Accelerated cardiac T1 mapping with recurrent networks and cyclic, model-based loss.

BACKGROUND: Using the spin-lattice relaxation time (T1) as a biomarker, the myocardium can be quantitatively characterized using cardiac T1 mapping. The modified Look-Locker inversion (MOLLI) recovery sequences have become the standard clinical method for cardiac T1 mapping. However, the MOLLI sequences require an 11-heartbeat breath-hold that can be difficult for subjects, particularly during exercise or pharmacologically induced stress. Although shorter cardiac T1 mapping sequences have been proposed, these methods suffer from reduced precision. As such, there is an unmet need for accelerated cardiac T1 mapping. PURPOSE: To accelerate cardiac T1 mapping MOLLI sequences by using neural networks to estimate T1 maps using a reduced number of T1-weighted images and their corresponding inversion times. MATERIALS AND METHODS: In this retrospective study, 911 pre-contrast T1 mapping datasets from 202 subjects (128 males, 56 ± 15 years; 74 females, 54 ± 17 years) and 574 T1 mapping post-contrast datasets from 193 subjects (122 males, 57 ± 15 years; 71 females, 54 ± 17 years) were acquired using the MOLLI-5(3)3 sequence and the MOLLI-4(1)3(1)2 sequence, respectively. All acquisition protocols used similar scan parameters: T R = 2.2 ms $TR\; = \;2.2\;{\rm{ms}}$ , T E = 1.12 ms $TE\; = \;1.12\;{\rm{ms}}$ , and F A = 35 ∘ $FA\; = \;35^\circ $ , gadoteridol (ProHance, Bracco Diagnostics) dose ∼ 0.075 mmol / kg $\sim 0.075\;\;{\rm{mmol/kg}}$ . A bidirectional multilayered long short-term memory (LSTM) network with fully connected output and cyclic model-based loss was used to estimate T1 maps from the first three T1-weighted images and their corresponding inversion times for pre- and post-contrast T1 mapping. The performance of the proposed architecture was compared to the three-parameter T1 recovery model using the same reduction of the number of T1-weighted images and inversion times. Reference T1 maps were generated from the scanner using the full MOLLI sequences and the three-parameter T1 recovery model. Correlation and Bland-Altman plots were used to evaluate network performance in which each point represents averaged regions of interest in the myocardium corresponding to the standard American Heart Association 16-segment model. The precision of the network was examined using consecutively repeated scans. Stress and rest pre-contrast MOLLI studies as well as various disease test cases, including amyloidosis, hypertrophic cardiomyopathy, and sarcoidosis were also examined. Paired t-tests were used to determine statistical significance with p < 0.05 $p < 0.05$ . RESULTS: Our proposed network demonstrated similar T1 estimations to the standard MOLLI sequences (pre-contrast: 1260 ± 94 ms $1260 \pm 94\;{\rm{ms}}$ vs. 1254 ± 91 ms $1254 \pm 91\;{\rm{ms}}$ with p = 0.13 $p\; = \;0.13$ ; post-contrast: 484 ± 92 ms $484 \pm 92\;{\rm{ms}}$ vs. 493 ± 91 ms $493 \pm 91\;{\rm{ms}}$ with p = 0.07 $p\; = \;0.07$ ). The precision of standard MOLLI sequences was well preserved with the proposed network architecture ( 24 ± 28 ms $24 \pm 28\;\;{\rm{ms}}$ vs. 18 ± 13 ms $18 \pm 13\;{\rm{ms}}$ ). Network-generated T1 reactivities are similar to stress and rest pre-contrast MOLLI studies ( 5.1 ± 4.0 % $5.1 \pm 4.0\;\% $ vs. 4.9 ± 4.4 % $4.9 \pm 4.4\;\% $ with p = 0.84 $p\; = \;0.84$ ). Amyloidosis T1 maps generated using the proposed network are also similar to the reference T1 maps (pre-contrast: 1243 ± 140 ms $1243 \pm 140\;\;{\rm{ms}}$ vs. 1231 ± 137 ms $1231 \pm 137\;{\rm{ms}}$ with p = 0.60 $p\; = \;0.60$ ; post-contrast: 348 ± 26 ms $348 \pm 26\;{\rm{ms}}$ vs. 346 ± 27 ms $346 \pm 27\;{\rm{ms}}$ with p = 0.89 $p\; = \;0.89$ ). CONCLUSIONS: A bidirectional multilayered LSTM network with fully connected output and cyclic model-based loss was used to generate high-quality pre- and post-contrast T1 maps using the first three T1-weighted images and their corresponding inversion times. This work demonstrates that combining deep learning with cardiac T1 mapping can potentially accelerate standard MOLLI sequences from 11 to 3 heartbeats.

Deep learning for radial SMS myocardial perfusion reconstruction using the 3D residual booster U-net.

