Unraveling contributions to the Z-spectrum signal at 3.5 ppm of human brain tumors.

PURPOSE: To evaluate the influence of the confounding factors, direct water saturation (DWS), and magnetization transfer contrast (MTC) effects on measured Z-spectra and amide proton transfer (APT) contrast in brain tumors. METHODS: High-grade glioma patients were scanned using an RF saturation-encoded 3D MR fingerprinting (MRF) sequence at 3 T. For MRF reconstruction, a recurrent neural network was designed to learn free water and semisolid macromolecule parameter mappings of the underlying multiple tissue properties from saturation-transfer MRF signals. The DWS spectra and MTC spectra were synthesized by solving Bloch-McConnell equations and evaluated in brain tumors. RESULTS: The dominant contribution to the saturation effect at 3.5 ppm was from DWS and MTC effects, but 25%-33% of the saturated signal in the gadolinium-enhancing tumor (13%-20% for normal tissue) was due to the APT effect. The APT# signal of the gadolinium-enhancing tumor was significantly higher than that of the normal-appearing white matter (10.1% vs. 8.3% at 1 µT and 11.2% vs. 7.8% at 1.5 µT). CONCLUSION: The RF saturation-encoded MRF allowed us to separate contributions to the saturation signal at 3.5 ppm in the Z-spectrum. Although free water and semisolid MTC are the main contributors, significant APT contrast between tumor and normal tissues was observed.

Fast and motion-robust saturation transfer MRI with inherent B0 correction using rosette trajectories and compressed sensing.

PURPOSE: To implement rosette readout trajectories with compressed sensing reconstruction for fast and motion-robust CEST and magnetization transfer contrast imaging with inherent correction of B0 inhomogeneity. METHODS: A pulse sequence was developed for fast saturation transfer imaging using a stack of rosette trajectories with a higher sampling density near the k-space center. Each rosette lobe was segmented into two halves to generate dual-echo images. B0 inhomogeneities were estimated using the phase difference between the images and corrected subsequently. The rosette-based imaging was evaluated in comparison to a fully sampled Cartesian trajectory and demonstrated on CEST phantoms (creatine solutions and egg white) and healthy volunteers at 3 T. RESULTS: Compared with the conventional Cartesian acquisition, compressed sensing reconstructed rosette images provided image quality with overall higher contrast-to-noise ratio and significantly faster readout time. Accurate B0 map estimation was achieved from the rosette acquisition with a negligible bias of 0.01 Hz between the rosette and dual-echo Cartesian gradient echo B0 maps, using the latter as ground truth. The water-saturation spectra (Z-spectra) and amide proton transfer weighted signals obtained from the rosette-based sequence were well preserved compared with the fully sampled data, both in the phantom and human studies. CONCLUSIONS: Fast, motion-robust, and inherent B0-corrected CEST and magnetization transfer contrast imaging using rosette trajectories could improve subject comfort and compliance, contrast-to-noise ratio, and provide inherent B0 homogeneity information. This work is expected to significantly accelerate the translation of CEST-MRI into a robust, clinically viable approach.

Self-supervised learning for denoising of multidimensional MRI data.

PURPOSE: To develop a fast denoising framework for high-dimensional MRI data based on a self-supervised learning scheme, which does not require ground truth clean image. THEORY AND METHODS: Quantitative MRI faces limitations in SNR, because the variation of signal amplitude in a large set of images is the key mechanism for quantification. In addition, the complex non-linear signal models make the fitting process vulnerable to noise. To address these issues, we propose a fast deep-learning framework for denoising, which efficiently exploits the redundancy in multidimensional MRI data. A self-supervised model was designed to use only noisy images for training, bypassing the challenge of clean data paucity in clinical practice. For validation, we used two different datasets of simulated magnetization transfer contrast MR fingerprinting (MTC-MRF) dataset and in vivo DWI image dataset to show the generalizability. RESULTS: The proposed method drastically improved denoising performance in the presence of mild-to-severe noise regardless of noise distributions compared to previous methods of the BM3D, tMPPCA, and Patch2self. The improvements were even pronounced in the following quantification results from the denoised images. CONCLUSION: The proposed MD-S2S (Multidimensional-Self2Self) denoising technique could be further applied to various multi-dimensional MRI data and improve the quantification accuracy of tissue parameter maps.

