. Several combinations of feature selection methods, oversampling techniques and machine learning (ML) classifiers were evaluated in a repeated fivefold cross-validation machine learning pipeline to yield the best-performing prediction model for MYCN. The best model used univariate feature selection, data oversampling (duplicating MYCN cases), and logistic regression classifier, achieving a mean AUC of 0.78 (SD 0.12). SHAP analysis highlighted lower sphericity, higher flatness, and greater gray-level heterogeneity as predictive for MYCNampRB1+/+ status, yielding an AUC of 0.81 (SD 0.11). This study shows the potential of MRI-based radiomics to distinguish MYCNampRB1+/+ and RB1-/- retinoblastoma subtypes.

An Effective Liposome-Based Nanodelivery System for Naphthalene Derivative Polyamines with Antitumor Activity

This study focuses on the development of a novel liposome-based nanodelivery system designed to encapsulate polyamine-1, a compound with potential anti-tumor properties. The main objective of this work was to enhance the therapeutic and imaging potential of polyamine-1 by incorporating it into liposome-based nanoparticles, which were functionalized with a gadolinium complex for imaging purposes and a fluorescent phospholipid for tracking applications. These nanoparticles were characterized by measuring their size, shape, polydispersity index, zeta potential and encapsulation efficiency. In vitro experiments were conducted to evaluate the antitumor activity, specifically determining the cytotoxicity of both free and encapsulated polyamine-1 in cancerous and non-cancerous cell lines. Additionally, the study shows the enhanced signal intensity of gadolinium-loaded liposomes by T1-weighted MRI, highlighting their imaging potential. The experimental results suggest that this liposome-based nanodelivery system not only has therapeutic potential in targeted cancer therapy but also could be advantageous for diagnostic imaging, particularly in MRI applications.

Preventive treatment effects on brain structures and functions in patients with chronic migraine: A multimodel magnetic resonance imaging study.

Patients with chronic migraine (CM) often exhibit structural and functional alterations in pain-matrix regions, but it remains unclear how preventive treatment affects these changes. Therefore, this study aimed to investigate the structural and functional changes in pain-matrix regions in CM patients after 6-month treatment. A total of 24 patients with CM and 15 healthy controls were recruited for this study. Patients were divided into responder group (N = 9) and non-responder group (N = 15). After completing the Migraine Disability Assessment (MIDAS) questionnaire, all patients underwent whole-brain high-resolution T1-weighted images, diffusion-weighted imaging, and resting-state functional magnetic resonance imaging at baseline and 6-month follow-up. Whole brain gray matter volume and white matter diffusion indices were analyzed using voxel-based analysis. Structural and functional connectivity analyses were performed to understand brain changes in patients after 6-month preventive treatment. The responder group exhibited significantly higher MIDAS scores than the non-responder group at baseline, but no significant difference between the two groups at follow-up. No significant interval change was noted in gray matter volume, white matter diffusion indices, and structural connectivity in CM patients after 6-month treatment. Nonetheless, the functional connectivity was significantly increased between occipital, temporal lobes and cerebellum, and was significantly decreased between parietal and temporal lobes after 6-month preventive treatment. We concluded that resting-state functional connectivity was suitable for investigating the preventive treatment effect on CM patients.

Correlations between Cerebrospinal Fluid Biomarkers and Gray Matter Atrophy in Alzheimer's and Behavioural Variant Frontotemporal Dementia.

