Purpose: To provide a robust whole-brain quantitative magnetization transfer (MT) imaging method that is not limited by Methods: Two variants of a spiral 2D interleaved multi-slice spoiled gradient echo (SPGR) sequence are used for rapid quantitative MT imaging of the brain at 3 T. A dual flip angle, steady-state prepared, double-contrast method is used for combined B1 and-T1 mapping in combination with a single-contrast MT-prepared acquisition over a range of different saturation flip angles (50 deg to 850 deg) and offset frequencies (1 kHz and 10 kHz). Five sets (containing minimum 6 to maximum 18 scans) with different MT-weightings were acquired. In addition, main magnetic field inhomogeneities (DB0) were measured from two Cartesian low-resolution 2D SPGR scans with different echo times. Quantitative MT model parameters were derived from all sets using a two-pool continuous-wave model analysis, yielding the pool-size ratio, F, their exchange rate, kf, and their transverse relaxation time, T2r. Results: Whole-brain quantitative MT imaging was feasible for all sets with total acquisition times ranging from 7:15 min down to 3:15 min. For accurate modeling, B1-correction was essential for all investigated sets, whereas DB0-correction showed limited bias for the observed maximum off-resonances at 3 T. Conclusion: The combination of rapid B1-T1 mapping and MT-weighted imaging using a 2D multi-slice spiral SPGR research sequence offers excellent prospects for rapid whole-brain quantitative MT imaging in the clinical setting.

PURPOSE:To develop a 3D MRI-thermometry technique for transcranial MR-guided focused ultrasound (MRgFUS). METHODS:A stack of retraced in-out (RIO) spirals was incorporated into a 3D, RF-spoiled, gradient recalled echo (GRE) sequence with a minimized energy deblurring strategy. Bloch simulations examined isochromat precession during RIO readout under temperature profiles matching those encountered during transcranial MRgFUS and measured the resulting effects on observed peak temperatures and FWHM of the heating patterns. The sequence then monitored temperatures in a phantom undergoing insonation from an MRgFUS device. Finally, the sequence monitored temperatures, without insonation, in four patients immediately after MRgFUS thalamotomy. For comparison, temperatures were also monitored using a frequently used, 2D, Cartesian, multi-echo, spoiled GRE sequence. RESULTS:The RIO sequence removed under-and over-estimation of peak temperatures produced by spiral-in and spiral-out portions, respectively, of the RIO readout. In phantoms, peak temperatures from the RIO sequence were statistically indistinguishable from those from the Cartesian sequence (p > 0.05). While, in silico, the RIO sequence accurately estimated heating FWHM (error <1 mm), in phantoms, the RIO sequence overestimated FWHM (error ˜ 1 mm, p < 0.05), when compared to the Cartesian sequence. In patients, the RIO sequence had a 40% improvement in temperature efficiency compared to the Cartesian sequence, with average temperature uncertainties of 1.32 and 1.75°C, respectively. The RIO sequence also produced residual blur artifacts at tissue interfaces. CONCLUSIONS:The 3D RIO strategy with deblurring monitored focal heating under transcranial MRgFUS conditions with similar accuracy and precision as an existing 2D Cartesian method.

MR images with high signal-to-noise ratio (SNR) provide more diagnostic information. Various methods for MRI denoising have been developed, but the majority of them operate on the magnitude image and neglect the phase information. Therefore, the goal of this work is to design and implement a complex-valued convolutional neural network (CNN) for MRI denoising. A complex-valued CNN incorporating the noise level map (non-blind ℂ $$ \mathbb{C} $$ DnCNN) was trained with ground truth and simulated noise-corrupted image pairs. The proposed method was validated using both simulated and in vivo data collected from low-field scanners. Its denoising performance was quantitively and qualitatively evaluated, and it was compared with the real-valued CNN and several other algorithms. For the simulated noise-corrupted testing dataset, the complex-valued models had superior normalized root-mean-square error, peak SNR, structural similarity index, and phase ABSD. By incorporating the noise level map, the non-blind ℂ $$ \mathbb{C} $$ DnCNN showed better performance in dealing with spatially varying parallel imaging noise. For in vivo low-field data, the non-blind ℂ $$ \mathbb{C} $$ DnCNN significantly improved the SNR and visual quality of the image. The proposed non-blind ℂ $$ \mathbb{C} $$ DnCNN provides an efficient and effective approach for MRI denoising. This is the first application of non-blind ℂ $$ \mathbb{C} $$ DnCNN to medical imaging. The method holds the potential to enable improved low-field MRI, facilitating enhanced diagnostic imaging in under-resourced areas.

