Image reconstruction is essential for imaging applications across the physical and life sciences, including optical and radar systems, magnetic resonance imaging, X-ray computed tomography, positron ...emission tomography, ultrasound imaging and radio astronomy. During image acquisition, the sensor encodes an intermediate representation of an object in the sensor domain, which is subsequently reconstructed into an image by an inversion of the encoding function. Image reconstruction is challenging because analytic knowledge of the exact inverse transform may not exist a priori, especially in the presence of sensor non-idealities and noise. Thus, the standard reconstruction approach involves approximating the inverse function with multiple ad hoc stages in a signal processing chain, the composition of which depends on the details of each acquisition strategy, and often requires expert parameter tuning to optimize reconstruction performance. Here we present a unified framework for image reconstruction-automated transform by manifold approximation (AUTOMAP)-which recasts image reconstruction as a data-driven supervised learning task that allows a mapping between the sensor and the image domain to emerge from an appropriate corpus of training data. We implement AUTOMAP with a deep neural network and exhibit its flexibility in learning reconstruction transforms for various magnetic resonance imaging acquisition strategies, using the same network architecture and hyperparameters. We further demonstrate that manifold learning during training results in sparse representations of domain transforms along low-dimensional data manifolds, and observe superior immunity to noise and a reduction in reconstruction artefacts compared with conventional handcrafted reconstruction methods. In addition to improving the reconstruction performance of existing acquisition methodologies, we anticipate that AUTOMAP and other learned reconstruction approaches will accelerate the development of new acquisition strategies across imaging modalities.
Purpose
Demonstrate a novel fast method for reconstruction of multi‐dimensional MR fingerprinting (MRF) data using deep learning methods.
Methods
A neural network (NN) is defined using the TensorFlow ...framework and trained on simulated MRF data computed with the extended phase graph formalism. The NN reconstruction accuracy for noiseless and noisy data is compared to conventional MRF template matching as a function of training data size and is quantified in simulated numerical brain phantom data and International Society for Magnetic Resonance in Medicine/National Institute of Standards and Technology phantom data measured on 1.5T and 3T scanners with an optimized MRF EPI and MRF fast imaging with steady state precession (FISP) sequences with spiral readout. The utility of the method is demonstrated in a healthy subject in vivo at 1.5T.
Results
Network training required 10 to 74 minutes; once trained, data reconstruction required approximately 10 ms for the MRF EPI and 76 ms for the MRF FISP sequence. Reconstruction of simulated, noiseless brain data using the NN resulted in a RMS error (RMSE) of 2.6 ms for T1 and 1.9 ms for T2. The reconstruction error in the presence of noise was less than 10% for both T1 and T2 for SNR greater than 25 dB. Phantom measurements yielded good agreement (R2 = 0.99/0.99 for MRF EPI T1/T2 and 0.94/0.98 for MRF FISP T1/T2) between the T1 and T2 estimated by the NN and reference values from the International Society for Magnetic Resonance in Medicine/National Institute of Standards and Technology phantom.
Conclusion
Reconstruction of MRF data with a NN is accurate, 300‐ to 5000‐fold faster, and more robust to noise and dictionary undersampling than conventional MRF dictionary‐matching.
Abstract In MR Fingerprinting, the flip angles and repetition times are chosen according to a pseudorandom schedule. In previous work, we have shown that maximizing the discrimination between ...different tissue types by optimizing the acquisition schedule allows reductions in the number of measurements required. The ideal optimization algorithm for this application remains unknown, however. In this work we examine several different optimization algorithms to determine the one best suited for optimizing MR Fingerprinting acquisition schedules.
We introduce a broadly applicable technique to create nuclear spin singlet states in organic molecules and other many-atom systems. We employ a novel pulse sequence to produce a spin-lock induced ...crossing (SLIC) of the spin singlet and triplet energy levels, which enables triplet-singlet polarization transfer and singlet-state preparation. We demonstrate the utility of the SLIC method by producing a long-lived nuclear spin singlet state on two strongly coupled proton pairs in the tripeptide molecule phenylalanine-glycine-glycine dissolved in D(2)O and by using SLIC to measure the J couplings, chemical shift differences, and singlet lifetimes of the proton pairs. We show that SLIC is more efficient at creating nearly equivalent nuclear spin singlet states than previous pulse sequence techniques, especially when triplet-singlet polarization transfer occurs on the same time scale as spin-lattice relaxation.
