Development and stability analysis of carpal kinematic metrics from 4D magnetic resonance imaging.

OBJECTIVE: To develop MRI-derived carpal kinematic metrics and investigating their stability. METHODS: The study used a 4D MRI method to track scaphoid, lunate, and capitate movements in the wrist. A panel of 120 metrics for radial-ulnar deviation and flexion-extension was created using polynomial models of scaphoid and lunate movements relative to the capitate. Intraclass correlation coefficients (ICCs) analyzed intra- and inter-subject stability in 49 subjects, 20 with and 29 without wrist injury history. RESULTS: Comparable degrees of stability were observed across the two different wrist movements. Among the total 120 derived metrics, distinct subsets demonstrated high stability within each type of movement. For asymptomatic subjects, 16 out of 17 metrics with high intra-subject stability also showed high inter-subject stability. The differential analysis of ICC values for each metric between asymptomatic and symptomatic cohorts revealed specific metrics (although relatively unstable) exhibiting greater variability in the symptomatic cohort, thereby highlighting the impact of wrist conditions on the variability of kinematic metrics. CONCLUSION: The findings demonstrate the developing potential of dynamic MRI for assessing and characterizing complex carpal bone dynamics. Stability analyses of the derived kinematic metrics revealed encouraging differences between cohorts with and without wrist injury histories. Although these broad metric stability variations highlight the potential utility of this approach for analyzing carpal instability, further studies are necessary to better characterize these observations.

Diffusion Weighted Magnetic Resonance Imaging of Spinal Cord Injuries After Instrumented Fusion Stabilization.

Diffusion-weighted magnetic resonance imaging (DW-MRI) is a promising technique for assessing spinal cord injury (SCI) that has historically been challenged by the presence of metallic stabilization hardware. This study leverages recent advances in metal-artifact resistant multi-spectral DW-MRI to enable diffusion quantification throughout the spinal cord even after fusion stabilization. Twelve participants with cervical spinal cord injuries treated with fusion stabilization and 49 asymptomatic able-bodied control participants underwent multi-spectral DW-MRI evaluation. Apparent diffusion coefficient (ADC) values were calculated in axial cord sections. Statistical modeling assessed ADC differences across cohorts and within distinct cord regions of the SCI participants (at, above, or below injured level). Computed models accounted for subject demographics and injury characteristics. ADC was found to be elevated at injured levels compared with non-injured levels (z = 3.2, p = 0.001), with ADC at injured levels decreasing over time since injury (z = -9.2, p < 0.001). Below the injury level, ADC was reduced relative to controls (z = -4.4, p < 0.001), with greater reductions after more severe injuries that correlated with lower extremity motor scores (z = 2.56, p = 0.012). No statistically significant differences in ADC above the level of injury were identified. By enabling diffusion analysis near fusion hardware, the multi-spectral DW-MRI technique allowed intuitive quantification of cord diffusion changes after SCI both at and away from injured levels. This demonstrates the approach's potential for assessing post-surgical spinal cord integrity throughout stabilized regions.

Association of neighborhood deprivation with white matter connectome abnormalities in temporal lobe epilepsy.