Distinguishing between frontotemporal dementia (FTD) and Alzheimer's disease (AD) in their early stages remains a significant clinical challenge. Cerebrospinal fluid (CSF) biomarkers (total Tau, phosphorylated Tau, and beta-amyloid) are promising candidates for identifying early differences between these conditions. This study investigates the relationship between grey matter density and CSF markers in the behavioural variant of frontotemporal dementia (bvFTD) and Alzheimer's disease (AD). CSF and 3D T1-weighted magnetic resonance (MR) images were acquired from 14 bvFTD patients, 15 AD patients, and 13 cognitively normal (CN) matched subjects. The CSF markers and their relative ratios (total Tau/beta-amyloid, phosphorylated Tau/beta-amyloid) were compared across the three groups. Voxel-based morphometry (VBM) was performed to characterize the anatomical changes in bvFTD and AD patients compared to CN subjects. Grey matter density maps were obtained by automatic segmentation of 3.0 Tesla 3D T1-Weighted MR Images, and their correlation with CSF markers and relative ratios was investigated. Results demonstrated that, as compared to CN subjects, AD patients are characterised by higher CSF total Tau levels and lower beta-amyloid levels; however, beta-amyloid and relative ratios discriminated AD from bvFTD. In addition, AD and bvFTD patients showed different patterns of atrophy, with AD exhibiting more central (temporal areas) and bvFTD more anterior (frontal areas) atrophy. A correlation was found between grey matter density maps and CSF marker concentrations in the AD group, with total Tau and phosphorylated Tau levels showing a high association with low grey matter density in the left superior temporal gyrus. The study concludes that while bvFTD lacks a CSF marker profile, CSF beta-amyloid levels are useful for differentiating AD from bvFTD. Furthermore, MR structural imaging can contribute significantly to distinguishing between the two pathologies.

Quality assessment and control of unprocessed anatomical, functional, and diffusion MRI of the human brain using MRIQC.

Quality control of MRI data prior to preprocessing is fundamental, as substandard data are known to increase variability spuriously. Currently, no automated or manual method reliably identifies subpar images, given pre-specified exclusion criteria. In this work, we propose a protocol describing how to carry out the visual assessment of T1-weighted, T2-weighted, functional, and diffusion MRI scans of the human brain with the visual reports generated by MRIQC. The protocol describes how to execute the software on all the images of the input dataset using typical research settings (i.e., a high-performance computing cluster). We then describe how to screen the visual reports generated with MRIQC to identify artifacts and potential quality issues and annotate the latter with the "rating widget" - a utility that enables rapid annotation and minimizes bookkeeping errors. Integrating proper quality control checks on the unprocessed data is fundamental to producing reliable statistical results and crucial to identifying faults in the scanning settings, preempting the acquisition of large datasets with persistent artifacts that should have been addressed as they emerged.

Spinal Cord Blood Perfusion Deficit is Associated with Clinical Impairment after Spinal Cord Injury.

Spinal cord injury (SCI) results in intramedullary microvasculature disruption and blood perfusion deficit at and remote from the injury site. However, the relationship between remote vascular impairment and functional recovery remains understudied. We characterized perfusion impairment invivo, rostral to the injury, using magnetic resonance imaging (MRI), and investigated its association with lesion extent and impairment following SCI. Twenty-one patients with chronic cervical SCI and 39 healthy controls (HC) underwent a high-resolution MRI protocol, including intravoxel incoherent motion (IVIM) and T2*-weighted MRI covering C1-C3 cervical levels, as well as T2-weighted MRI to determine lesion volumes. IVIM matrices (i.e., blood volume fraction, velocity, flow indices, and diffusion) and cord structural characteristics were calculated to assess perfusion changes and cervical cord atrophy, respectively. Patients with SCI additionally underwent a standard clinical examination protocol to assess functional impairment. Correlation analysis was used to investigate associations between IVIM parameters with lesion volume and sensorimotor dysfunction. Cervical cord white and gray matter were atrophied (27.60% and 21.10%, p < 0.0001, respectively) above the cervical cord injury, accompanied by a lower blood volume fraction (-22.05%, p < 0.001) and a higher blood velocity-related index (+38.72%, p < 0.0001) in patients with SCI compared with HC. Crucially, gray matter remote perfusion deficit correlated with larger lesion volumes and clinical impairment. This study shows clinically eloquent perfusion deficit rostral to a SCI, its magnitude driven by injury severity. These findings indicate trauma-induced widespread microvascular alterations beyond the injury site. Perfusion MRI matrices in the spinal cord hold promise as biomarkers for monitoring treatment effects and dynamic changes in microvasculature integrity following SCI.