Magnetic Resonance Imaging (MRI) is unparalleled in its ability to visualize anatomical structure and function non-invasively with high spatial and temporal resolution. Yet to overcome the low ...sensitivity inherent in inductive detection of weakly polarized nuclear spins, the vast majority of clinical MRI scanners employ superconducting magnets producing very high magnetic fields. Commonly found at 1.5-3 tesla (T), these powerful magnets are massive and have very strict infrastructure demands that preclude operation in many environments. MRI scanners are costly to purchase, site, and maintain, with the purchase price approaching $1 M per tesla (T) of magnetic field. We present here a remarkably simple, non-cryogenic approach to high-performance human MRI at ultra-low magnetic field, whereby modern under-sampling strategies are combined with fully-refocused dynamic spin control using steady-state free precession techniques. At 6.5 mT (more than 450 times lower than clinical MRI scanners) we demonstrate (2.5 × 3.5 × 8.5) mm(3) imaging resolution in the living human brain using a simple, open-geometry electromagnet, with 3D image acquisition over the entire brain in 6 minutes. We contend that these practical ultra-low magnetic field implementations of MRI (<10 mT) will complement traditional MRI, providing clinically relevant images and setting new standards for affordable (<$50,000) and robust portable devices.
Purpose
To develop an automated machine‐learning‐based method for the discovery of rapid and quantitative chemical exchange saturation transfer (CEST) MR fingerprinting acquisition and reconstruction ...protocols.
Methods
An MR physics‐governed AI system was trained to generate optimized acquisition schedules and the corresponding quantitative reconstruction neural network. The system (termed AutoCEST) is composed of a CEST saturation block, a spin dynamics module, and a deep reconstruction network, all differentiable and jointly connected. The method was validated using a variety of chemical exchange phantoms and in vivo mouse brains at 9.4T.
Results
The acquisition times for AutoCEST optimized schedules ranged from 35 to 71 s, with a quantitative image reconstruction time of only 29 ms. The resulting exchangeable proton concentration maps for the phantoms were in good agreement with the known solute concentrations for AutoCEST sequences (mean absolute error = 2.42 mM; Pearson’s r=0.992, p<0.0001), but not for an unoptimized sequence (mean absolute error = 65.19 mM; Pearson’s r=‐0.161, p=0.522). Similarly, improved exchange rate agreement was observed between AutoCEST and quantification of exchange using saturation power (QUESP) methods (mean absolute error: 35.8 Hz, Pearson’s r=0.971, p<0.0001) compared to an unoptimized schedule and QUESP (mean absolute error = 58.2 Hz; Pearson’s r=0.959, p<0.0001). The AutoCEST in vivo mouse brain semi‐solid proton volume fractions were lower in the cortex (12.77% ± 0.75%) compared to the white matter (19.80% ± 0.50%), as expected.
Conclusion
AutoCEST can automatically generate optimized CEST/MT acquisition protocols that can be rapidly reconstructed into quantitative exchange parameter maps.