OBJECTIVE: Social determinants of health, including the effects of neighborhood disadvantage, impact epilepsy prevalence, treatment, and outcomes. This study characterized the association between aberrant white matter connectivity in temporal lobe epilepsy (TLE) and disadvantage using a US census-based neighborhood disadvantage metric, the Area Deprivation Index (ADI), derived from measures of income, education, employment, and housing quality. METHODS: Participants including 74 TLE patients (47 male, mean age = 39.2 years) and 45 healthy controls (27 male, mean age = 31.9 years) from the Epilepsy Connectome Project were classified into ADI-defined low and high disadvantage groups. Graph theoretic metrics were applied to multishell connectome diffusion-weighted imaging (DWI) measurements to derive 162 × 162 structural connectivity matrices (SCMs). The SCMs were harmonized using neuroCombat to account for interscanner differences. Threshold-free network-based statistics were used for analysis, and findings were correlated with ADI quintile metrics. A decrease in cross-sectional area (CSA) indicates reduced white matter integrity. RESULTS: Sex- and age-adjusted CSA in TLE groups was significantly reduced compared to controls regardless of disadvantage status, revealing discrete aberrant white matter tract connectivity abnormalities in addition to apparent differences in graph measures of connectivity and network-based statistics. When comparing broadly defined disadvantaged TLE groups, differences were at trend level. Sensitivity analyses of ADI quintile extremes revealed significantly lower CSA in the most compared to least disadvantaged TLE group. SIGNIFICANCE: Our findings demonstrate (1) the general impact of TLE on DWI connectome status is larger than the association with neighborhood disadvantage; however, (2) neighborhood disadvantage, indexed by ADI, revealed modest relationships with white matter structure and integrity on sensitivity analysis in TLE. Further studies are needed to explore this relationship and determine whether the white matter relationship with ADI is driven by social drift or environmental influences on brain development. Understanding the etiology and course of the disadvantage-brain integrity relationship may serve to inform care, management, and policy for patients.

Diffusion-weighted MRI of the spinal cord in cervical spondylotic myelopathy after instrumented fusion.

Introduction: This study investigated tissue diffusion properties within the spinal cord of individuals treated for cervical spondylotic myelopathy (CSM) using post-decompression stabilization hardware. While previous research has indicated the potential of diffusion-weighted MRI (DW-MRI) markers of CSM, the metallic implants often used to stabilize the decompressed spine hamper conventional DW-MRI. Methods: Utilizing recent developments in DW-MRI metal-artifact suppression technologies, imaging data was acquired from 38 CSM study participants who had undergone instrumented fusion, as well as asymptomatic (non-instrumented) control participants. Apparent diffusion coefficients were determined in axial slice sections and split into four categories: a) instrumented levels, b) non-instrumented CSM levels, c) adjacent-segment (to instrumentation) CSM levels, and d) non-instrumented control levels. Multi-linear regression models accounting for age, sex, and body mass index were used to investigate ADC measures within each category. Furthermore, the cord diffusivity within CSM subjects was correlated with symptom scores and the duration since fusion procedures. Results: ADC measures of the spinal cord in CSM subjects were globally reduced relative to control subjects (p = 0.005). In addition, instrumented levels within the CSM subjects showed reduced diffusivity relative to controls (p = 0.003), while ADC within non-instrumented CSM levels did not statistically deviate from control levels (p = 0.107). Discussion: Multi-spectral DW-MRI technology can be effectively employed to evaluate cord diffusivity near fusion hardware in subjects who have undergone surgery for CSM. Leveraging this advanced technology, this study had identified significant reductions in cord diffusivity, relative to control subjects, in CSM patients treated with conventional metallic fusion instrumentation.

Development and Stability Analysis of Carpal Kinematic Metrics from 4D Magnetic Resonance Imaging.

INTRODUCTION: Wrist instability remains a common health concern. The potential of dynamic Magnetic Resonance Imaging (MRI) in assessing carpal dynamics associated with this condition is a field of ongoing research. This study contributes to this line of inquiry by developing MRI-derived carpal kinematic metrics and investigating their stability. METHODS: A previously described 4D MRI approach for tracking the movements of carpal bones in the wrist was deployed in this study. A panel of 120 metrics characterizing radial/ulnar deviation and flexion extension movements was constructed by fitting low order polynomial models of scaphoid and lunate degrees of freedom against that of the capitate. Intraclass Correlation Coefficients were utilized to analyze intra- and inter-subject stability within a mixed cohort of 49 subjects, including 20 with and 29 without a history of wrist injury. RESULTS: A comparable degree of stability across the two different wrist movements. Out of the total 120 derived metrics, distinct subsets demonstrated high stability within each type of movement. Among asymptomatic subjects, 16 out of 17 metrics with high intra-subject stability also showed high inter-subject stability. Interestingly, some quadratic term metrics, although relatively unstable within asymptomatic subjects, showed increased stability within this group, hinting at potential differentiation in their behavior across different cohorts. CONCLUSION: This study showed the developing potential of dynamic MRI to characterize complex carpal bone dynamics. Stability analyses of the derived kinematic metrics showed encouraging differences between cohorts with and without a history of wrist injury. Although these broad metric stability variations highlight the potential utility of this approach for analysis of carpal instability, further studies are necessary to better characterize these observations.