Markers of low field NMR relaxation features of tissues

This work presents an approach to exploiting Nuclear Magnetic Resonance (NMR) relaxometry data (1H spin-lattice relaxation rates covering the frequency range from below 1 kHz to 10 MHz) for the purpose of differentiating between pathological and reference tissues. Characteristic quantities (markers) that can be obtained in a straightforward manner, not resorting to an advanced analysis of 1H spin-lattice relaxation data, have been identified and compared for pathological and reference colon tissues. Moreover, the relaxation data have been parametrised in terms of Lorentzian spectral densities and the possibility of using the obtained dipolar relaxation constants and correlation times as biomarkers to assess the state of tissues has been discussed. It has also been demonstrated that the relaxation data for the reference and the pathological tissues can be attributed to two groups (for each case). The studies are a step towards exploiting the potential of NMR relaxometry for characterisation of pathological changes in tissues.

Comprehensive segmentation of gray matter structures on T1-weighted brain MRI: A Comparative Study of CNN, CNN hybrid-transformer or -Mamba architectures.

Recent advances in deep learning have shown promising results in medical image analysis and segmentation. However, most brain MRI segmentation models are limited by the size of their datasets and/or the number of structures they can identify. This study evaluates the performance of six advanced deep learning models in segmenting 122 brain structures from T1-weighted MRI scans, aiming to identify the most effective model for clinical and research applications. 1,510 T1-weighted MRIs were used to compare six deep-learning models for the segmentation of 122 distinct gray matter structures: nnU-Net, SegResNet, SwinUNETR, UNETR, U-Mamba_BOT and U-Mamba_ Enc. Each model was rigorously tested for accuracy using the Dice Similarity Coefficient (DSC) and the 95th percentile Hausdorff Distance (HD95). Additionally, the volume of each structure was calculated and compared between normal control (NC) and Alzheimer's Disease (AD) patients. U-Mamba_Bot achieved the highest performance with a median DSC of 0.9112 [IQR:0.8957, 0.9250]. nnU-Net achieved a median DSC of 0.9027 [IQR: 0.8847, 0.9205] and had the highest HD95 of 1.392[IQR: 1.174, 2.029]. The value of each HD95 (<3mm) indicates its superior capability in capturing detailed brain structures accurately. Following segmentation, volume calculations were performed, and the resultant volumes of normal controls and AD patients were compared. The volume changes observed in thirteen brain substructures were all consistent with those reported in existing literature, reinforcing the reliability of the segmentation outputs. This study underscores the efficacy of U-Mamba_Bot as a robust tool for detailed brain structure segmentation in T1-weighted MRI scans. The congruence of our volumetric analysis with the literature further validates the potential of advanced deep-learning models to enhance the understanding of neurodegenerative diseases such as AD. Future research should consider larger datasets to validate these findings further and explore the applicability of these models in other neurological conditions. AD = Alzheimer's Disease; ADNI = Alzheimer's Disease Neuroimaging Initiative; DSC = Dice Similarity Coefficient; HD95 = the 95th Percentile Hausdorff Distance; IQR = Interquartile Range; NC = Normal Control; SSMs = State-space Sequence Models.

Study on the classification of benign and malignant breast lesions using a multi-sequence breast MRI fusion radiomics and deep learning model

PurposeTo develop a multi-modal model combining multi-sequence breast MRI fusion radiomics and deep learning for the classification of benign and malignant breast lesions, to assist clinicians in better selecting treatment plans. MethodsA total of 314 patients who underwent breast MRI examinations were included. They were randomly divided into training, validation, and test sets in a ratio of 7:1:2. Subsequently, features of T1-weighted images (T1WI), T2-weighted images (T2WI), and dynamic contrast-enhanced MRI (DCE-MRI) were extracted using the convolutional neural network ResNet50 for fusion, and then combined with radiomic features from the three sequences. The following models were established: T1 model, T2 model, DCE model, DCE_T1_T2 model, and DCE_T1_T2_rad model. The performance of the models was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. The differences between the DCE_T1_T2_rad model and the other four models were compared using the Delong test, with a P-value < 0.05 considered statistically significant. ResultsThe five models established in this study performed well, with AUC values of 0.53 for the T1 model, 0.62 for the T2 model, 0.79 for the DCE model, 0.94 for the DCE_T1_T2 model, and 0.98 for the DCE_T1_T2_rad model. The DCE_T1_T2_rad model showed statistically significant differences (P < 0.05) compared to the other four models. ConclusionThe use of a multi-modal model combining multi-sequence breast MRI fusion radiomics and deep learning can effectively improve the diagnostic performance of breast lesion classification.