Summary Background Selective BCL2 inhibition with venetoclax has substantial activity in patients with relapsed or refractory chronic lymphocytic leukaemia. Combination therapy with rituximab ...enhanced activity in preclinical models. The aim of this study was to assess the safety, pharmacokinetics, and activity of venetoclax in combination with rituximab. Methods Adult patients with relapsed or refractory chronic lymphocytic leukaemia (according to the 2008 Modified International Workshop on CLL guidelines) or small lymphocytic lymphoma were eligible for this phase 1b, dose-escalation trial. The primary outcomes were to assess the safety profile, to determine the maximum tolerated dose, and to establish the recommended phase 2 dose of venetoclax when given in combination with rituximab. Secondary outcomes were to assess the pharmacokinetic profile and analyse efficacy, including overall response, duration of response, and time to tumour progression. Minimal residual disease was a protocol-specified exploratory objective. Central review of the endpoints was not done. Venetoclax was dosed daily using a stepwise escalation to target doses (200–600 mg) and then monthly rituximab commenced (375 mg/m2 in month 1 and 500 mg/m2 in months 2–6). Adverse events were graded according to the National Cancer Institute Common Terminology Criteria for adverse events version 4.0. Protocol-guided drug cessation was allowed for patients who achieved complete response (including complete response with incomplete marrow recovery) or negative bone marrow minimal residual disease. Analyses were done per protocol for all patients who commenced drug and included all patients who received at least one dose of venetoclax. Data were pooled across dose cohorts. Patients are still receiving therapy and follow-up is ongoing. The trial is registered at ClinicalTrials.gov , number NCT01682616. Findings Between Aug 6, 2012, and May 28, 2014, we enrolled 49 patients. Common grade 1–2 toxicities included upper respiratory tract infections (in 28 57% of 49 patients), diarrhoea (27 55%), and nausea (25 51%). Grade 3–4 adverse events occurred in 37 (76%) of 49 patients; most common were neutropenia (26 53%), thrombocytopenia (eight 16%), anaemia (seven 14%), febrile neutropenia (six 12%), and leucopenia (six 12%). The most common serious adverse events were pyrexia (six 12%), febrile neutropenia (five 10%), lower respiratory tract infection, and pneumonia (each three 6%). Clinical tumour lysis syndrome occurred in two patients (resulting in one death) who initiated venetoclax at 50 mg. After enhancing tumour lysis syndrome prophylaxis measures and commencing venetoclax at 20 mg, clinical tumour lysis syndrome did not occur. The maximum tolerated dose was not identified; the recommended phase 2 dose of venetoclax in combination with rituximab was 400 mg. Overall, 42 (86%) of 49 patients achieved a response, including a complete response in 25 (51%) of 49 patients. 2 year estimates for progression-free survival and ongoing response were 82% (95% CI 66–91) and 89% (95% CI 72–96), respectively. Negative marrow minimal residual disease was attained in 20 (80%) of 25 complete responders and 28 (57%) of 49 patients overall. 13 responders ceased all therapy; among these all 11 minimal residual disease-negative responders remain progression-free off therapy. Two with minimal residual disease-positive complete response progressed after 24 months off therapy and re-attained response after re-initiation of venetoclax. Interpretation A substantial proportion of patients achieved an overall response with the combination of venetoclax and rituximab including 25 (51%) of 49 patients who achieved a complete response and 28 (57%) of 49 patients who achieved negative marrow minimal residual disease with acceptable safety. The depth and durability of responses observed with the combination offers an attractive potential treatment option for patients with relapsed or refractory chronic lymphocytic leukaemia and could allow some patients to maintain response after discontinuing therapy, a strategy that warrants further investigation in randomised studies. Funding AbbVie Inc and Genentech Inc.
Purpose
To develop a fast magnetic resonance fingerprinting (MRF) method for quantitative chemical exchange saturation transfer (CEST) imaging.
Methods
We implemented a CEST‐MRF method to quantify ...the chemical exchange rate and volume fraction of the Nα‐amine protons of L‐arginine (L‐Arg) phantoms and the amide and semi‐solid exchangeable protons of in vivo rat brain tissue. L‐Arg phantoms were made with different concentrations (25–100 mM) and pH (pH 4–6). The MRF acquisition schedule varied the saturation power randomly for 30 iterations (phantom: 0–6 μT; in vivo: 0–4 μT) with a total acquisition time of ≤2 min. The signal trajectories were pattern‐matched to a large dictionary of signal trajectories simulated using the Bloch‐McConnell equations for different combinations of exchange rate, exchangeable proton volume fraction, and water T1 and T2 relaxation times.