Functional Connectivity Magnetic Resonance Imaging Sequences in Patients With Postsurgical Persistent Spinal Pain Syndrome Type 2 With Implanted Spinal Cord Stimulation Systems: A Safety, Feasibility, and Validity Study.

BACKGROUND: Chronic pain has been associated with alterations in brain connectivity, both within networks (regional) and between networks (cross-network connectivity). Functional connectivity (FC) data on chronic back pain are limited and based on heterogeneous pain populations. Patients with postsurgical persistent spinal pain syndrome (PSPS) type 2 are good candidates for spinal cord stimulation (SCS) therapy. We hypothesize that 1) FC magnetic resonance imaging (fcMRI) scans can be safely obtained in patients with PSPS type 2 with implanted therapeutic SCS devices and that 2) their cross-network connectivity patterns are altered and involve emotion and reward/aversion functions. MATERIALS AND METHODS: Resting-state (RS) fcMRI (rsfcMRI) scans were obtained from nine patients with PSPS type 2 implanted with therapeutic SCS systems and 13 age-matched controls. Seven RS networks were analyzed, including the striatum. RESULTS: Cross-network FC sequences were safely obtained on a 3T MRI scanner in all nine patients with PSPS type 2 with implanted SCS systems. FC patterns involving emotion/reward brain circuitry were altered as compared with controls. Patients with a history of constant neuropathic pain, experiencing longer therapeutic effects of SCS, had fewer alterations in their connectivity patterns. CONCLUSIONS: To our knowledge, this is the first report of altered cross-network FC involving emotion/reward brain circuitry in a homogeneous population of patients with chronic pain with fully implanted SCS systems, on a 3T MRI scanner. All rsfcMRI studies were safe and well tolerated by all nine patients, with no detectable effects on the implanted devices.

Predicting Neoadjuvant Treatment Response in Rectal Cancer Using Machine Learning: Evaluation of MRI-Based Radiomic and Clinical Models.

BACKGROUND: Radiomics is an approach to medical imaging that quantifies the features normally translated into visual display. While both radiomic and clinical markers have shown promise in predicting response to neoadjuvant chemoradiation therapy (nCRT) for rectal cancer, the interrelationship is not yet clear. METHODS: A retrospective, single-institution study of patients treated with nCRT for locally advanced rectal cancer was performed. Clinical and radiomic features were extracted from electronic medical record and pre-treatment magnetic resonance imaging, respectively. Machine learning models were created and assessed for complete response and positive treatment effect using the area under the receiver operating curves. RESULTS: Of 131 rectal cancer patients evaluated, 68 (51.9%) were identified to have a positive treatment effect and 35 (26.7%) had a complete response. On univariate analysis, clinical T-stage (OR 0.46, p = 0.02), lymphovascular/perineural invasion (OR 0.11, p = 0.03), and statin use (OR 2.45, p = 0.049) were associated with a complete response. Clinical T-stage (OR 0.37, p = 0.01), lymphovascular/perineural invasion (OR 0.16, p = 0.001), and abnormal carcinoembryonic antigen level (OR 0.28, p = 0.002) were significantly associated with a positive treatment effect. The clinical model was the strongest individual predictor of both positive treatment effect (AUC = 0.64) and complete response (AUC = 0.69). The predictive ability of a positive treatment effect increased by adding tumor and mesorectal radiomic features to the clinical model (AUC = 0.73). CONCLUSIONS: The use of a combined model with both clinical and radiomic features resulted in the strongest predictive capability. With the eventual goal of tailoring treatment to the individual, both clinical and radiologic markers offer insight into identifying patients likely to respond favorably to nCRT.