Whole-cerebrum guanidino and amide CEST mapping at 3 T by a 3D stack-of-spirals gradient echo acquisition.

PURPOSE: To develop a 3D, high-sensitivity CEST mapping technique based on the 3D stack-of-spirals (SOS) gradient echo readout, the proposed approach was compared with conventional acquisition techniques and evaluated for its efficacy in concurrently mapping of guanidino (Guan) and amide CEST in human brain at 3 T, leveraging the polynomial Lorentzian line-shape fitting (PLOF) method. METHODS: Saturation time and recovery delay were optimized to achieve maximum CEST time efficiency. The 3DSOS method was compared with segmented 3D EPI (3DEPI), turbo spin echo, and gradient- and spin-echo techniques. Image quality, temporal SNR (tSNR), and test-retest reliability were assessed. Maps of Guan and amide CEST derived from 3DSOS were demonstrated on a low-grade glioma patient. RESULTS: The optimized recovery delay/saturation time was determined to be 1.4/2 s for Guan and amide CEST. In addition to nearly doubling the slice number, the gradient echo techniques also outperformed spin echo sequences in tSNR: 3DEPI (193.8 ± 6.6), 3DSOS (173.9 ± 5.6), and GRASE (141.0 ± 2.7). 3DSOS, compared with 3DEPI, demonstrated comparable GuanCEST signal in gray matter (GM) (3DSOS: [2.14%-2.59%] vs. 3DEPI: [2.15%-2.61%]), and white matter (WM) (3DSOS: [1.49%-2.11%] vs. 3DEPI: [1.64%-2.09%]). 3DSOS also achieves significantly higher amideCEST in both GM (3DSOS: [2.29%-3.00%] vs. 3DEPI: [2.06%-2.92%]) and WM (3DSOS: [2.23%-2.66%] vs. 3DEPI: [1.95%-2.57%]). 3DSOS outperforms 3DEPI in terms of scan-rescan reliability (correlation coefficient: 3DSOS: 0.58-0.96 vs. 3DEPI: -0.02 to 0.75) and robustness to motion as well. CONCLUSION: The 3DSOS CEST technique shows promise for whole-cerebrum CEST imaging, offering uniform contrast and robustness against motion artifacts.

Compartmental anisotropy of filtered exchange imaging (FEXI) in human white matter: What is happening in FEXI?

PURPOSE: To investigate the effects of compartmental anisotropy on filtered exchange imaging (FEXI) in white matter (WM). THEORY AND METHODS: FEXI signals were measured using multiple combinations of diffusion filter and detection directions in five healthy volunteers. Additional filters, including a trace-weighted diffusion filter with trapezoidal gradients, a spherical b-tensor encoded diffusion filter, and a T2 filter, were tested with trace-weighted diffusion detection. RESULTS: A large range of apparent exchange rates (AXR) and both positive and negative filter efficiencies (σ) were found depending on the mutual orientation of the filter and detection gradients relative to WM fiber orientation. The data demonstrated that the fast-diffusion compartment suppressed by diffusional filtering is not exclusively extra-cellular, but also intra-cellular. While not comprehensive, a simple two-compartment diffusion tensor model with water exchange was able to account qualitatively for the trends in positive and negative filtering efficiencies, while standard model imaging (SMI) without exchange could not. This two-compartment diffusion tensor model also demonstrated smaller AXR variances across subjects. When employing trace-weighted diffusion detection, AXR values were on the order of the R1 (=1/T1) of water at 3T for crossing fibers, while being less than R1 for parallel fibers. CONCLUSION: Orientation-dependent AXR and σ values were observed when using multi-orientation filter and detection gradients in FEXI, indicating that WM FEXI models need to account for compartmental anisotropy. When using trace-weighted detection, AXR values were on the order of or less than R1, complicating the interpretation of FEXI results in WM in terms of biological exchange properties. These findings may contribute toward better understanding of FEXI results in WM.

Predictors of mTICI 2c/3 over 2b in patients successfully recanalized with mechanical thrombectomy.