Results
The chemical exchange rates of the Nα‐amine protons of L‐Arg were significantly (P < 0.0001) correlated with the rates measured with the quantitation of exchange using saturation power method. Similarly, the L‐Arg concentrations determined using MRF were significantly (P < 0.0001) correlated with the known concentrations. The pH dependence of the exchange rate was well fit (R2 = 0.9186) by a base catalyzed exchange model. The amide proton exchange rate measured in rat brain cortex (34.8 ± 11.7 Hz) was in good agreement with that measured previously with the water exchange spectroscopy method (28.6 ± 7.4 Hz). The semi‐solid proton volume fraction was elevated in white (12.2 ± 1.7%) compared to gray (8.1 ± 1.1%) matter brain regions in agreement with previous magnetization transfer studies.
Conclusion
CEST‐MRF provides a method for fast, quantitative CEST imaging.
Objective: The use of a close-fitting roughly head-shaped volume coil for MRI (magnetic resonance imaging) has the merit of improving the filling factor and thus the SNR (signal-to-noise ratio) from ...the brain. However, the surface of the RF coil follows that of the head which makes it difficult to determine an optimal coil winding pattern. We describe here a new method to optimize a head-shaped RF coil with the objective of maximizing its SNR and RF-magnetic-field homogeneity for operation at ultra-low magnetic field (6.5 mT, 276 kHz). Methods: The approach consists of FEM (finite-element-method) simulation and linear programing based optimization. Results: We have implemented the optimization and further studied the relationship between the design requirements and the performance of the RF coil. Finally, we constructed an optimal RF coil and scanned both a head-shaped phantom and a human subject. Conclusion: The method we outline here provide new insight into the conductor layout needed for magnetic optimization of structurally complex coils, especially when tradeoffs between competing attributes (SNR and homogeneity in this case) must be made.
IMPORTANCE: Neuroimaging is a key step in the clinical evaluation of brain injury. Conventional magnetic resonance imaging (MRI) systems operate at high-strength magnetic fields (1.5-3 T) that ...require strict, access-controlled environments. Limited access to timely neuroimaging remains a key structural barrier to effectively monitor the occurrence and progression of neurological injury in intensive care settings. Recent advances in low-field MRI technology have allowed for the acquisition of clinically meaningful imaging outside of radiology suites and in the presence of ferromagnetic materials at the bedside. OBJECTIVE: To perform an assessment of brain injury in critically ill patients in intensive care unit settings, using a portable, low-field MRI device at the bedside. DESIGN, SETTING, AND PARTICIPANTS: This was a prospective, single-center cohort study of 50 patients admitted to the neuroscience or coronavirus disease 2019 (COVID-19) intensive care units at Yale New Haven Hospital in New Haven, Connecticut, from October 30, 2019, to May 20, 2020. Patients were eligible if they presented with neurological injury or alteration, no contraindications for conventional MRI, and a body habitus not exceeding the scanner’s 30-cm vertical opening. Diagnosis of COVID-19 was determined by positive severe acute respiratory syndrome coronavirus 2 polymerase chain reaction nasopharyngeal swab result. EXPOSURES: Portable MRI in an intensive care unit room. MAIN OUTCOMES AND MEASURES: Demographic, clinical, radiological, and treatment data were collected and analyzed. Brain imaging findings are described. RESULTS: Point-of-care MRI examinations were performed on 50 patients (16 women 32%; mean SD age, 59 12 years range, 20-89 years). Patients presented with ischemic stroke (n = 9), hemorrhagic stroke (n = 12), subarachnoid hemorrhage (n = 2), traumatic brain injury (n = 3), brain tumor (n = 4), and COVID-19 with altered mental status (n = 20). Examinations were acquired at a median of 5 (range, 0-37) days after intensive care unit admission. Diagnostic-grade T1-weighted, T2-weighted, T2 fluid-attenuated inversion recovery, and diffusion-weighted imaging sequences were obtained for 37, 48, 45, and 32 patients, respectively. Neuroimaging findings were detected in 29 of 30 patients who did not have COVID-19 (97%), and 8 of 20 patients with COVID-19 (40%) demonstrated abnormalities. There were no adverse events or complications during deployment of the portable MRI or scanning in an intensive care unit room. CONCLUSIONS AND RELEVANCE: This single-center series of patients with critical illness in an intensive care setting demonstrated the feasibility of low-field, portable MRI. These findings demonstrate the potential role of portable MRI to obtain neuroimaging in complex clinical care settings.