Dynamic tracking of scaphoid, lunate, and capitate carpal bones using four-dimensional MRI.

A preliminary exploration of technical methodology for dynamic analysis of scaphoid, capitate, and lunate during unconstrained movements is performed in this study. A heavily accelerated and fat-saturated 3D Cartesian MRI acquisition was used to capture temporal frames of the unconstrained moving wrist of 5 healthy subjects. A slab-to-volume point-cloud based registration was then utilized to register the moving volumes to a high-resolution image volume collected at a neutral resting position. Comprehensive in-silico error analyses for different acquisition parameter settings were performed to evaluate the performance limits of several dynamic metrics derived from the registration parameters. Computational analysis suggested that sufficient volume coverage for the dynamic acquisitions was reached when collecting 12 slice-encodes at 2.5mm resolution, which yielded a temporal resolution of and 2.6 seconds per volumetric frame. These acquisition parameters resulted in total in-silico errors of 1.9°±1.8° and 3°±4.6° in derived principal rotation angles within ulnar-radial deviation and flexion-extension motion, respectively. Rotation components of the carpal bones in the radius coordinate system were calculated and found to be consistent with earlier 4D-CT studies. Temporal metric profiles derived from ulnar-radial deviation motion demonstrated better performance than those derived from flexion/extension movements. Future work will continue to explore the use of these methods in deriving more complex dynamic metrics and their application to subjects with symptomatic carpal dysfunction.

Head Impact Exposure, Gray Matter Volume, and Moderating Effects of Estimated Intelligence Quotient and Educational Attainment in Former Athletes at Midlife.

Repetitive head impact (RHI) exposure has been associated with differences in brain structure among younger active athletes, most often within the hippocampus. Studies of former athletes at early-midlife are limited. We investigated the association between RHI exposure and gray matter (GM) structure, as well as moderating factors, among former athletes in early-midlife. Former collegiate football players (n = 55; age = 37.9 + 1.5 years) completed magnetic resonance imaging to quantify GM morphometry and extensive structured interviews of RHI history (Head Impact Exposure Estimate). Linear regression models tested the association between RHI exposure and GM structures of interest. Interactions were tested for moderators: two estimates of intelligence quotient (IQ) (single word reading and picture vocabulary) and education history. Greater RHI exposure was associated with smaller hippocampal volume, ß = -0.36, p = 0.004. Conversely, RHI exposure was not significantly associated with other GM outcomes ps > 0.05. Education history significantly moderated the association between RHI exposure and hippocampal volume, ß = 0.31, p = 0.047. Among those with a bachelor's degree, greater RHI exposure was significantly associated with smaller hippocampal volumes, ß = -0.58, p < 0.001. For those with graduate/professional degrees, the association between RHI and hippocampal volume was not significant, ß = -0.33, p = 0.134. Consistent with studies involving younger, active athletes, smaller hippocampal volumes were selectively associated with greater RHI exposure among former collegiate football players at midlife. This relationship was moderated by higher levels of education. Future longitudinal studies are needed to investigate the course of possible changes that can occur between early-midlife and older ages, as well as the continued protective effect of education and other potential influential factors.

Diffusion propagator metrics are biased when simultaneous multi-slice acceleration is used.

