Weakly supervised deep learning for diagnosis of multiple vertebral compression fractures in CT.

OBJECTIVE: This study aims to develop a weakly supervised deep learning (DL) model for vertebral-level vertebral compression fracture (VCF) classification using image-level labelled data. METHODS: The training set included 815 patients with normal (n = 507, 62%) or VCFs (n = 308, 38%). Our proposed model was trained on image-level labelled data for vertebral-level classification. Another supervised DL model was trained with vertebral-level labelled data to compare the performance of the proposed model. RESULTS: The test set included 227 patients with normal (n = 117, 52%) or VCFs (n = 110, 48%). For a fair comparison of the two models, we compared sensitivities with the same specificities of the proposed model and the vertebral-level supervised model. The specificity for overall L1-L5 performance was 0.981. The proposed model may outperform the vertebral-level supervised model with sensitivities of 0.770 vs 0.705 (p = 0.080), respectively. For vertebral-level analysis, the specificities for each L1-L5 were 0.974, 0.973, 0.970, 0.991, and 0.995, respectively. The proposed model yielded the same or better sensitivity than the vertebral-level supervised model in L1 (0.750 vs 0.694, p = 0.480), L3 (0.793 vs 0.586, p < 0.05), L4 (0.833 vs 0.667, p = 0.480), and L5 (0.600 vs 0.600, p = 1.000), respectively. The proposed model showed lower sensitivity than the vertebral-level supervised model for L2, but there was no significant difference (0.775 vs 0.825, p = 0.617). CONCLUSIONS: The proposed model may have a comparable or better performance than the supervised model in vertebral-level VCF classification. CLINICAL RELEVANCE STATEMENT: Vertebral-level vertebral compression fracture classification aids in devising patient-specific treatment plans by identifying the precise vertebrae affected by compression fractures. KEY POINTS: • Our proposed weakly supervised method may have comparable or better performance than the supervised method for vertebral-level vertebral compression fracture classification. • The weakly supervised model could have classified cases with multiple vertebral compression fractures at the vertebral-level, even if the model was trained with image-level labels. • Our proposed method could help reduce radiologists' labour because it enables vertebral-level classification from image-level labels.

Intelligent noninvasive meningioma grading with a fully automatic segmentation using interpretable multiparametric deep learning.

OBJECTIVES: To establish a robust interpretable multiparametric deep learning (DL) model for automatic noninvasive grading of meningiomas along with segmentation. METHODS: In total, 257 patients with pathologically confirmed meningiomas (162 low-grade, 95 high-grade) who underwent a preoperative brain MRI, including T2-weighted (T2) and contrast-enhanced T1-weighted images (T1C), were included in the institutional training set. A two-stage DL grading model was constructed for segmentation and classification based on multiparametric three-dimensional U-net and ResNet. The models were validated in the external validation set consisting of 61 patients with meningiomas (46 low-grade, 15 high-grade). Relevance-weighted Class Activation Mapping (RCAM) method was used to interpret the DL features contributing to the prediction of the DL grading model. RESULTS: On external validation, the combined T1C and T2 model showed a Dice coefficient of 0.910 in segmentation and the highest performance for meningioma grading compared to the T2 or T1C only models, with an area under the curve (AUC) of 0.770 (95% confidence interval: 0.644-0.895) and accuracy, sensitivity, and specificity of 72.1%, 73.3%, and 71.7%, respectively. The AUC and accuracy of the combined DL grading model were higher than those of the human readers (AUCs of 0.675-0.690 and accuracies of 65.6-68.9%, respectively). The RCAM of the DL grading model showed activated maps at the surface regions of meningiomas indicating that the model recognized the features at the tumor margin for grading. CONCLUSIONS: An interpretable multiparametric DL model combining T1C and T2 can enable fully automatic grading of meningiomas along with segmentation. KEY POINTS: • The multiparametric DL model showed robustness in grading and segmentation on external validation. • The diagnostic performance of the combined DL grading model was higher than that of the human readers. • The RCAM interpreted that DL grading model recognized the meaningful features at the tumor margin for grading.

Deep learning referral suggestion and tumour discrimination using explainable artificial intelligence applied to multiparametric MRI.