OBJECTIVE: For patients presenting with acute ischemic stroke (AIS) caused by large vessel occlusions (LVO), mechanical thrombectomy (MT) is the treatment standard of care in eligible patients. Modified Thrombolysis in Cerebral Infarction (mTICI) grades of 2b, 2c, and 3 are all considered successful reperfusion; however, recent studies have shown achieving mTICI 2c/3 leads to better outcomes than mTICI 2b. This study aims to investigate whether any baseline preprocedural or periprocedural parameters are predictive of achieving mTICI 2c/3 in successfully recanalized LVO patients. METHODS: We conducted a retrospective multicenter cohort study of consecutive patients presenting with AIS caused by a LVO from 1 January 2017 to 1 January 2023. Baseline and procedural data were collected through chart review. Univariate and multivariate analysis were applied to determine significant predictors of mTICI 2c/3. RESULTS: A total of 216 patients were included in the study, with 159 (73.6%) achieving mTICI 2c/3 recanalization and 57 (26.4%) achieving mTICI 2b recanalization. We found that a higher groin puncture to first pass time (OR = 0.976, 95%CI: 0.960-0.992, p = 0.004), a higher first pass to recanalization time (OR = 0.985, 95%CI: 0.972-0.998, p = 0.029), a higher admission NIHSS (OR = 0.949, 95%CI: 0.904-0.995, p = 0.031), and a lower age (OR = 1.032, 95%CI: 1.01-1.055, p = 0.005) were associated with a decreased probability of achieving mTICI 2c/3. INTERPRETATION: A lower groin puncture to first pass time, a lower first pass to recanalization time, a lower admission NIHSS, and a higher age were independent predictors of mTICI 2c/3 recanalization.

CEST and nuclear Overhauser enhancement imaging with deep learning-extrapolated semisolid magnetization transfer reference: Scan-rescan reproducibility and reliability studies.

PURPOSE: To develop a novel MR physics-driven, deep-learning, extrapolated semisolid magnetization transfer reference (DeepEMR) framework to provide fast, reliable magnetization transfer contrast (MTC) and CEST signal estimations, and to determine the reproducibility and reliability of the estimates from the DeepEMR. METHODS: A neural network was designed to predict a direct water saturation and MTC-dominated signal at a certain CEST frequency offset using a few high-frequency offset features in the Z-spectrum. The accuracy, scan-rescan reproducibility, and reliability of MTC, CEST, and relayed nuclear Overhauser enhancement (rNOE) signals estimated from the DeepEMR were evaluated on numerical phantoms and in heathy volunteers at 3 T. In addition, we applied the DeepEMR method to brain tumor patients and compared tissue contrast with other CEST calculation metrics. RESULTS: The DeepEMR method demonstrated a high degree of accuracy in the estimation of reference MTC signals at ±3.5 ppm for APT and rNOE imaging, and computational efficiency (˜190-fold) compared with a conventional fitting approach. In addition, the DeepEMR method achieved high reproducibility and reliability (intraclass correlation coefficient = 0.97, intersubject coefficient of variation = 3.5%, and intrasubject coefficient of variation = 1.3%) of the estimation of MTC signals at ±3.5 ppm. In tumor patients, DeepEMR-based amide proton transfer images provided higher tumor contrast than a conventional MT ratio asymmetry image, particularly at higher B1 strengths (>1.5 µT), with a distinct delineation of the tumor core from normal tissue or peritumoral edema. CONCLUSION: The DeepEMR approach is feasible for measuring clean APT and rNOE effects in longitudinal and cross-sectional studies with low scan-rescan variability.

Bloch simulator-driven deep recurrent neural network for magnetization transfer contrast MR fingerprinting and CEST imaging.