Advanced diffusion MRI models are being explored to study the complex microstructure of the brain with higher accuracy. However, these techniques require long acquisition times. Simultaneous Multi-Slice (SMS) accelerates data acquisition by exciting multiple image slices simultaneously and separating the overlapping slices using a mathematical model, which makes use of the distinct information coming from an array of receive coils. However, SMS acceleration introduces increased noise in reconstructed images and crosstalk between simultaneously excited slices. These compounded effects from SMS acceleration could affect quantitative MRI techniques such as diffusion imaging. In this study, the effects of SMS acceleration on the accuracy of propagator metrics obtained from a model-free advanced diffusion technique called Mean Apparent Propagator MRI (MAP-MRI) was investigated. Ten healthy volunteers were scanned with SMS accelerated multi-shell diffusion MRI acquisitions. Group analyses were performed to study brain regions affected by SMS acceleration. In addition, diffusion metrics from atlas-based fiber tracts of interest were analyzed to investigate how propagator metrics in major fiber tracts were biased by 2- and 3-band SMS acceleration. Both zero-displacement metrics and non-Gaussianity metrics were significantly altered when SMS acceleration was used. MAP-MRI metrics calculated from SMS-3 showed significant differences with respect to SMS-2. Furthermore, with the shorter TR afforded by SMS acceleration, the characteristics of this bias have changed. This has implications for studies using diffusion MRI with SMS acceleration to investigate the effects of a disease or injury on the brain tissues.

Analysis and Evaluation of a Deep Learning Reconstruction Approach with Denoising for Orthopedic MRI.

PURPOSE: To evaluate two settings (noise reduction of 50% or 75%) of a deep learning (DL) reconstruction model relative to each other and to conventional MR image reconstructions on clinical orthopedic MRI datasets. MATERIALS AND METHODS: This retrospective study included 54 patients who underwent two-dimensional fast spin-echo MRI for hip (n = 22; mean age, 44 years ± 13 [standard deviation]; nine men) or shoulder (n = 32; mean age, 56 years ± 17; 17 men) conditions between March 2019 and June 2020. MR images were reconstructed with conventional methods and the vendor-provided and commercially available DL model applied with 50% and 75% noise reduction settings (DL 50 and DL 75, respectively). Quantitative analytics, including relative anatomic edge sharpness, relative signal-to-noise ratio (rSNR), and relative contrast-to-noise ratio (rCNR) were computed for each dataset. In addition, the image sets were randomized, blinded, and presented to three board-certified musculoskeletal radiologists for ranking based on overall image quality and diagnostic confidence. Statistical analysis was performed with a nonparametric hypothesis comparing derived quantitative metrics from each reconstruction approach. In addition, inter- and intrarater agreement analysis was performed on the radiologists' rankings. RESULTS: Both denoising settings of the DL reconstruction showed improved edge sharpness, rSNR, and rCNR relative to the conventional reconstructions. The reader rankings demonstrated strong agreement, with both DL reconstructions outperforming the conventional approach (Gwet agreement coefficient = 0.98). However, there was lower agreement between the readers on which DL reconstruction denoising setting produced higher-quality images (Gwet agreement coefficient = 0.31 for DL 50 and 0.35 for DL 75). CONCLUSION: The vendor-provided DL MRI reconstruction showed higher edge sharpness, rSNR, and rCNR in comparison with conventional methods; however, optimal levels of denoising may need to be further assessed.Keywords: MRI Reconstruction Method, Deep Learning, Image Analysis, Signal-to-Noise Ratio, MR-Imaging, Neural Networks, Hip, Shoulder, Physics, Observer Performance, Technology Assessment Supplemental material is available for this article. © RSNA, 2021.

Value CMR: Towards a Comprehensive, Rapid, Cost-Effective Cardiovascular Magnetic Resonance Imaging.