OBJECTIVES: An appropriate and fast clinical referral suggestion is important for intra-axial mass-like lesions (IMLLs) in the emergency setting. We aimed to apply an interpretable deep learning (DL) system to multiparametric MRI to obtain clinical referral suggestion for IMLLs, and to validate it in the setting of nontraumatic emergency neuroradiology. METHODS: A DL system was developed in 747 patients with IMLLs ranging 30 diseases who underwent pre- and post-contrast T1-weighted (T1CE), FLAIR, and diffusion-weighted imaging (DWI). A DL system that segments IMLLs, classifies tumourous conditions, and suggests clinical referral among surgery, systematic work-up, medical treatment, and conservative treatment, was developed. The system was validated in an independent cohort of 130 emergency patients, and performance in referral suggestion and tumour discrimination was compared with that of radiologists using receiver operating characteristics curve, precision-recall curve analysis, and confusion matrices. Multiparametric interpretable visualisation of high-relevance regions from layer-wise relevance propagation overlaid on contrast-enhanced T1WI and DWI was analysed. RESULTS: The DL system provided correct referral suggestions in 94 of 130 patients (72.3%) and performed comparably to radiologists (accuracy 72.6%, McNemar test; p = .942). For distinguishing tumours from non-tumourous conditions, the DL system (AUC, 0.90 and AUPRC, 0.94) performed similarly to human readers (AUC, 0.81~0.92, and AUPRC, 0.88~0.95). Solid portions of tumours showed a high overlap of relevance, but non-tumours did not (Dice coefficient 0.77 vs. 0.33, p < .001), demonstrating the DL's decision. CONCLUSIONS: Our DL system could appropriately triage patients using multiparametric MRI and provide interpretability through multiparametric heatmaps, and may thereby aid neuroradiologic diagnoses in emergency settings. CLINICAL RELEVANCE STATEMENT: Our AI triages patients with raw MRI images to clinical referral pathways in brain intra-axial mass-like lesions. We demonstrate that the decision is based on the relative relevance between contrast-enhanced T1-weighted and diffusion-weighted images, providing explainability across multiparametric MRI data. KEY POINTS: • A deep learning (DL) system using multiparametric MRI suggested clinical referral to patients with intra-axial mass-like lesions (IMLLs) similar to radiologists (accuracy 72.3% vs. 72.6%). • In the differentiation of tumourous and non-tumourous conditions, the DL system (AUC, 0.90) performed similar with radiologists (AUC, 0.81-0.92). • The DL's decision basis for differentiating tumours from non-tumours can be quantified using multiparametric heatmaps obtained via the layer-wise relevance propagation method.

Deep-Learning-Aided Evaluation of Spondylolysis Imaged with Ultrashort Echo Time Magnetic Resonance Imaging.

Isthmic spondylolysis results in fracture of pars interarticularis of the lumbar spine, found in as many as half of adolescent athletes with persistent low back pain. While computed tomography (CT) is the gold standard for the diagnosis of spondylolysis, the use of ionizing radiation near reproductive organs in young subjects is undesirable. While magnetic resonance imaging (MRI) is preferable, it has lowered sensitivity for detecting the condition. Recently, it has been shown that ultrashort echo time (UTE) MRI can provide markedly improved bone contrast compared to conventional MRI. To take UTE MRI further, we developed supervised deep learning tools to generate (1) CT-like images and (2) saliency maps of fracture probability from UTE MRI, using ex vivo preparation of cadaveric spines. We further compared quantitative metrics of the contrast-to-noise ratio (CNR), mean squared error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM) between UTE MRI (inverted to make the appearance similar to CT) and CT and between CT-like images and CT. Qualitative results demonstrated the feasibility of successfully generating CT-like images from UTE MRI to provide easier interpretability for bone fractures thanks to improved image contrast and CNR. Quantitatively, the mean CNR of bone against defect-filled tissue was 35, 97, and 146 for UTE MRI, CT-like, and CT images, respectively, being significantly higher for CT-like than UTE MRI images. For the image similarity metrics using the CT image as the reference, CT-like images provided a significantly lower mean MSE (0.038 vs. 0.0528), higher mean PSNR (28.6 vs. 16.5), and higher SSIM (0.73 vs. 0.68) compared to UTE MRI images. Additionally, the saliency maps enabled quick detection of the location with probable pars fracture by providing visual cues to the reader. This proof-of-concept study is limited to the data from ex vivo samples, and additional work in human subjects with spondylolysis would be necessary to refine the models for clinical use. Nonetheless, this study shows that the utilization of UTE MRI and deep learning tools could be highly useful for the evaluation of isthmic spondylolysis.

Importance of CT image normalization in radiomics analysis: prediction of 3-year recurrence-free survival in non-small cell lung cancer.

OBJECTIVES: To analyze whether CT image normalization can improve 3-year recurrence-free survival (RFS) prediction performance in patients with non-small cell lung cancer (NSCLC) relative to the use of unnormalized CT images. METHODS: A total of 106 patients with NSCLC were included in the training set. For each patient, 851 radiomic features were extracted from the normalized and the unnormalized CT images, respectively. After the feature selection, random forest models were constructed with selected radiomic features and clinical features. The models were then externally validated in the test set consisting of 79 patients with NSCLC. RESULTS: The model using normalized CT images yielded better performance than the model using unnormalized CT images (with an area under the receiver operating characteristic curve of 0.802 vs 0.702, p = 0.01), with the model performing especially well among patients with adenocarcinoma (with an area under the receiver operating characteristic curve of 0.880 vs 0.720, p < 0.01). CONCLUSIONS: CT image normalization may improve prediction performance among patients with NSCLC, especially for patients with adenocarcinoma. KEY POINTS: • After CT image normalization, more radiomic features were able to be identified. • Prognostic performance in patients was improved significantly after CT image normalization compared with before the CT image normalization. • The improvement in prognostic performance following CT image normalization was superior in patients with adenocarcinoma.