Age-specific optimization of the T2-weighted MRI contrast in infant and toddler brain.

In 0-2-year-old brains, the T2-weighted (T2w) contrast between white matter (WM) and gray matter (GM) is weaker compared with that in adult brains and rapidly changes with age. This study aims to design variable-flip-angle (VFA) trains in 3D fast spin-echo sequence that adapt to the dynamically changing relaxation times to improve the contrast in the T2w images of the developing brains. T1 and T2 relaxation times in 0-2-year-old brains were measured, and several age groups were defined according to the age-dependent pattern of T2 values. Based on the static pseudo-steady-state theory and the extended phase graph algorithm, VFA trains were designed for each age group to maximize WM/GM contrast, constrained by the maximum specific absorption rate and overall signal intensity. The optimized VFA trains were compared with the default one used for adult brains based on the relative contrast between WM and GM. Dice coefficient was used to demonstrate the advantage of contrast-improved images as inputs for automatic tissue segmentation in infant brains. The 0-2-year-old pool was divided into groups of 0-8 months, 8-12 months, and 12-24 months. The optimal VFA trains were tested in each age group in comparison with the default sequence. Quantitative analyses demonstrated improved relative contrasts in infant and toddler brains by 1.5-2.3-fold at different ages. The Dice coefficient for contrast-optimized images was improved compared with default images (p < 0.001). An effective strategy was proposed to improve the 3D T2w contrast in 0-2-year-old brains.

Refined Analysis of the Crystal Structure, Thermodynamic Property, and Structural Geometry of Perovskite (CH3NH3)2ZnBr4.

Organic-inorganic hybrid single crystals of (CH3NH3)2ZnBr4 were grown by the aqueous solution technique, and their structure, phase transition temperature, and thermal stability were discussed. The crystals were determined to have a monoclinic structure using X-ray diffraction on single crystals, and a transition temperature of 458 K was confirmed from the differential scanning calorimetry and X-ray diffraction analyses on powders. Moreover, the compound exhibited excellent thermal stabilities up to 550 K. The 1H and 13C nuclear magnetic resonance chemical shifts of CH3 shifted in the positive direction as the temperature increased, while the 1H chemical shift of NH2 exhibited minor variations in N-H···Br hydrogen bonding, which was influenced by minor changes in the bromide ions surrounding Zn in the ZnBr4 anion. The spin-lattice relaxation time, which indicates the energy transfer near 1H and 13C, exhibited similar values, which is because 1H and 13C are bonded to each other in the CH3NH3 cation. This compound exhibits high structural and thermal stabilities and is expected to pave the way for future applications.

Large-scale structural covariance networks changes relate to executive function deficit in betel quid-dependent chewers.

Previous studies demonstrate deficits in executive function for betel quid-dependent (BQD) patients. Large-scale structural covariance network (SCN) based on gray matter (GM) morphometry may be able to explore the neural mechanism of executive dysfunction in BQD individuals. This study aims to identify spatial covariance patterns of GM volume and to investigate their association with executive dysfunction in BQD individuals. Sixty-four BQD individuals and 48 sex- and age-matched healthy controls (HCs) underwent T1-weighted structural MRI examination and executive function assessments, including the Backward Digit Span (BDS) test and Stroop Color and Word (SCW) test. Seventy SCNs based on GM volume covariance patterns were defined using independent component analysis. An SCN-based classifier was constructed to differentiate between BQD and HC individuals. Receiver operating characteristic (ROC) curves were applied to evaluate the performance of the SCN-based classifier. Linear regression analyses were performed to investigate the association between SCN network indices and executive function indices. Six SCNs had higher classifications for differentiating between BQD and HC individuals. The area under the ROC curve of the SCN-based classifier was 0.812 in the training set and 0.771 in the testing set. Furthermore, linear regression analyses demonstrated that the network indices in the thalamus were associated with BDS scores adjusted for age, sex, and education. Large-scale SCNs could provide potential imaging markers for differentiating BQD and HC groups. The loss of network index in the thalamus was associated with working memory, indicating that SCNs could reveal executive dysfunction in BQD individuals.

