scholarly journals Accelerated cardiac T1 mapping in four heartbeats with inline MyoMapNet: a deep learning-based T1 estimation approach

2022 ◽  
Vol 24 (1) ◽  
Author(s):  
Rui Guo ◽  
Hossam El-Rewaidy ◽  
Salah Assana ◽  
Xiaoying Cai ◽  
Amine Amyar ◽  
...  
Keyword(s):  
Neural Networks ◽  
T1 Mapping ◽  
Inversion Recovery ◽  
Inversion Pulse ◽  
Mapping Data ◽  
Post Contrast ◽  
Recovery Times ◽  
Similar Accuracy ◽  
Fully Connected

Abstract Purpose To develop and evaluate MyoMapNet, a rapid myocardial T1 mapping approach that uses fully connected neural networks (FCNN) to estimate T1 values from four T1-weighted images collected after a single inversion pulse in four heartbeats (Look-Locker, LL4). Method We implemented an FCNN for MyoMapNet to estimate T1 values from a reduced number of T1-weighted images and corresponding inversion-recovery times. We studied MyoMapNet performance when trained using native, post-contrast T1, or a combination of both. We also explored the effects of number of T1-weighted images (four and five) for native T1. After rigorous training using in-vivo modified Look-Locker inversion recovery (MOLLI) T1 mapping data of 607 patients, MyoMapNet performance was evaluated using MOLLI T1 data from 61 patients by discarding the additional T1-weighted images. Subsequently, we implemented a prototype MyoMapNet and LL4 on a 3 T scanner. LL4 was used to collect T1 mapping data in 27 subjects with inline T1 map reconstruction by MyoMapNet. The resulting T1 values were compared to MOLLI. Results MyoMapNet trained using a combination of native and post-contrast T1-weighted images had excellent native and post-contrast T1 accuracy compared to MOLLI. The FCNN model using four T1-weighted images yields similar performance compared to five T1-weighted images, suggesting that four T1 weighted images may be sufficient. The inline implementation of LL4 and MyoMapNet enables successful acquisition and reconstruction of T1 maps on the scanner. Native and post-contrast myocardium T1 by MOLLI and MyoMapNet was 1170 ± 55 ms vs. 1183 ± 57 ms (P = 0.03), and 645 ± 26 ms vs. 630 ± 30 ms (P = 0.60), and native and post-contrast blood T1 was 1820 ± 29 ms vs. 1854 ± 34 ms (P = 0.14), and 508 ± 9 ms vs. 514 ± 15 ms (P = 0.02), respectively. Conclusion A FCNN, trained using MOLLI data, can estimate T1 values from only four T1-weighted images. MyoMapNet enables myocardial T1 mapping in four heartbeats with similar accuracy as MOLLI with inline map reconstruction.

Real‐time Triggered RAdial Single‐Shot Inversion recovery for arrhythmia‐insensitive myocardial T1 mapping: motion phantom validation and in vivo comparison

2018 ◽  
Vol 81 (3) ◽  
pp. 1714-1725 ◽  
Author(s):  
Daniel Gensler ◽  
Tim Salinger ◽  
Markus Düring ◽  
Kristina Lorenz ◽  
Roland Jahns ◽  
...  
Keyword(s):  
Real Time ◽  
T1 Mapping ◽  
Inversion Recovery ◽  
Single Shot ◽  
Myocardial T1 ◽  
Motion Phantom ◽  
Phantom Validation

scholarly journals Fast Interleaved Multislice T1 Mapping: Model-Based Reconstruction of Single-Shot Inversion-Recovery Radial FLASH

2018 ◽  
Vol 2018 ◽  
pp. 1-8 ◽  
Author(s):  
Xiaoqing Wang ◽  
Dirk Voit ◽  
Volkert Roeloffs ◽  
Martin Uecker ◽  
Jens Frahm
Keyword(s):  
High Speed ◽  
T1 Mapping ◽  
Inversion Recovery ◽  
Inversion Pulse ◽  
Breath Hold ◽  
Single Shot ◽  
Radial Sampling ◽  
Model Based ◽  
Inverse Reconstruction ◽  
Space Data

Purpose. To develop a high-speed multislice T1 mapping method based on a single-shot inversion-recovery (IR) radial FLASH acquisition and a regularized model-based reconstruction. Methods. Multislice radial k-space data are continuously acquired after a single nonselective inversion pulse using a golden-angle sampling scheme in a spoke-interleaved manner with optimized flip angles. Parameter maps and coil sensitivities of each slice are estimated directly from highly undersampled radial k-space data using a model-based nonlinear inverse reconstruction in conjunction with joint sparsity constraints. The performance of the method has been validated using a numerical and experimental T1 phantom as well as demonstrated for studies of the human brain and liver at 3T. Results. The proposed method allows for 7 simultaneous T1 maps of the brain at 0.5 × 0.5 × 4 mm3 resolution within a single IR experiment of 4 s duration. Phantom studies confirm similar accuracy and precision as obtained for a single-slice acquisition. For abdominal applications, the proposed method yields three simultaneous T1 maps at 1.25 × 1.25 × 6 mm3 resolution within a 4 s breath hold. Conclusion. Rapid, robust, accurate, and precise multislice T1 mapping may be achieved by combining the advantages of a model-based nonlinear inverse reconstruction, radial sampling, parallel imaging, and compressed sensing.

scholarly journals Relationship between training database size and structural connectivity in artificial neural networks: Do specific local connections improve prediction accuracy?