Cardiac magnetic resonance imaging (CMR) is considered the gold standard for measuring cardiac function. Further, in a single CMR exam, information about cardiac structure, tissue composition, and blood flow could be obtained. Nevertheless, CMR is underutilized due to long scanning times, the need for multiple breath-holds, use of a contrast agent, and relatively high cost. In this work, we propose a rapid, comprehensive, contrast-free CMR exam that does not require repeated breath-holds, based on recent developments in imaging sequences. Time-consuming conventional sequences have been replaced by advanced sequences in the proposed CMR exam. Specifically, conventional 2D cine and phase-contrast (PC) sequences have been replaced by optimized 3D-cine and 4D-flow sequences, respectively. Furthermore, conventional myocardial tagging has been replaced by fast strain-encoding (SENC) imaging. Finally, T1 and T2 mapping sequences are included in the proposed exam, which allows for myocardial tissue characterization. The proposed rapid exam has been tested in vivo. The proposed exam reduced the scan time from >1 hour with conventional sequences to <20 minutes. Corresponding cardiovascular measurements from the proposed rapid CMR exam showed good agreement with those from conventional sequences and showed that they can differentiate between healthy volunteers and patients. Compared to 2D cine imaging that requires 12-16 separate breath-holds, the implemented 3D-cine sequence allows for whole heart coverage in 1-2 breath-holds. The 4D-flow sequence allows for whole-chest coverage in less than 10 minutes. Finally, SENC imaging reduces scan time to only one slice per heartbeat. In conclusion, the proposed rapid, contrast-free, and comprehensive cardiovascular exam does not require repeated breath-holds or to be supervised by a cardiac imager. These improvements make it tolerable by patients and would help improve cost effectiveness of CMR and increase its adoption in clinical practice.

Split-slice training and hyperparameter tuning of RAKI networks for simultaneous multi-slice reconstruction.

PURPOSE: Simultaneous multi-slice acquisitions are essential for modern neuroimaging research, enabling high temporal resolution functional and high-resolution q-space sampling diffusion acquisitions. Recently, deep learning reconstruction techniques have been introduced for unaliasing these accelerated acquisitions, and robust artificial-neural-networks for k-space interpolation (RAKI) have shown promising capabilities. This study systematically examines the impacts of hyperparameter selections for RAKI networks, and introduces a novel technique for training data generation which is analogous to the split-slice formalism used in slice-GRAPPA. METHODS: RAKI networks were developed with variable hyperparameters and with and without split-slice training data generation. Each network was trained and applied to five different datasets including acquisitions harmonized with Human Connectome Project lifespan protocol. Unaliasing performance was assessed through L1 errors computed between unaliased and calibration frequency-space data. RESULTS: Split-slice training significantly improved network performance in nearly all hyperparameter configurations. Best unaliasing results were achieved with three layer RAKI networks using at least 64 convolutional filters with receptive fields of 7 voxels, 128 single-voxel filters in the penultimate RAKI layer, batch normalization, and no training dropout with the split-slice augmented training dataset. Networks trained without the split-slice technique showed symptoms of network over-fitting. CONCLUSIONS: Split-slice training for simultaneous multi-slice RAKI networks positively impacts network performance. Hyperparameter tuning of such reconstruction networks can lead to further improvements in unaliasing performance.

Acute Post-Concussive Assessments of Brain Tissue Magnetism Using Magnetic Resonance Imaging.

Recent studies have demonstrated the promising capabilities of magnetic resonance imaging (MRI)-based quantitative susceptibility maps (QSM) in producing biomarkers of brain injury. The present study aims to further explore acute QSM changes in athletes after sports concussion and investigate prognostication capabilities of QSM-derived imaging metrics. The QSM were derived from neurological MRI data acquired on a cohort (n = 78) of concussed male American football athletes within 48 h of injury. The MRI-derived QSM values in subcortical gray and white matter compartments after concussion showed differences relative to a matched uninjured control group (white matter: z = 3.04, p = 0.002, subcortical gray matter: z = -2.07, p = 0.04). Subcortical gray matter QSM MRI measurements also correlated strongly with duration of symptoms (ρ = -0.46, p = 0.002) within a subcohort of subjects who had symptom durations for at least one week (n = 39). The acute QSM MRI metrics showed promising prognostication capabilities, with subcortical gray matter compartment QSM values yielding a mean classification area under the curve performance of 0.78 when predicting symptoms of more than two weeks in duration. The results of the study reproduce previous acute post-concussion group QSM findings and provide promising initial prognostication capabilities of acute QSM measurements in a post-concussion setting.