Background: Clinical practice guidelines recommend adjuvant therapy for patients with early nonsmall cell lung cancer (eNSCLC), especially those with lymph node metastasis. This study evaluated the prevalence of lymph node examination and its association with adjuvant treatment rates, overall survival (OS), and healthcare costs among United States (US) Medicare patients with resected eNSCLC. Methods: This retrospective observational cohort study used Surveillance, Epidemiology, and End Results cancer registry data linked with Medicare claims data. Eligible patients were aged >= 65 years with newly diagnosed non -small cell lung cancer (NSCLC) stages IA to IIIB [the American Joint Committee on Cancer (AJCC) Cancer Staging Manual , 7th edition] between January 2010 and December 2017 with surgery <= 1 month prior to or <= 12 months after diagnosis. Patients were grouped by lymph node examination status: no examination (pNX), examination and no metastasis (pN0), or metastasis staging in N1 (pN1) or N2 (pN2). OS and costs were evaluated by examination status and number of lymph node examined. OS was analyzed using extended Cox proportional hazards models for specific time periods and time interaction with examination status, and adjusted for patient characteristics. Adjusted post -surgical healthcare costs per patient per month (PPPM) were analyzed using gamma -log regression models. Results: Among the 14,648 patients included in the study, approximately 11% were pNX, whereas most were pN0 (68%), followed by pN1 (11%) and pN2 (10%). Adjuvant treatment rates were higher for pNX (35%) than pN0 (18%), but lower than pN1 (68%) and pN2 (74%) patients (P<0.001). Unadjusted OS for pNX patients was nearly identical to pN2, and significantly worse compared to pN0 and pN1 (P<0.0001). After adjusting for patient characteristics, pNX patients had higher risk of death relative to pN0 patients (P<0.001). Marginal mean adjusted total costs were comparable across pNX ($15,827 PPPM), pN0 ($12,712 PPPM) and pN1 ($17,089 PPPM), but significantly less for pN0 compared to pN2 ($23,566 PPPM) (P=0.002). Conclusions: Inadequate lymph node examination is associated with underutilization of adjuvant treatment and poor OS in resected NSCLC. In the current era of targeted and immunotherapies, lymph node examination is more important than ever, implicating the need for Quality Improvement practices and multidisciplinary coordination.

Cardiac magnetic resonance (CMR) plays an important role in clinically assessing cardiovascular diseases. However, CMR is inherently slow leading to patient discomfort and degraded image quality. Compared with parallel imaging (PI) and compressed sensing (CS), deep-learning-based methods have demonstrated superior image reconstruction performance, in terms of image quality and substantially reduced reconstruction times from highly undersampled CMR data. In this work, we proposed a novel complex-valued cascading cross-domain convolutional neural network, dubbed “C3-Net”, for improved image quality for accelerated CMR. C3-Net outperformed L1-ESPIRiT reconstruction, a baseline U-Net, and a real-valued cascading cross-domain CNN, especially with high acceleration factors (>8). The short-axis results from C3-Net showed reduced residual artifacts and improved temporal fidelity of cardiac motion. In long-axis results, C3-Net excelled in mitigating artifacts surrounding the heart wall and adipose regions in 2-chamber and 4-chamber views, while in a 3-chamber view all the listed methods resulted in suboptimal performance compared to the reference. The quantitative assessment indicated results consistent with assessment of the reconstructed images.