2021 ◽  
Author(s):  
Min Sik Park
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Prediction Accuracy ◽  
Structural Connectivity ◽  
Large Size ◽  
Artificial Neural ◽  
Neuronal Systems ◽  
Similar Accuracy ◽  
Fully Connected

Abstract Training of artificial neural networks is very expensive, as a large-size database is necessary. Moreover, it is usually difficult to find such large-size training databases. Hence, it will be interesting to design artificial neural networks that can be used for training with a small-size database, while maintaining a similar accuracy for prediction compared to fully connected neural networks. We studied neural networks with partial disconnections, additional bypass connections, and negative activation nodes, which are found in the neuronal systems of the human brain. By combining the fully connected neural network and the above three brain-like elements, we found that the modified neural network showed improved prediction accuracy of 13% compared to the fully connected one despite the small size of the training database. To analyze the improved neural network, the contribution of each node in the hidden layers affecting the total prediction accuracy of the neural networks was studied. We also found important local connections that improve the prediction accuracy, and discussed the design of a neural network with a small-size training database without reduction in prediction accuracy.

scholarly journals Fast calculation software for modified Look-Locker inversion recovery (MOLLI) T1 mapping

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Yoon-Chul Kim ◽  
Khu Rai Kim ◽  
Hyelee Lee ◽  
Yeon Hyeon Choe
Keyword(s):  
T1 Mapping ◽  
Volume Fraction ◽  
Extracellular Volume ◽  
Computation Time ◽  
Software Tool ◽  
Extracellular Volume Fraction ◽  
Inversion Recovery ◽  
Least Square ◽  
Post Contrast ◽  
T1 Map

Abstract Background The purpose of this study was to develop a software tool and evaluate different T1 map calculation methods in terms of computation time in cardiac magnetic resonance imaging. Methods The modified Look-Locker inversion recovery (MOLLI) sequence was used to acquire multiple inversion time (TI) images for pre- and post-contrast T1 mapping. The T1 map calculation involved pixel-wise curve fitting based on the T1 relaxation model. A variety of methods were evaluated using data from 30 subjects for computational efficiency: MRmap, python Levenberg–Marquardt (LM), python reduced-dimension (RD) non-linear least square, C++ single- and multi-core LM, and C++ single- and multi-core RD. Results Median (interquartile range) computation time was 126 s (98–141) for the publicly available software MRmap, 261 s (249–282) for python LM, 77 s (74–80) for python RD, 3.4 s (3.1–3.6) for C++ multi-core LM, and 1.9 s (1.9–2.0) for C++ multi-core RD. The fastest C++ multi-core RD and the publicly available MRmap showed good agreement of myocardial T1 values, resulting in 95% Bland–Altman limits of agreement of (− 0.83 to 0.58 ms) and (− 6.57 to 7.36 ms) with mean differences of − 0.13 ms and 0.39 ms, for the pre- and post-contrast, respectively. Conclusion The C++ multi-core RD was the fastest method on a regular eight-core personal computer for pre- or post-contrast T1 map calculation. The presented software tool (fT1fit) facilitated rapid T1 map and extracellular volume fraction map calculations.

scholarly journals Separable Fully Connected Layers Improve Deep Learning Models For Genomics

2017 ◽  
Author(s):  
Amr Mohamed Alexandari ◽  
Avanti Shrikumar ◽  
Anshul Kundaje
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Biologically Inspired ◽  
Outer Product ◽  
Link Type ◽  
Current State ◽  
Conventional Models ◽  
Model Setup ◽  
Fully Connected

ABSTRACTConvolutional neural networks are rapidly gaining popularity in regulatory genomics. Typically, these networks have a stack of convolutional and pooling layers, followed by one or more fully connected layers. In genomics, the same positional patterns are often present across multiple convolutional channels. Therefore, in current state-of-the-art networks, there exists significant redundancy in the representations learned by standard fully connected layers. We present a new separable fully connected layer that learns a weights tensor that is the outer product of positional weights and cross-channel weights, thereby allowing the same positional patterns to be applied across multiple convolutional channels. Decomposing positional and cross-channel weights further enables us to readily impose biologically-inspired constraints on positional weights, such as symmetry. We also propose a novel regularizer and constraint that act on curvature in the positional weights. Using experiments on simulated and in vivo datasets, we show that networks that incorporate our separable fully connected layer outperform conventional models with analogous architectures and the same number of parameters. Additionally, our networks are more robust to hyperparameter tuning, have more informative gradients, and produce importance scores that are more consistent with known biology than conventional deep neural networks.AvailabilityImplementation: https://github.com/kundajelab/keras/tree/keras_1A gist illustrating model setup is at: goo.gl/gYooaa

scholarly journals T1 mapping-derived signature of myocardial involvement in idiopathic inflammatory myopathy compared to acute viral myocarditis: a texture-based analysis