PURPOSE: To develop a unified deep-learning framework by combining an ultrafast Bloch simulator and a semisolid macromolecular magnetization transfer contrast (MTC) MR fingerprinting (MRF) reconstruction for estimation of MTC effects. METHODS: The Bloch simulator and MRF reconstruction architectures were designed with recurrent neural networks and convolutional neural networks, evaluated with numerical phantoms with known ground truths and cross-linked bovine serum albumin phantoms, and demonstrated in the brain of healthy volunteers at 3 T. In addition, the inherent magnetization-transfer ratio asymmetry effect was evaluated in MTC-MRF, CEST, and relayed nuclear Overhauser enhancement imaging. A test-retest study was performed to evaluate the repeatability of MTC parameters, CEST, and relayed nuclear Overhauser enhancement signals estimated by the unified deep-learning framework. RESULTS: Compared with a conventional Bloch simulation, the deep Bloch simulator for generation of the MTC-MRF dictionary or a training data set reduced the computation time by 181-fold, without compromising MRF profile accuracy. The recurrent neural network-based MRF reconstruction outperformed existing methods in terms of reconstruction accuracy and noise robustness. Using the proposed MTC-MRF framework for tissue-parameter quantification, the test-retest study showed a high degree of repeatability in which the coefficients of variance were less than 7% for all tissue parameters. CONCLUSION: Bloch simulator-driven, deep-learning MTC-MRF can provide robust and repeatable multiple-tissue parameter quantification in a clinically feasible scan time on a 3T scanner.

CEST2022: Amide proton transfer-weighted MRI improves the diagnostic performance of multiparametric non-contrast-enhanced MRI techniques in patients with post-treatment high-grade gliomas.

New or enlarged lesions in malignant gliomas after surgery and chemoradiation can be associated with tumor recurrence or treatment effect. Due to similar radiographic characteristics, conventional-and even some advanced MRI techniques-are limited in distinguishing these two pathologies. Amide proton transfer-weighted (APTw) MRI, a protein-based molecular imaging technique that does not require the administration of any exogenous contrast agent, was recently introduced into the clinical setting. In this study, we evaluated and compared the diagnostic performances of APTw MRI with several non-contrast-enhanced MRI sequences, such as diffusion-weighted imaging, susceptibility-weighted imaging, and pseudo-continuous arterial spin labeling. Thirty-nine scans from 28 glioma patients were obtained on a 3 T MRI scanner. A histogram analysis approach was employed to extract parameters from each tumor area. Statistically significant parameters (P < 0.05) were selected to train multivariate logistic regression models to evaluate the performance of MRI sequences. Multiple histogram parameters, particularly from APTw and pseudo-continuous arterial spin labeling images, demonstrated significant differences between treatment effect and recurrent tumor. The regression model trained on the combination of all significant histogram parameters achieved the best result (area under the curve = 0.89). We found that APTw images added value to other advanced MR images for the differentiation of treatment effect and tumor recurrence.

Pretreatment brain CT perfusion thresholds for predicting final infarct volume in distal medium vessel occlusions.

BACKGROUND AND PURPOSE: Quantitative CT perfusion (CTP) thresholds for assessing the extent of ischemia in patients with acute ischemic stroke (AIS) have been established; relative cerebral blood flow (rCBF) <30% is typically used for estimating estimated ischemic core volume and Tmax (time to maximum) >6 seconds for critical hypoperfused volume in AIS patients with large vessel occlusion (LVO). In this study, we aimed to identify the optimal threshold values for patients presenting with AIS secondary to distal medium vessel occlusions (DMVOs). METHODS: In this retrospective study, consecutive AIS patients with anterior circulation DMVO who underwent pretreatment CTP and follow-up MRI/CT were included. The CTP data were processed by RAPID (iSchemaView, Menlo Park, CA) to generate estimated ischemic core volumes using rCBF <20%, <30%, <34%, and <38% and critical hypoperfused volumes using Tmax (seconds) >4, >6, >8, and >10. Final infarct volumes (FIVs) were obtained from follow-up MRI/CT within 5 days of symptom onset. Diagnostic performance between CTP thresholds and FIV was assessed in the successfully and unsuccessfully recanalized groups. RESULTS: Fifty-five patients met our inclusion criteria (32 female [58.2%], 68.0 ± 12.1 years old [mean ± SD]). Recanalization was attempted with intravenous tissue-type plasminogen activator and mechanical thrombectomy in 27.7% and 38.1% of patients, respectively. Twenty-five patients (45.4%) were successfully recanalized. In the successfully recanalized patients, no CTP threshold significantly outperformed what is used in LVO setting (rCBF < 30%). All rCBF CTP thresholds demonstrated fair diagnostic performances for predicting FIV. In unsuccessfully recanalized patients, all Tmax CTP thresholds strongly predicted FIV with relative superiority of Tmax >10 seconds (area under the receiver operating characteristic curve = .875, p = .001). CONCLUSION: In AIS patients with DMVOs, longer Tmax delays than Tmax  > 6 seconds, most notably, Tmax  > 10 seconds, best predict FIV in unsuccessfully recanalized patients. No CTP threshold reliably predicts FIV in the successfully recanalized group nor significantly outperformed rCBF < 30%.