The cerebrospinal fluid (CSF), primarily generated by the choroid plexus (ChP), is the major carrier of the glymphatic system. The alternations of CSF production and the ChP can be associated with the Alzheimer’s disease (AD). The present work investigated the roles of the ChP in the AD based on a proposed ChP image segmentation pipeline. A human-in-the-loop ChP image segmentation pipeline was implemented with intermediate and active learning datasets. The performance of the proposed pipeline was evaluated on manual contours by five radiologists, compared to the FreeSurfer and FastSurfer toolboxes. The ChP volume and blood flow were investigated among AD groups. The correlations between the ChP volume and AD CSF biomarkers including phosphorylated tau (p-tau), total tau (t-tau), amyloid-β42 (Aβ42), and amyloid-β40 (Aβ40) was investigated using three models (univariate, multiple variables, and stepwise regression) on two datasets with 806 and 320 subjects. The proposed ChP segmentation pipeline achieved superior performance with a Dice coefficient of 0.620 on the test dataset, compared to the FreeSurfer (0.342) and FastSurfer (0.371). Significantly larger volumes (p < 0.001) and higher perfusion (p = 0.032) at the ChP were found in AD compared to CN groups. Significant correlations were found between the tau and the relative ChP volume (the ChP volume and ChP/parenchyma ratio) in each patient groups and in the univariate regression analysis (p < 0.001), the multiple regression model (p < 0.05 except for the t-tau in the LMCI), and in the step-wise regression model (p < 0.021). In addition, the correlation coefficients changed from − 0.32 to − 0.21 along with the AD progression in the multiple regression model. In contrast, the Aβ42 and Aβ40 shows consistent and significant associations with the lateral ventricle related measures in the step-wise regression model (p < 0.027). The proposed pipeline provided accurate ChP segmentation which revealed the associations between the ChP and tau level in the AD. The proposed pipeline is available on GitHub (https://github.com/princeleeee/ChP-Seg).

Rationale and Objectives: The current clinical standard for functional imaging of patients with lung ailments is nuclear medicine scintigraphy and Single Photon Emission Computed Tomography (SPECT) which detect the gamma decay of inhaled radioactive tracers. Hyperpolarized (HP) Xenon-129 MRI (XeMRI) of the lungs has recently been FDA approved and provides similar functional images of the lungs with higher spatial resolution than scintigraphy and SPECT. Here we compare Technetium-99m (Tc-99m) diethylene-triamine-pentaacetate scintigraphy and SPECT with HP XeMRI in healthy controls, asthma, and chronic obstructive pulmonary disorder (COPD) patients. Materials and Methods: 59 subjects, healthy, with asthma, and with COPD, underwent Tc-99m scintigraphy/SPECT, standard spirometry, and HP XeMRI. XeMRI and SPECT images were registered for direct voxel-wise signal comparisons. Images were also compared using ventilation defect percentage (VDP), and a standard 6-compartment method. VDP calculated from XeMRI and SPECT images was compared to spirometry. Results: Median Pearson correlation coefficient for voxel-wise signal comparison was 0.698 (0.613-0.782) between scintigraphy and XeMRI and 0.398 (0.286-0.502) between SPECT and XeMRI. Correlation between VDP measures was r = 0.853, p < 0.05. VDP separated asthma and COPD from the control group and was significantly correlated with FEV1, FEV1/FVC, and FEF 25-75. Conclusion: HP XeMRI provides equivalent information to Tc-99m SPECT and standard spirometry measures. Additionally, XeMRI is non-invasive, hence it could be used for longitudinal studies for evaluating emerging treatment for lung ailments.

To develop two spiral-based bSSFP pulse sequences combined with L + S reconstruction for accelerated ungated, free-breathing dynamic cardiac imaging at 1.5 T. Tiny golden angle rotated spiral-out and spiral-in/out bSSFP sequences combined with view-sharing (VS), compressed sensing (CS), and low-rank plus sparse (L + S) reconstruction were evaluated and compared via simulation and in vivo dynamic cardiac imaging studies. The proposed methods were then validated against the standard cine, in terms of quantitative image assessment and qualitative quality rating. The L + S method yielded the least residual artifacts and the best image sharpness among the three methods. Both spiral cine techniques showed clinically diagnostic images (score > 3). Compared to standard cine, there were significant differences in global image quality and edge sharpness for spiral cine techniques, while there was significant difference in image contrast for the spiral-out cine but no significant difference for the spiral-in/out cine. There was good agreement in left ventricular ejection fraction for both the spiral-out cine (− 1.6 ± 3.1 ± 2.8

This paper hypothesizes that a combination of long-duration, retraced spiral sampling strategy and an energy-based variation of multi-frequency reconstruction will improve MR thermometry spatial coverage and precision for MR-guided focused ultrasound applications. We develop this thermometry method through MRI performance characterization, focused ultrasound sonications in phantoms, and MRI thermometry scans in healthy volunteers.