Directly monitoring the dynamic in vivo metabolisms of hyperpolarized 13C-oligopeptides.

Peptides play essential roles in biological phenomena, and, thus, there is a growing interest in detecting in vivo dynamics of peptide metabolisms. Dissolution-dynamic nuclear polarization (d-DNP) is a state-of-the-art technology that can markedly enhance the sensitivity of nuclear magnetic resonance (NMR), providing metabolic and physiological information in vivo. However, the hyperpolarized state exponentially decays back to the thermal equilibrium, depending on the spin-lattice relaxation time (T1). Because of the limitation in T1, peptide-based DNP NMR molecular probes applicable in vivo have been limited to amino acids or dipeptides. Here, we report the direct detection of in vivo metabolic conversions of hyperpolarized 13C-oligopeptides. Structure-based T1 relaxation analysis suggests that the C-terminal [1-13C]Gly-d2 residue affords sufficient T1 for biological uses, even in relatively large oligopeptides, and allowed us to develop 13C-β-casomorphin-5 and 13C-glutathione. It was found that the metabolic response and perfusion of the hyperpolarized 13C-glutathione in the mouse kidney were significantly altered in a model of cisplatin-induced acute kidney injury.

A multicenter bladder cancer MRI dataset and baseline evaluation of federated learning in clinical application

Bladder cancer (BCa), as the most common malignant tumor of the urinary system, has received significant attention in research on the clinical application of artificial intelligence algorithms. Nevertheless, it has been observed that certain investigations use data from various medical facilities to train models for BCa, which may pose a privacy risk. Given this concern, protecting patient privacy during machine learning algorithm training is a crucial aspect that requires substantial attention. One emerging machine learning paradigm that addresses this concern is federated learning (FL). FL enables multiple entities to collaboratively build machine learning models while preserving data privacy and security. In this study, we present a multicenter BCa magnetic resonance imaging (MRI) dataset. The dataset comprises 275 three-dimensional bladder T2-weighted MRI scans collected from four medical centers, and each scan includes diagnostic pathological labels for muscle invasion and pixel-level annotations of tumor contours. Four FL methods are used to assess the baseline of the dataset for both the task of diagnosing muscle-invasive bladder cancer and automatic bladder tumor lesion segmentation.

Benchmarking Stroke Outcome Prediction through Comprehensive Data Analysis - NeuralCup 2023.

Stroke is a significant cause of mortality and long-term disability worldwide, with variable recovery trajectories posing substantial challenges in anticipating post-event care and rehabilitation planning. The NeuralCup 2023 consortium was established to address these challenges by comparing the predictability of stroke outcome models through a collaborative, data-driven approach. This study presents the consortium's findings, which involved 15 participating teams worldwide. Using a comprehensive dataset, which included clinical and imaging data, we conducted an open competition to identify and compare predictors of motor, cognitive, and neuropsychological (emotional) outcomes one-year post-stroke. Analyses incorporated both traditional and novel methods, including machine learning algorithms. These efforts culminated in the search for 'optimal recipes' for predicting each domain through an exhaustive exploration of the features of all the approaches. Key predictors included lesion characteristics, T1-weighted MRI sequences, and demographic factors. Notably, integrating FLAIR imaging and white matter tract analysis emerged as crucial to improving the accuracy of cognitive and motor outcome predictions, respectively. These findings advocate for a tailored, multifaceted approach to stroke outcome prediction, underscoring the potential of collaborative data science in addressing complex neurological prognostication challenges. This study also sets a new benchmark methodology in stroke research, offering a foundational step toward personalized care strategies that could significantly impact recovery planning and quality of life for stroke survivors.

Assessing suitability and stability of materials for a head and neck anthropomorphic multimodality (MRI/CT) phantoms for radiotherapy.