Deep learning-based Lorentzian fitting of water saturation shift referencing spectra in MRI.

PURPOSE: Water saturation shift referencing (WASSR) Z-spectra are used commonly for field referencing in chemical exchange saturation transfer (CEST) MRI. However, their analysis using least-squares (LS) Lorentzian fitting is time-consuming and prone to errors because of the unavoidable noise in vivo. A deep learning-based single Lorentzian Fitting Network (sLoFNet) is proposed to overcome these shortcomings. METHODS: A neural network architecture was constructed and its hyperparameters optimized. Training was conducted on a simulated and in vivo-paired data sets of discrete signal values and their corresponding Lorentzian shape parameters. The sLoFNet performance was compared with LS on several WASSR data sets (both simulated and in vivo 3T brain scans). Prediction errors, robustness against noise, effects of sampling density, and time consumption were compared. RESULTS: LS and sLoFNet performed comparably in terms of RMS error and mean absolute error on all in vivo data with no statistically significant difference. Although the LS method fitted well on samples with low noise, its error increased rapidly when increasing sample noise up to 4.5%, whereas the error of sLoFNet increased only marginally. With the reduction of Z-spectral sampling density, prediction errors increased for both methods, but the increase occurred earlier (at 25 vs. 15 frequency points) and was more pronounced for LS. Furthermore, sLoFNet performed, on average, 70 times faster than the LS-method. CONCLUSION: Comparisons between LS and sLoFNet on simulated and in vivo WASSR MRI Z-spectra in terms of robustness against noise and decreased sample resolution, as well as time consumption, showed significant advantages for sLoFNet.

Only-train-once MR fingerprinting for B0 and B1 inhomogeneity correction in quantitative magnetization-transfer contrast.

PURPOSE: To develop a fast, deep-learning approach for quantitative magnetization-transfer contrast (MTC)-MR fingerprinting (MRF) that simultaneously estimates multiple tissue parameters and corrects the effects of B0 and B1 variations. METHODS: An only-train-once recurrent neural network was designed to perform the fast tissue-parameter quantification for a large range of different MRF acquisition schedules. It enabled a dynamic scan-wise linear calibration of the scan parameters using the measured B0 and B1 maps, which allowed accurate, multiple-tissue parameter mapping. MRF images were acquired from 8 healthy volunteers at 3 T. Estimated parameter maps from the MRF images were used to synthesize the MTC reference signal (Zref ) through Bloch equations at multiple saturation power levels. RESULTS: The B0 and B1 errors in MR fingerprints, if not corrected, would impair the tissue quantification and subsequently corrupt the synthesized MTC reference images. Bloch equation-based numerical phantom studies and synthetic MRI analysis demonstrated that the proposed approach could correctly estimate water and semisolid macromolecule parameters, even with severe B0 and B1 inhomogeneities. CONCLUSION: The only-train-once deep-learning framework can improve the reconstruction accuracy of brain-tissue parameter maps and be further combined with any conventional MRF or CEST-MRF method.

Lateralization of temporal lobe epileptic foci with automated chemical exchange saturation transfer measurements at 3 Tesla.