OBJECTIVE:A coupling bath of circulating, chilled, degassed water is essential to safe and precise acoustic transmittance during transcranial magnetic resonance-guided focused ultrasound (tMRgFUS) procedures, but the circulating water impairs the critical real-time magnetic resonance imaging (MRI). An iron-based coupling medium (IBCM) using iron oxide nanoparticles previously developed by our group increased the relaxivity of the coupling bath such that it appears to be invisible on MRI compared with degassed water. However, the nanoparticles also reduced the pressure threshold for cavitation. To address this concern for prefocal cavitation, our group recently developed an IBCM of electrosterically stabilized and aggregation-resistant poly(methacrylic acid)-coated iron oxide nanoparticles (PMAA-FeOX) with a similar capability to reduce the MR signal of degassed water. This study examines the effect of the PMAA-FeOX IBCM on the cavitation threshold.METHODS:Increasing concentrations of PMAA-FeOX nanoparticles in degassed, deionized water were placed at the focus of two different transducers to assess low and high duty-cycle pulsing parameters which are representative of two modes of focused ultrasound being investigated for tMRgFUS. Passive cavitation detection and high-speed optical imaging were used to measure cavitation threshold pressures.RESULTS:The mean cavitation threshold was determined in both cases to be indistinguishable from the degassed water control, between 6-8 MPa for high duty-cycle pulsing (CW) and between 25.5-26.5 MPa for very low duty-cycle pulsing.CONCLUSION:The findings of this study indicate that an IBCM of PMAA-FeOX nanoparticles is a possible solution to reducing MRI interference from the coupling bath without increasing the risk of prefocal cavitation.

Background: The distal superficial femoral artery (SFA) is most commonly affected in peripheral artery disease (PAD). The effects of the proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor alirocumab added to statin therapy on SFA atherosclerosis, downstream flow, and walking performance are unknown. Methods: Thirty-five patients with PAD on maximally tolerated statin therapy were recruited. Patients were randomized to alirocumab 150 mg subcutaneously ( n = 18) or matching placebo ( n = 17) therapy every 2 weeks for 1 year. The primary outcome was change in SFA plaque volume by black blood magnetic resonance imaging (MRI). Secondary outcomes were changes in calf muscle perfusion by cuff/occlusion hyperemia arterial spin labeling MRI, 6-minute walk distance (6MWD), low-density lipoprotein (LDL) cholesterol, and other biomarkers. Results: Age (mean ± SD) was 64 ± 8 years, 20 (57%) patients were women, 17 (49%) were Black individuals, LDL was 107 ± 36 mg/dL, and the ankle–brachial index 0.71 ± 0.20. The LDL fell more with alirocumab than placebo (mean [95% CI]) (−49.8 [−66.1 to −33.6] vs −7.7 [−19.7 to 4.3] mg/dL; p < 0.0001). Changes in SFA plaque volume and calf perfusion showed no difference between groups when adjusted for baseline (+0.25 [−0.29 to 0.79] vs −0.04 [−0.47 to 0.38] cm 3 ; p = 0.37 and 0.22 [−8.67 to 9.11] vs 3.81 [−1.45 to 9.08] mL/min/100 g; p = 0.46, respectively), nor did 6MWD. Conclusion: In this exploratory study, the addition of alirocumab therapy to statins did not alter SFA plaque volume, calf perfusion or 6MWD despite significant LDL lowering. Larger studies with longer follow up that include plaque characterization may improve understanding of the effects of intensive LDL-lowering therapy in PAD (ClinicalTrials.gov Identifier: NCT02959047).