&#xD;This study aims to identify and evaluate suitable and stable materials for developing a head and neck anthropomorphic multimodality phantom for radiotherapy purposes. These materials must mimic human head and neck tissues in both computed tomography (CT) and magnetic resonance imaging (MRI) and maintain stable imaging properties over time and after radiation exposure, including the high levels associated with linear accelerator (linac) use.&#xD;Approach:&#xD;Various materials were assessed by measuring their CT numbers and T1 and T2 relaxation times. These measurements were compared to literature values to determine how closely the properties of the candidate materials resemble those of human tissues in the head and neck region. The stability of these properties was evaluated monthly over a year and after radiation exposure to doses up to 1000 Gy. Statistical analyses were conducted to identify any significant changes over time and after radiation exposure.&#xD;Main Results:&#xD;10% and 12.6% Polyvinyl alcohol cryogel (PVA-c) both exhibited T1 and T2 relaxation times and CT numbers within the range appropriate for brain grey matter. 14.3% PVA-c and some plastic-based materials matched the MRI properties of brain white matter, with CT numbers close to the clinical range. Additionally, some plastic-based materials showed T1 and T2 relaxation times consistent with MRI properties of fat, although their CT numbers were not suitable. Over time and after irradiation, 10% PVA-c maintained consistent properties for brain grey matter. 12.6% PVA-c's T1 relaxation time decreased beyond the range after the first month.&#xD;Significance:&#xD;This study identified 10% PVA-c as a substitute for brain grey matter, demonstrating stable imaging properties over a year and after radiation exposure up to 1000 Gy. However, the results highlight a need for further research to find additional materials to accurately simulate a wider range of human tissues.&#xD.

Higher effect sizes for the detection of accelerated brain volume loss and disability progression in multiple sclerosis using deep-learning

PurposeClinical validation of “BrainLossNet”, a deep learning-based method for fast and robust estimation of brain volume loss (BVL) from longitudinal T1-weighted MRI, for the detection of accelerated BVL in multiple sclerosis (MS) and for the discrimination between MS patients with versus without disability progression. Materials and MethodsA longitudinal normative database of healthy controls (n=563), two mono-scanner MS cohorts (n=414, 156) and a mixed-scanner cohort acquired for various indications (n=216) were included retrospectively. Mean observation period from the baseline (BL) to the last follow-up (FU) MRI scan was 2-3 years. Expanded Disability Status Scale (EDSS) at BL and FU was available in 149 MS patients. Annual BVL was computed using BrainLossNet and Siena and then adjusted for age. Repeated-measures ANOVA and Cohen’s effect size were used to compare BrainLossNet and Siena regarding the detection of accelerated BVL and the differentiation between MS patients with versus without EDSS progression. ResultsCohen’s effect size for the differentiation of patients from healthy controls based on the age-adjusted annual BVL was larger with BrainLossNet than with Siena (MS cohort 1: 0.927 versus 0.495, MS cohort 2: 0.671 versus 0.456, mixed-scanner cohort: 0.918 versus 0.730, all p<0.001). Cohen’s effect size for the discrimination between MS patients with (n=51) versus without (n=98) EDSS progression was larger with BrainLossNet (0.503 versus 0.400, p=0.048). ConclusionBrainLossNet can provide added value in clinical routine and MS therapy trials regarding the detection of accelerated BVL in MS and the differentiation between MS patients with versus without disability progression.

MRI recovery with self-calibrated denoisers without fully-sampled data.

Acquiring fully sampled training data is challenging for many MRI applications. We present a self-supervised image reconstruction method, termed ReSiDe, capable of recovering images solely from undersampled data. ReSiDe is inspired by plug-and-play (PnP) methods, but unlike traditional PnP approaches that utilize pre-trained denoisers, ReSiDe iteratively trains the denoiser on the image or images that are being reconstructed. We introduce two variations of our method: ReSiDe-S and ReSiDe-M. ReSiDe-S is scan-specific and works with a single set of undersampled measurements, while ReSiDe-M operates on multiple sets of undersampled measurements and provides faster inference. Studies I, II, and III compare ReSiDe-S and ReSiDe-M against other self-supervised or unsupervised methods using data from T1- and T2-weighted brain MRI, MRXCAT digital perfusion phantom, and first-pass cardiac perfusion, respectively. ReSiDe-S and ReSiDe-M outperform other methods in terms of peak signal-to-noise ratio and structural similarity index measure for Studies I and II, and in terms of expert scoring for Study III. We present a self-supervised image reconstruction method and validate it in both static and dynamic MRI applications. These developments can benefit MRI applications where the availability of fully sampled training data is limited.