BACKGROUND: Magnetic Resonance Imaging (MRI) is an indispensable tool for the diagnosis of temporal lobe epilepsy (TLE). However, about 30% of TLE patients show no lesion on structural MRI (sMRI-negative), posing a significant challenge for presurgical evaluation. This study aimed to investigate whether chemical exchange saturation transfer (CEST) MRI at 3 Tesla can lateralize the epileptic focus of TLE and study the metabolic contributors to the CEST signal measured. METHODS: Forty TLE subjects (16 males and 24 females) were included in this study. An automated data analysis pipeline was established, including segmentation of the hippocampus and amygdala (HA), calculation of four CEST metrics and quantitative relaxation times (T1 and T2), and construction of prediction models by logistic regression. Furthermore, a modified two-stage Bloch-McConnell fitting method was developed to investigate the molecular imaging mechanism of 3 T CEST in identifying epileptic foci of TLE. FINDINGS: The mean CEST ratio (CESTR) metric within 2.25-3.25 ppm in the HA was the most powerful index in predicting seizure laterality, with an area under the receiver-operating characteristic curve (AUC) of 0.84. And, the combination of T2 and CESTR further increased the AUC to 0.92. Amine and guanidinium moieties were the two leading contributors to the CEST contrast between the epileptogenic HA and the normal HA. INTERPRETATION: CEST at 3 Tesla is a powerful modality that can predict seizure laterality with high accuracy. This study can potentially facilitate the clinical translation of CEST MRI in identifying the epileptic foci of TLE or other localization-related epilepsies. FUNDING: National Natural Science Foundation of China, Science Technology Department of Zhejiang Province, and Zhejiang University.

Amide proton transfer imaging in stroke.

Amide proton transfer (APT) imaging, a variant of chemical exchange saturation transfer MRI, has shown promise in detecting ischemic tissue acidosis following impaired aerobic metabolism in animal models and in human stroke patients due to the sensitivity of the amide proton exchange rate to changes in pH within the physiological range. Recent studies have demonstrated the possibility of using APT-MRI to detect acidosis of the ischemic penumbra, enabling the assessment of stroke severity and risk of progression, monitoring of treatment progress, and prognostication of clinical outcome. This paper reviews current APT imaging methods actively used in ischemic stroke research and explores the clinical aspects of ischemic stroke and future applications for these methods.

MR fingerprinting for semisolid magnetization transfer and chemical exchange saturation transfer quantification.

Chemical exchange saturation transfer (CEST) MRI has positioned itself as a promising contrast mechanism, capable of providing molecular information at sufficient resolution and amplified sensitivity. However, it has not yet become a routinely employed clinical technique, due to a variety of confounding factors affecting its contrast-weighted image interpretation and the inherently long scan time. CEST MR fingerprinting (MRF) is a novel approach for addressing these challenges, allowing simultaneous quantitation of several proton exchange parameters using rapid acquisition schemes. Recently, a number of deep-learning algorithms have been developed to further boost the performance and speed of CEST and semi-solid macromolecule magnetization transfer (MT) MRF. This review article describes the fundamental theory behind semisolid MT/CEST-MRF and its main applications. It then details supervised and unsupervised learning approaches for MRF image reconstruction and describes artificial intelligence (AI)-based pipelines for protocol optimization. Finally, practical considerations are discussed, and future perspectives are given, accompanied by basic demonstration code and data.

Radiomics analysis of amide proton transfer-weighted and structural MR images for treatment response assessment in malignant gliomas.

The purpose of this study was to evaluate the value of amide proton transfer-weighted (APTw) MRI radiomic features for the differentiation of tumor recurrence from treatment effect in malignant gliomas. Eighty-six patients who had suspected tumor recurrence after completion of chemoradiation or radiotherapy, and who had APTw-MRI data acquired at 3 T, were retrospectively analyzed. Using a fluid-attenuated inversion recovery (FLAIR) image-based mask, radiomics analysis was applied to the processed APTw and structural MR images. A chi-square automatic interaction detector decision tree was used for classification analysis. Models with and without APTw features were built using the same strategy. Tenfold cross-validation was applied to obtain the overall classification performance of each model. Sixty patients were confirmed as having tumor recurrence, and the remainder were confirmed as having treatment effect, at median time points of 190 and 171 days after therapy, respectively. There were 525 radiomic features extracted from each of the processed APTw and structural MR images. Based on these, the APTw-based model yielded the highest accuracy (86.0%) for the differentiation of tumor recurrence from treatment effect, compared with 74.4%, 76.7%, 83.7%, and 76.7% for T1 w, T2 w, FLAIR, and Gd-T1 w, respectively. Model classification accuracy was 82.6% when using the combined structural MR images (T1 w, T2 w, FLAIR, Gd-T1 w), and increased to 89.5% when using these structural plus APTw images. The corresponding sensitivity and specificity were 85.0% and 76.9% for the combination of structural MR images, and 85.0% and 100% after adding APTw image features. Adding APTw-based radiomic features increased MRI accuracy in the assessment of the treatment response in post-treatment malignant gliomas.