Purpose To develop 2D turbo spin‐echo (TSE) imaging using annular spiral rings (abbreviated “SPRING‐RIO TSE”) with compensation of concomitant gradient fields and B 0 inhomogeneity at both 0.55T and 1.5T for fast T 2 ‐weighted imaging. Methods Strategies of gradient waveform modifications were implemented in SPRING‐RIO TSE for compensation of self‐squared concomitant gradient terms at the TE and across echo spacings, along with reconstruction‐based corrections to simultaneously compensate for the residual concomitant gradient and B 0 field induced phase accruals along the readout. The signal pathway disturbance caused by time‐varying and spatially dependent concomitant fields was simulated, and echo‐to‐echo phase variations before and after sequence‐based compensation were compared. Images from SPRING‐RIO TSE with no compensation, with compensation, and Cartesian TSE were also compared via phantom and in vivo acquisitions. Results Simulation showed how concomitant fields affected the signal evolution with no compensation, and both simulation and phantom studies demonstrated the performance of the proposed sequence modifications, as well as the readout off‐resonance corrections. Volunteer data showed that after full correction, the SPRING‐RIO TSE sequence achieved high image quality with improved SNR efficiency (15%–20% increase), and reduced RF SAR (˜50% reduction), compared to the standard Cartesian TSE, presenting potential benefits, especially in regaining SNR at low‐field (0.55T). Conclusion Implementation of SPRING‐RIO TSE with concomitant field compensation was tested at 0.55T and 1.5T. The compensation principles can be extended to correct for other trajectory types that are time‐varying along the echo train and temporally asymmetric in TSE‐based imaging.

Pulmonary hypertension (PH) is being diagnosed with increasing frequency in modern cohorts. 1 Wijeratne DT Lajkosz K Brogly SB Lougheed MD Jiang L Housin A Barber D Johnson A Doliszny KM Archer SL. Increasing incidence and prevalence of World Health Organization Groups 1 to 4 pulmonary hypertension: a population-based cohort study in Ontario. Circ Cardiovasc Qual Outcomes. 2018; 11 (Canada.)e003973 Crossref PubMed Scopus (159) Google Scholar The clinical classification of PH consists of 5 groups based on the pathophysiological mechanisms, including (1) group 1, pulmonary arterial hypertension (PAH); (2) group 2, PH associated with left heart disease (i.e., heart failure); (3) group 3, PH associated with lung diseases or hypoxia (i.e., chronic obstructive pulmonary disease); (4) group 4, PH associated with pulmonary artery obstructions (i.e., chronic thromboembolic PH); and (5) group 5, PH with unclear or multifactorial mechanisms. 2 Simonneau G Montani D Celermajer DS Denton CP Gatzoulis MA Krowka M Williams PG Souza R. Haemodynamic definitions and updated clinical classification of pulmonary hypertension. Eur Respir J. 2019; 531801913 Crossref Scopus (2287) Google Scholar

Aim: Biomarker testing detects actionable driver mutations to inform first-line treatment in advanced non-small-cell lung cancer (aNSCLC) and metastatic colorectal cancer (mCRC). This study evaluated biomarker testing in a nationwide database (NAT) versus the OneOncology (OneOnc) community network. Patients & methods: Patients with aNSCLC or mCRC with ≥1 biomarker test in a de-identified electronic health record-derived database were evaluated. OneOnc oncologists were surveyed. Results: Biomarker testing rates were high and comparable between OneOnc and NAT; next-generation sequencing (NGS) rates were higher at OneOnc. Patients with NGS versus other biomarker testing were more likely to receive targeted treatment. Operational challenges and insufficient tissue were barriers to NGS testing. Conclusion: Community cancer centers delivered personalized healthcare through biomarker testing.

