Dataset Information

Rapid 4D-MRI reconstruction using a deep radial convolutional neural network: Dracula.

ABSTRACT:

Background and purpose

4D and midposition MRI could inform plan adaptation in lung and abdominal MR-guided radiotherapy. We present deep learning-based solutions to overcome long 4D-MRI reconstruction times while maintaining high image quality and short scan times.

Methods

Two 3D U-net deep convolutional neural networks were trained to accelerate the 4D joint MoCo-HDTV reconstruction. For the first network, gridded and joint MoCo-HDTV-reconstructed 4D-MRI were used as input and target data, respectively, whereas the second network was trained to directly calculate the midposition image. For both networks, input and target data had dimensions of 256 × 256 voxels (2D) and 16 respiratory phases. Deep learning-based MRI were verified against joint MoCo-HDTV-reconstructed MRI using the structural similarity index (SSIM) and the naturalness image quality evaluator (NIQE). Moreover, two experienced observers contoured the gross tumour volume and scored the images in a blinded study.

Results

For 12 subjects, previously unseen by the networks, high-quality 4D and midposition MRI (1.25 × 1.25 × 3.3 mm³) were each reconstructed from gridded images in only 28 seconds per subject. Excellent agreement was found between deep-learning-based and joint MoCo-HDTV-reconstructed MRI (average SSIM ≥ 0.96, NIQE scores 7.94 and 5.66). Deep-learning-based 4D-MRI were clinically acceptable for target and organ-at-risk delineation. Tumour positions agreed within 0.7 mm on midposition images.

Conclusion

Our results suggest that the joint MoCo-HDTV and midposition algorithms can each be approximated by a deep convolutional neural network. This rapid reconstruction of 4D and midposition MRI facilitates online treatment adaptation in thoracic or abdominal MR-guided radiotherapy.

SUBMITTER: Freedman JN

PROVIDER: S-EPMC8216429 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundRecently, deep learning technologies have rapidly expanded into medical image analysis, including both disease detection and classification. As far as we know, migraine is a disabling and common neurological disorder, typically characterized by unilateral, throbbing and pulsating headaches. Unfortunately, a large number of migraineurs do not receive the accurate diagnosis when using traditional diagnostic criteria based on the guidelines of the International Headache Society. As such, there is substantial interest in developing automated methods to assist in the diagnosis of migraine.MethodsTo the best of our knowledge, no studies have evaluated the potential of deep learning technologies in assisting with the classification of migraine patients. Here, we used deep learning methods in combination with three functional measures (the amplitude of low-frequency fluctuations, regional homogeneity and regional functional correlation strength) based on rs-fMRI data to distinguish not only between migraineurs and healthy controls, but also between the two subtypes of migraine. We employed 21 migraine patients without aura, 15 migraineurs with aura, and 28 healthy controls.ResultsCompared with the traditional support vector machine classifier, which has an accuracy of 83.67%, our Inception module-based convolutional neural network approach showed a significant improvement in classification output (over 86.18%). Our data also indicate that the Inception module-based CNN performs better than the AlexNet-based CNN (Inception module-based CNN reached an accuracy of 99.25%). Finally, we also found that regional functional correlation strength (RFCS) could be regarded as the optimum input out of the three indices (ALFF, ReHo, RFCS).ConclusionsOverall, our study shows that combining the three functional measures of rs-fMRI with deep learning classification is a powerful method to distinguish between migraineurs and healthy individuals. Our data also highlight that deep learning-based frameworks could be used to develop more complicated models or systems to aid in clinical decision making in the future.

Project description:PurposeThe existence of multicontrast magnetic resonance (MR) images increases the level of clinical information available for the diagnosis and treatment of brain cancer patients. However, acquiring the complete set of multicontrast MR images is not always practically feasible. In this study, we developed a state-of-the-art deep learning convolutional neural network (CNN) for image-to-image translation across three standards MRI contrasts for the brain.MethodsBRATS'2018 MRI dataset of 477 patients clinically diagnosed with glioma brain cancer was used in this study, with each patient having T1-weighted (T1), T2-weighted (T2), and FLAIR contrasts. It was randomly split into 64%, 16%, and 20% as training, validation, and test set, respectively. We developed a U-Net model to learn the nonlinear mapping of a source image contrast to a target image contrast across three MRI contrasts. The model was trained and validated with 2D paired MR images using a mean-squared error (MSE) cost function, Adam optimizer with 0.001 learning rate, and 120 epochs with a batch size of 32. The generated synthetic-MR images were evaluated against the ground-truth images by computing the MSE, mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM).ResultsThe generated synthetic-MR images with our model were nearly indistinguishable from the real images on the testing dataset for all translations, except synthetic FLAIR images had slightly lower quality and exhibited loss of details. The range of average PSNR, MSE, MAE, and SSIM values over the six translations were 29.44-33.25 dB, 0.0005-0.0012, 0.0086-0.0149, and 0.932-0.946, respectively. Our results were as good as the best-reported results by other deep learning models on BRATS datasets.ConclusionsOur U-Net model exhibited that it can accurately perform image-to-image translation across brain MRI contrasts. It could hold great promise for clinical use for improved clinical decision-making and better diagnosis of brain cancer patients due to the availability of multicontrast MRIs. This approach may be clinically relevant and setting a significant step to efficiently fill a gap of absent MR sequences without additional scanning.

Project description:PurposeTo implement an integrated reconstruction pipeline including a graphics processing unit (GPU)-based convolutional neural network (CNN) architecture and test whether it reconstructs four-dimensional non-Cartesian, non-contrast material-enhanced MR angiographic k-space data faster than a central processing unit (CPU)-based compressed sensing (CS) reconstruction pipeline, without significant losses in data fidelity, summed visual score (SVS), or arterial vessel-diameter measurements.Materials and methodsRaw k-space data of 24 patients (18 men and six women; mean age, 56.8 years ± 11.8 [standard deviation]) suspected of having thoracic aortic disease were used to evaluate the proposed reconstruction pipeline derived from an open-source three-dimensional CNN. For training, 4800 zero-filled images and the corresponding CS-reconstructed images from 10 patients were used as input-output pairs. For testing, 6720 zero-filled images from 14 different patients were used as inputs to a trained CNN. Metrics for evaluating the agreement between the CNN and CS images included reconstruction times, structural similarity index (SSIM) and normalized root-mean-square error (NRMSE), SVS (3 = nondiagnostic, 9 = clinically acceptable, 15 = excellent), and vessel diameters.ResultsThe mean reconstruction time was 65 times and 69 times shorter for the CPU-based and GPU-based CNN pipelines (216.6 seconds ± 40.5 and 204.9 seconds ± 40.5), respectively, than for CS (14 152.3 seconds ± 1708.6) (P < .001). Compared with CS as practical ground truth, CNNs produced high data fidelity (SSIM = 0.94 ± 0.02, NRMSE = 2.8% ± 0.4) and not significantly different (P = .25) SVS and aortic diameters, except at one out of seven locations, where the percentage difference was only 3% (ie, clinically irrelevant).ConclusionThe proposed integrated reconstruction pipeline including a CNN architecture is capable of rapidly reconstructing time-resolved volumetric cardiovascular MRI k-space data, without a significant loss in data quality, thereby supporting clinical translation of said non-contrast-enhanced MR angiograms. Supplemental material is available for this article. © RSNA, 2020.