Learning-based analysis of amide proton transfer-weighted MRI to identify true progression in glioma patients.

The purpose of this study was to develop and verify a convolutional neural network (CNN)-based deep-learning algorithm to identify tumor progression versus response by adding amide proton transfer-weighted (APTw) MRI data to structural MR images as the proposed model input. 145 scans with 2175 MR instances from 98 patients with malignant glioma (acquired between April 2010 and February 2018) were re-analyzed. An end-to-end classification framework based on a ResNet backbone was developed. The architecture includes a learnable subtraction layer and a hierarchical classification paradigm, and synthesizes information over multiple MR slices using a long short-term memory. Areas under the receiver-operating-characteristic curves (AUCs) were used to assess the impact of adding APTw MRI to structural MRI (T1w, T2w, FLAIR, and GdT1w) on classification of tumor response vs. progression, both on the slice- and scan-level. With both APTw and structural MRI data, adding a learnable subtraction layer and a hierarchical classification paradigm to the backbone ResNet model improved the slice-level classification performance from an AUC of 0.85 to 0.90. Adding APTw data to structural MR images as input to our proposed CNN classification framework led to an increase in AUCs from 0.88 to 0.90 for the slice-level classification (P < 0.001), and from 0.85 to 0.90 for the scan-level classification (P < 0.05). Generated saliency maps highlighted the vast majority of lesions. Complementing structural MRI sequences with protein-based APTw MRI enhanced CNN-based classification of recurrent glioma at the slice and scan levels. Addition of APTw MRI to structural MRI sequences enhanced CNN-based classification of recurrent glioma at the slice and scan levels.

Review and consensus recommendations on clinical APT-weighted imaging approaches at 3T: Application to brain tumors.

Amide proton transfer-weighted (APTw) MR imaging shows promise as a biomarker of brain tumor status. Currently used APTw MRI pulse sequences and protocols vary substantially among different institutes, and there are no agreed-on standards in the imaging community. Therefore, the results acquired from different research centers are difficult to compare, which hampers uniform clinical application and interpretation. This paper reviews current clinical APTw imaging approaches and provides a rationale for optimized APTw brain tumor imaging at 3 T, including specific recommendations for pulse sequences, acquisition protocols, and data processing methods. We expect that these consensus recommendations will become the first broadly accepted guidelines for APTw imaging of brain tumors on 3 T MRI systems from different vendors. This will allow more medical centers to use the same or comparable APTw MRI techniques for the detection, characterization, and monitoring of brain tumors, enabling multi-center trials in larger patient cohorts and, ultimately, routine clinical use.

Learning-based optimization of acquisition schedule for magnetization transfer contrast MR fingerprinting.

Magnetization transfer contrast MR fingerprinting (MTC-MRF) is a novel quantitative imaging method that simultaneously quantifies free bulk water and semisolid macromolecule parameters using pseudo-randomized scan parameters. To improve acquisition efficiency and reconstruction accuracy, the optimization of MRF sequence design has been of recent interest in the MRF field, but has been challenging due to the large number of degrees of freedom to be optimized in the sequence. Herein, we propose a framework for learning-based optimization of the acquisition schedule (LOAS), which optimizes RF saturation-encoded MRF acquisitions with a minimal number of scan parameters for tissue parameter determination. In a supervised learning framework, scan parameters were subsequently updated to minimize a predefined loss function that can directly represent tissue quantification errors. We evaluated the performance of the proposed approach with a numerical phantom and in in vivo experiments. For validation, MRF images were synthesized using the tissue parameters estimated from a fully connected neural network framework and compared with references. Our results showed that LOAS outperformed existing indirect optimization methods with regard to quantification accuracy and acquisition efficiency. The proposed LOAS method could be a powerful optimization tool in the design of MRF pulse sequences.