Current FDA-approved transcranial MR-guided focused ultrasound (tcMRgFUS) transducers cause a curved dark band in 3 T brain images that runs through midbrain targets of ablative treatments for essential tremor and other applications, and signal is reduced by at least 25% elsewhere in the brain. This limits the set of scans that can be performed to guide and assess the effects of treatment. An electromagnetic simulation study was performed to elucidate the mechanisms causing the dark band. Based on the results, a pair of passive antennas in a "propellerbeanie" configuration were designed to manipulate the reflected waves to avoid signal cancellation within the brain. The antennas were optimized and validated with in-vivo experiments and hydrophone measurements. The simulation study revealed that the dark band is caused by RF waves reflected from the transducer's ground plane, which cancel with incoming waves from the scanner's body coil. The passive antennas shifted the dark band out of the brain and increased transmit efficiency in the center of brain 2.3 times while improving field homogeneity by 50%. They also increased receive sensitivity and SNR in anatomic and temperature imaging. They caused no detectable distortion in hydrophone-measured focal pressure profiles. The conductive ground planes and coupling media used in tcMRgFUS and other piezoelectric FUS transducers interact with a 3 T scanner's RF fields to reduce transmit efficiency and SNR. For tcMRgFUS scenario, "propeller beanie" passive reflecting antennas alleviated these effects. This could make a broader set of imaging sequences available to guide tcMRgFUS treatment.

INTRODUCTION:Athletes use their skeletal muscles to demonstrate performance. Muscle force generating capacity is correlated with volume, meaning that variations in sizes of different muscles may be indicative of how athletes meet different demands in their sports. Medical imaging enables in vivo quantification of muscle volumes; however, muscle volume distribution has not been compared across athletes of different sports. PURPOSE:The goal of this work was to define "muscular phenotypes" in athletes of different sports and compare these using hierarchical clustering. METHODS:Muscle volumes normalized by body mass of athletes (football, baseball, basketball, or track) were compared with control participants to quantify size differences using z -scores. z -Scores of 35 muscles described the pattern of volume deviation within each athlete's lower limb, characterizing their muscular phenotype. Data-driven high-dimensional clustering analysis was used to group athletes presenting similar phenotypes. Efficacy of clustering to identify similar phenotypes was demonstrated by grouping athletes' contralateral limbs before other athletes' limbs. RESULTS:Analyses revealed that athletes did not tend to cluster with others competing in the same sport. Basketball players with similar phenotypes grouped by clustering also demonstrated similarities in performance. Clustering also identified muscles with similar volume variation patterns across athletes, and principal component analysis revealed specific muscles that accounted for most of the variance (gluteus maximus, sartorius, semitendinosus, vastus medialis, vastus lateralis, and rectus femoris). CONCLUSIONS:Athletes exhibit heterogeneous lower limb muscle volumes that can be characterized and compared as individual muscular phenotypes. Clustering revealed that athletes with the most similar phenotypes do not always play the same sport such that patterns of muscular heterogeneity across a group of athletes reflect factors beyond their specific sports.

The objective of the current study was to develop and evaluate a DEep learning-based rapid Spiral Image REconstruction (DESIRE) and deep learning (DL)-based segmentation approach to quantify the left ventricular ejection fraction (LVEF) for high-resolution spiral real-time cine imaging, including 2D balanced steady-state free precession imaging at 1.5 T and gradient echo (GRE) imaging at 1.5 and 3 T. A 3D U-Net-based image reconstruction network and 2D U-Net-based image segmentation network were proposed and evaluated. Low-rank plus sparse (L+S) served as the reference for the image reconstruction network and manual contouring of the left ventricle was the reference of the segmentation network. To assess the image reconstruction quality, structural similarity index, peak signal-to-noise ratio, normalized root-mean-square error, and blind grading by two experienced cardiologists (5: excellent; 1: poor) were performed. To assess the segmentation performance, quantification of the LVEF on GRE imaging at 3 T was compared with the quantification from manual contouring. Excellent performance was demonstrated by the proposed technique. In terms of image quality, there was no difference between L+S and the proposed DESIRE technique. For quantification analysis, the proposed DL method was not different to the manual segmentation method (p > 0.05) in terms of quantification of LVEF. The reconstruction time for DESIRE was ~32 s (including nonuniform fast Fourier transform [NUFFT]) per dynamic series (40 frames), while the reconstruction time of L+S with GPU acceleration was approximately 3 min. The DL segmentation takes less than 5 s. In conclusion, the proposed DL-based image reconstruction and quantification techniques enabled 1-min image reconstruction for the whole heart and quantification with automatic reconstruction and quantification of the left ventricle function for high-resolution spiral real-time cine imaging with excellent performance.