Cardiac functional magnetic resonance imaging at 7T: Image quality optimization and ultra-high field capabilities.

BACKGROUND: 7T cardiac magnetic resonance imaging (MRI) introduces several advantages, as well as some limitations, compared to lower-field imaging. The capabilities of ultra-high field (UHF) MRI have not been fully exploited in cardiac functional imaging. AIM: To optimize 7T cardiac MRI functional imaging without the need for conducting B1 shimming or subject-specific tuning, which improves scan efficiency. In this study, we provide results from phantom and in vivo scans using a multi-channel transceiver modular coil. METHODS: We investigated the effects of adding a dielectric pad at different locations next to the imaged region of interest on improving image quality in subjects with different body habitus. We also investigated the effects of adjusting the imaging flip angle in cine and tagging sequences on improving image quality, B1 field homogeneity, signal-to-noise ratio (SNR), blood-myocardium contrast-to-noise ratio (CNR), and tagging persistence throughout the cardiac cycle. RESULTS: The results showed the capability of achieving improved image quality with high spatial resolution (0.75 mm × 0.75 mm × 2 mm), high temporal resolution (20 ms), and increased tagging persistence (for up to 1200 ms cardiac cycle duration) at 7T cardiac MRI after adjusting scan set-up and imaging parameters. Adjusting the imaging flip angle was essential for achieving optimal SNR and myocardium-to-blood CNR. Placing a dielectric pad at the anterior left position of the chest resulted in improved B1 homogeneity compared to other positions, especially in subjects with small chest size. CONCLUSION: Improved regional and global cardiac functional imaging can be achieved at 7T MRI through simple scan set-up adjustment and imaging parameter optimization, which would allow for more streamlined and efficient UHF cardiac MRI.

Accurate segmentation of prostate cancer histomorphometric features using a weakly supervised convolutional neural network.

Purpose: Prostate cancer primarily arises from the glandular epithelium. Histomophometric techniques have been used to assess the glandular epithelium in automated detection and classification pipelines; however, they are often rigid in their implementation, and their performance suffers on large datasets where variation in staining, imaging, and preparation is difficult to control. The purpose of this study is to quantify performance of a pixelwise segmentation algorithm that was trained using different combinations of weak and strong stroma, epithelium, and lumen labels in a prostate histology dataset. Approach: We have combined weakly labeled datasets generated using simple morphometric techniques and high-quality labeled datasets from human observers in prostate biopsy cores to train a convolutional neural network for use in whole mount prostate labeling pipelines. With trained networks, we characterize pixelwise segmentation of stromal, epithelium, and lumen (SEL) regions on both biopsy core and whole-mount H&E-stained tissue. Results: We provide evidence that by simply training a deep learning algorithm on weakly labeled data generated from rigid morphometric methods, we can improve the robustness of classification over the morphometric methods used to train the classifier. Conclusions: We show that not only does our approach of combining weak and strong labels for training the CNN improve qualitative SEL labeling within tissue but also the deep learning generated labels are superior for cancer classification in a higher-order algorithm over the morphometrically derived labels it was trained on.

Longitudinal white-matter abnormalities in sports-related concussion: A diffusion MRI study.