PURPOSE: To develop an end-to-end deep learning solution for quickly reconstructing radial simultaneous multi-slice (SMS) myocardial perfusion datasets with comparable quality to the pixel tracking spatiotemporal constrained reconstruction (PT-STCR) method. METHODS: Dynamic contrast enhanced (DCE) radial SMS myocardial perfusion data were obtained from 20 subjects who were scanned at rest and/or stress with or without ECG gating using a saturation recovery radial CAIPI turboFLASH sequence. Input to the networks consisted of complex coil combined images reconstructed using the inverse Fourier transform of undersampled radial SMS k-space data. Ground truth images were reconstructed using the PT-STCR pipeline. The performance of the residual booster 3D U-Net was tested by comparing it to state-of-the-art network architectures including MoDL, CRNN-MRI, and other U-Net variants. RESULTS: Results demonstrate significant improvements in speed requiring approximately 8 seconds to reconstruct one radial SMS dataset which is approximately 200 times faster than the PT-STCR method. Images reconstructed with the residual booster 3D U-Net retain quality of ground truth PT-STCR images (0.963 SSIM/40.238 PSNR/0.147 NRMSE). The residual booster 3D U-Net has superior performance compared to existing network architectures in terms of image quality, temporal dynamics, and reconstruction time. CONCLUSION: Residual and booster learning combined with the 3D U-Net architecture was shown to be an effective network for reconstructing high-quality images from undersampled radial SMS datasets while bypassing the reconstruction time of the PT-STCR method.

Late Gadolinium Enhancement Magnetic Resonance Imaging Evaluation of Post-Atrial Fibrillation Ablation Esophageal Thermal Injury Across the Spectrum of Severity.

Background Esophageal thermal injury (ETI) is a byproduct of atrial fibrillation (AF) ablation using thermal sources. The most severe form of ETI is represented by atrioesophageal fistula, which has a high mortality rate. Late gadolinium enhancement (LGE) magnetic resonance imaging (MRI) allows identification of ETI. Hence, we sought to evaluate the utility of LGE-MRI as a method to identify ETI across the entire spectrum of severity. Methods and Results All AF radiofrequency ablations performed at the University of Utah between January 2009 and December 2017 were reviewed. Patients with LGE-MRI within 24 hours following AF ablation as well as patients who had esophagogastroduodenoscopy in addition to LGE-MRI were identified. An additional patient with atrioesophageal fistula who had AF ablation at a different institution and had MRI and esophagogastroduodenoscopy at the University of Utah was identified. A total of 1269 AF radiofrequency ablations were identified. ETI severity was classified on the basis of esophageal LGE pattern (none, 60.9%; mild, 27.5%; moderate, 9.9%; severe, 1.7%). ETI resolved in most patients who underwent repeat LGE-MRI at 3 months. All patients with esophagogastroduodenoscopy-confirmed ETI had moderate-to-severe LGE 24 hours after ablation MRI. Moderate-to-severe LGE had 100% sensitivity and 58.1% specificity in detecting ETI, and a negative predictive value of 100%. Atrioesophageal fistula was visualized by both computed tomography and LGE-MRI in one patient. Conclusions LGE-MRI is useful in detecting and characterizing ETI across the entire severity spectrum. LGE-MRI exhibits an extremely high sensitivity and negative predictive value in screening for ETI after AF ablation.

Preventing or Minimizing Acute Kidney Injury in Patients Undergoing Transcatheter Aortic Valve Replacement.

BACKGROUND: Transcatheter aortic valve implantation (TAVI) is now routinely performed in patients with aortic stenosis with low mortality and complication rates. Although periprocedural risks have been substantially minimized, procedure- and contrast-induced acute kidney injury (AKI) remains a major concern. AKI remains a frequent complication of contrast-guided interventional procedures and is associated with a significantly adverse prognosis. We review the currently available clinical data related to AKI, with emphasis on contrast-induced nephropathy (CIN), and discuss a novel, integrated approach aiming to minimize AKI risk in high-risk patients. A stepwise algorithm is also proposed for the management of these complex patients.

Accelerated 3D Left Atrial Late Gadolinium Enhancement in Patients with Atrial Fibrillation at 1.5 T: Technical Development.