2021 ◽  
Vol 22 (Supplement_1) ◽  
Author(s):  
E Bollache ◽  
AT Huber ◽  
J Lamy ◽  
E Afari ◽  
TM Bacoyannis ◽  
...  
Keyword(s):  
Contrast Agent ◽  
Texture Analysis ◽  
T1 Mapping ◽  
Inflammatory Myopathy ◽  
Viral Myocarditis ◽  
Texture Features ◽  
Idiopathic Inflammatory Myopathy ◽  
Post Contrast ◽  
Myocardial T1 ◽  
Myocardial Involvement

Abstract Funding Acknowledgements Type of funding sources: None. Background. Recent studies revealed the ability of MRI T1 mapping to characterize myocardial involvement in both idiopathic inflammatory myopathy (IIM) and acute viral myocarditis (AVM), as compared to healthy controls. However, neither myocardial T1 nor T2 maps were able to discriminate between IIM and AVM patients, when considering conventional myocardial mean values and derived indices such as lambda and extracellular volume. Purpose. To investigate the ability of T1 mapping-derived texture analysis to differentiate IIM from AVM. Methods. Forty patients, 20 with IIM (51 ± 17 years, 9 men) and 20 with AVM (34 ± 13 years, 16 men) underwent 1.5T MRI T1 mapping using a modified Look-Locker inversion-recovery sequence before and 15 minutes after injection of a gadolinium contrast agent. After manual delineation of endocardial and epicardial borders and co-registration of all inversion time images, native and post-contrast T1 maps were estimated. Myocardial texture analysis was performed on native T1 maps. Textural features such as: autocorrelation, contrast, dissimilarity, energy and sum entropy were used to build a least squares-based linear regression model. Finally, receiver operating characteristic (ROC) analysis was used to investigate the ability of such texture features score to classify IIM vs. AVM patients, compared to the performance of mean myocardial T1. A Wilcoxon rank-sum test was also used to test difference significance between groups. Results. Both native and post-contrast mean myocardial T1 values were comparable between IIM (native: 1022 ± 43 ms; post-contrast: 319 ± 44 ms) and AVM (1056 ± 59 ms, p = 0.07; 318 ± 35 ms, p = 0.90, respectively) groups. Results of ROC analyses are provided in the Table, indicating that a better discrimination between IIM and AVM patients was obtained when using texture features, with higher AUC and accuracy than mean T1 values (Figure). Conclusion. Texture analysis derived from MRI T1 maps without contrast agent injection was able to discriminate between IIM and AVM with higher accuracy, sensitivity and specificity than conventional T1 indices. Such analysis could provide a useful myocardial signature to help diagnose and manage cardiac alterations associated with IIM in patients presenting with myocarditis and primarily suspected of AVM. Table Area under curve (AUC) Accuracy Sensitivity Specificity Native T1 0.67 0.70 0.65 0.75 Post-contrast T1 0.49 0.60 0.25 0.95 Texture features score 0.85 0.82 0.90 0.75 ROC analyses for classification between IIM and AVM patients Abstract Figure

scholarly journals Predicting Absorption-Distribution Properties of Neuroprotective Phosphine-Borane Compounds Using In Silico Modeling and Machine Learning

Molecules ◽  
2021 ◽  
Vol 26 (9) ◽  
pp. 2505
Author(s):  
Raheem Remtulla ◽  
Sanjoy Kumar Das ◽  
Leonard A. Levin
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
In Silico ◽  
Disulfide Bonds ◽  
Oral Absorption ◽  
In Vivo Studies ◽  
Drug Candidates ◽  
In Silico Methods

Phosphine-borane complexes are novel chemical entities with preclinical efficacy in neuronal and ophthalmic disease models. In vitro and in vivo studies showed that the metabolites of these compounds are capable of cleaving disulfide bonds implicated in the downstream effects of axonal injury. A difficulty in using standard in silico methods for studying these drugs is that most computational tools are not designed for borane-containing compounds. Using in silico and machine learning methodologies, the absorption-distribution properties of these unique compounds were assessed. Features examined with in silico methods included cellular permeability, octanol-water partition coefficient, blood-brain barrier permeability, oral absorption and serum protein binding. The resultant neural networks demonstrated an appropriate level of accuracy and were comparable to existing in silico methodologies. Specifically, they were able to reliably predict pharmacokinetic features of known boron-containing compounds. These methods predicted that phosphine-borane compounds and their metabolites meet the necessary pharmacokinetic features for orally active drug candidates. This study showed that the combination of standard in silico predictive and machine learning models with neural networks is effective in predicting pharmacokinetic features of novel boron-containing compounds as neuroprotective drugs.

Sign in / Sign up

Export Citation Format

Share Document