OBJECTIVE: To study longitudinal recovery trajectories of white matter after sports-related concussion (SRC) by performing diffusion tensor imaging (DTI) on collegiate athletes who sustained SRC. METHODS: Collegiate athletes (n = 219, 82 concussed athletes, 68 contact-sport controls, and 69 non-contact-sport controls) were included from the Concussion Assessment, Research and Education Consortium. The participants completed clinical assessments and DTI at 4 time points: 24 to 48 hours after injury, asymptomatic state, 7 days after return-to-play, and 6 months after injury. Tract-based spatial statistics was used to investigate group differences in DTI metrics and to identify white-matter areas with persistent abnormalities. Generalized linear mixed models were used to study longitudinal changes and associations between outcome measures and DTI metrics. Cox proportional hazards model was used to study effects of white-matter abnormalities on recovery time. RESULTS: In the white matter of concussed athletes, DTI-derived mean diffusivity was significantly higher than in the controls at 24 to 48 hours after injury and beyond the point when the concussed athletes became asymptomatic. While the extent of affected white matter decreased over time, part of the corpus callosum had persistent group differences across all the time points. Furthermore, greater elevation of mean diffusivity at acute concussion was associated with worse clinical outcome measures (i.e., Brief Symptom Inventory scores and symptom severity scores) and prolonged recovery time. No significant differences in DTI metrics were observed between the contact-sport and non-contact-sport controls. CONCLUSIONS: Changes in white matter were evident after SRC at 6 months after injury but were not observed in contact-sport exposure. Furthermore, the persistent white-matter abnormalities were associated with clinical outcomes and delayed recovery time.

Radiomic Features of Multiparametric MRI Present Stable Associations With Analogous Histological Features in Patients With Brain Cancer.

Magnetic resonance (MR)-derived radiomic features have shown substantial predictive utility in modeling different prognostic factors of glioblastoma and other brain cancers. However, the biological relationship underpinning these predictive models has been largely unstudied, and the generalizability of these models had been called into question. Here, we examine the localized relationship between MR-derived radiomic features and histology-derived "histomic" features using a data set of 16 patients with brain cancer. Tile-based radiomic features were collected on T1, post-contrast T1, FLAIR, and diffusion-weighted imaging (DWI)-derived apparent diffusion coefficient (ADC) images acquired before patient death, with analogous histomic features collected for autopsy samples coregistered to the magnetic resonance imaging. Features were collected for each original image, as well as a 3D wavelet decomposition of each image, resulting in 837 features per MR and histology image. Correlative analyses were used to assess the degree of association between radiomic-histomic pairs for each magnetic resonance imaging. The influence of several confounds was also assessed using linear mixed-effect models for the normalized radiomic-histomic distance, testing for main effects of different acquisition field strengths. Results as a whole were largely heterogeneous, but several features showed substantial associations with their histomic analogs, particularly those derived from the FLAIR and postcontrast T1W images. These features with the strongest association typically presented as stable across field strengths as well. These data suggest that a subset of radiomic features can consistently capture texture information on underlying tissue histology.

Functional connectivity and structural analysis of trial spinal cord stimulation responders in failed back surgery syndrome.

BACKGROUND: Chronic pain has been associated with alterations in brain structure and function that appear dependent on pain phenotype. Functional connectivity (FC) data on chronic back pain (CBP) is limited and based on heterogeneous pain populations. We hypothesize that failed back surgery syndrome (FBSS) patients being considered for spinal cord stimulation (SCS) therapy have altered resting state (RS) FC cross-network patterns that 1) specifically involve emotion and reward/aversion functions and 2) are related to pain scores. METHODS: RS functional MRI (fMRI) scans were obtained for 10 FBSS patients who are being considered for but who have not yet undergone implantation of a permanent SCS device and 12 healthy age-matched controls. Seven RS networks were analyzed including the striatum (STM). The Wilcoxon signed-rank test evaluated differences in cross-network FC strength (FCS). Differences in periaqueductal grey (PAG) FC were assessed with seed-based analysis. RESULTS: Cross-network FCS was decreased (p<0.05) between the STM and all other networks in these FBSS patients. There was a negative linear relationship (R2 = 0.76, p<0.0022) between STMFCS index and pain scores. The PAG showed decreased FC with network elements and amygdala but increased FC with the sensorimotor cortex and cingulate gyrus. CONCLUSIONS: Decreased FC between STM and other RS networks in FBSS has not been previously reported. This STMFCS index may represent a more objective measure of chronic pain specific to FBSS which may help guide patient selection for SCS and subsequent management.