PURPOSE: To develop an accelerated three-dimensional (3D) late gadolinium enhancement (LGE) pulse sequence using balanced steady-state free precession readout with stack-of-stars k-space sampling and extra motion-state golden-angle radial sparse parallel (XD-GRASP) reconstruction and test the performance for detecting atrial scar and fibrosis in patients with atrial fibrillation (AF). MATERIALS AND METHODS: Twenty-five patients with AF (20 paroxysmal and five persistent; 65 years ± 7 [standard deviation]; 18 men) were imaged at 1.5 T using the proposed LGE sequence with 1.3 mm × 1.3 mm × 2-mm spatial resolution and predictable imaging time. The resulting images were compared with historic images of 25 patients with AF (18 paroxysmal and seven persistent; 67 years ± 10; 14 men) obtained using a reference 3D left atrial (LA) LGE sequence with 1.3 mm × 1.3 mm × 2.5-mm spatial resolution. Two readers visually graded the 3D LGE images (conspicuity, artifact, noise) on a five-point Likert scale (1 = worst, 3 = acceptable, 5 = best), in which the summed visual score (SVS) of 9 or greater was defined as clinically acceptable. Appropriate statistical analyses (Cohen κ coefficient, Mann-Whitney U test, t tests, and intraclass correlation) were performed, where a P value < .05 was considered significant. RESULTS: Mean imaging time was significantly shorter (P < .01) for the proposed pulse sequence (5.9 minutes ± 1.3) than for the reference pulse sequence (10.6 minutes ± 2). Median SVS was significantly higher (P < .01) for the proposed (SVS = 11) than reference (SVS = 9.5) 3D LA LGE images. Interrater reproducibility in visual scores was higher for the proposed (κ = 0.78-1) than reference 3D LA LGE (κ = 0.44-0.75). Intrareader repeatability in fibrosis quantification was higher for the reference cohort (intraclass correlation coefficient [ICC] = 0.94) than the prospective cohort (ICC = 0.79). CONCLUSION: The proposed 3D LA LGE method produced clinically acceptable image quality with 1.5 mm × 1.5 mm × 2-mm nominal spatial resolution and 6-minute predictable imaging time for quantification of LA scar and fibrosis in patients with AF. Supplemental material is available for this article. © RSNA, 2020.

Feasibility of multiple-view myocardial perfusion MRI using radial simultaneous multi-slice acquisitions.

PURPOSE: Dynamic contrast enhanced MRI of the heart typically acquires 2-4 short-axis (SA) slices to detect and characterize coronary artery disease. This acquisition scheme is limited by incomplete coverage of the left ventricle. We studied the feasibility of using radial simultaneous multi-slice (SMS) technique to achieve SA, 2-chamber and/or 4-chamber long-axis (2CH LA and/or 4CH LA) coverage with and without electrocardiography (ECG) gating using a motion-robust reconstruction framework. METHODS: 12 subjects were scanned at rest and/or stress, free breathing, with or without ECG gating. Multiple sets of radial SMS k-space were acquired within each cardiac cycle, and each SMS set sampled 3 parallel slices that were either SA, 2CH LA, or 4CH LA slices. The radial data was interpolated onto Cartesian space using an SMS GRAPPA operator gridding method. Self-gating and respiratory states binning of the data were done. The binning information as well as a pixel tracking spatiotemporal constrained reconstruction method were applied to obtain motion-robust image reconstructions. Reconstructions with and without the pixel tracking method were compared for signal-to-noise ratio and contrast-to-noise ratio. RESULTS: Full coverage of the heart (at least 3 SA and 3 LA slices) during the first pass of contrast at every heartbeat was achieved by using the radial SMS acquisition. The proposed pixel tracking reconstruction improves the average SNR and CNR by 21% and 30% respectively, and reduces temporal blurring for both gated and ungated acquisitions. CONCLUSION: Acquiring simultaneous multi-slice SA, 2CH LA and/or 4CH LA myocardial perfusion images in every heartbeat is feasible in both gated and ungated acquisitions. This can add confidence when detecting and characterizing coronary artery disease by revealing ischemia in different views, and by providing apical coverage that is improved relative to SA slices alone. The proposed pixel tracking framework improves the reconstruction while adding little computational cost.

Design for occupational safety and health of workers in construction in developing countries: a study of architects in Nigeria.

PURPOSE: Design for safety (DfS) of workers is amongst the prominent ways of tackling poor occupational safety and health performance in construction. However, in developing countries there is extremely limited research on DfS. This study thus makes an important contribution to the subject of DfS in developing countries by specifically examining the awareness and practice of DfS amongst architects within the construction sector of Nigeria. MATERIALS AND METHODS: A survey of architects, yielding 161 valid responses, was conducted. RESULTS: While there is high awareness of the concept of DfS, the actual practice is low. Additionally, although there is high interest in DfS training, the engagement in DfS training is low. Significantly, awareness of DfS, training and education related to DfS, and membership of a design professional body have very limited bearing on the practice of DfS by architects. CONCLUSIONS: The findings are thus symptomatic of the prevalence of influential DfS implementation barriers within the construction sector. Industry stakeholders should seek to raise the profile of DfS practice within the sector. Furthermore, similar empirical studies in the construction sector of other developing countries would be useful in shedding light on the status of DfS in these countries.