Robust performance of deep learning for automatic detection and segmentation of brain metastases using three-dimensional black-blood and three-dimensional gradient echo imaging.

OBJECTIVES: To evaluate whether a deep learning (DL) model using both three-dimensional (3D) black-blood (BB) imaging and 3D gradient echo (GRE) imaging may improve the detection and segmentation performance of brain metastases compared to that using only 3D GRE imaging. METHODS: A total of 188 patients with brain metastases (917 lesions) who underwent a brain metastasis MRI protocol including contrast-enhanced 3D BB and 3D GRE were included in the training set. DL models based on 3D U-net were constructed. The models were validated in the test set consisting of 45 patients with brain metastases (203 lesions) and 49 patients without brain metastases. RESULTS: The combined 3D BB and 3D GRE model yielded better performance than the 3D GRE model (sensitivities of 93.1% vs 76.8%, p < 0.001), and this effect was significantly stronger in subgroups with small metastases (p interaction < 0.001). For metastases < 3 mm, ≥ 3 mm and < 10 mm, and ≥ 10 mm, the sensitivities were 82.4%, 93.2%, and 100%, respectively. The combined 3D BB and 3D GRE model showed a false-positive per case of 0.59 in the test set. The combined 3D BB and 3D GRE model showed a Dice coefficient of 0.822, while 3D GRE model showed a lower Dice coefficient of 0.756. CONCLUSIONS: The combined 3D BB and 3D GRE DL model may improve the detection and segmentation performance of brain metastases, especially in detecting small metastases. KEY POINTS: • The combined 3D BB and 3D GRE model yielded better performance for the detection of brain metastases than the 3D GRE model (p < 0.001), with sensitivities of 93.1% and 76.8%, respectively. • The combined 3D BB and 3D GRE model showed a false-positive rate per case of 0.59 in the test set. • The combined 3D BB and 3D GRE model showed a Dice coefficient of 0.822, while the 3D GRE model showed a lower Dice coefficient of 0.756.

Deep-learned short tau inversion recovery imaging using multi-contrast MR images.

PURPOSE: To generate short tau, or short inversion time (TI), inversion recovery (STIR) images from three multi-contrast MR images, without additional scanning, using a deep neural network. METHODS: For simulation studies, we used multi-contrast simulation images. For in-vivo studies, we acquired knee MR images including 288 slices of T1 -weighted (T1 -w), T2 -weighted (T2 -w), gradient-recalled echo (GRE), and STIR images taken from 12 healthy volunteers. Our MR image synthesis method generates a new contrast MR image from multi-contrast MR images. We used a deep neural network to identify the complex relationships between MR images that show various contrasts for the same tissues. Our contrast-conversion deep neural network (CC-DNN) is an end-to-end architecture that trains the model to create one image from three (T1 -w, T2 -w, and GRE images). We propose a new loss function to take into account intensity differences, misregistration, and local intensity variations. The CC-DNN-generated STIR images were evaluated with four quantitative evaluation metrics, including mean squared error, peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and multi-scale SSIM (MS-SSIM). Furthermore, a subjective evaluation was performed by musculoskeletal radiologists. RESULTS: Our method showed improved results in all quantitative evaluations compared with other methods and received the highest scores in subjective evaluations by musculoskeletal radiologists. CONCLUSION: This study suggests the feasibility of our method for generating STIR sequence images without additional scanning that offered a potential alternative to the STIR pulse sequence when additional scanning is limited or STIR artifacts are severe.

Parallel imaging in time-of-flight magnetic resonance angiography using deep multistream convolutional neural networks.

PURPOSE: To develop and evaluate a method of parallel imaging time-of-flight (TOF) MRA using deep multistream convolutional neural networks (CNNs). METHODS: A deep parallel imaging network ("DPI-net") was developed to reconstruct 3D multichannel MRA from undersampled data. It comprises 2 deep-learning networks: a network of multistream CNNs for extracting feature maps of multichannel images and a network of reconstruction CNNs for reconstructing images from the multistream network output feature maps. The images were evaluated using normalized root mean square error (NRMSE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) values, and the visibility of blood vessels was assessed by measuring the vessel sharpness of middle and posterior cerebral arteries on axial maximum intensity projection (MIP) images. Vessel sharpness was compared using paired t tests, between DPI-net, 2 conventional parallel imaging methods (SAKE and ESPIRiT), and a deep-learning method (U-net). RESULTS: DPI-net showed superior performance in reconstructing vessel signals in both axial slices and MIP images for all reduction factors. This was supported by the quantitative metrics, with DPI-net showing the lowest NRMSE, the highest PSNR and SSIM (except R = 3.8 on sagittal MIP images, and R = 5.7 on axial slices and sagittal MIP images), and significantly higher vessel sharpness values than the other methods. CONCLUSION: DPI-net was effective in reconstructing 3D TOF MRA from highly undersampled multichannel MR data, achieving superior performance, both quantitatively and qualitatively, over conventional parallel imaging and other deep-learning methods.

KIKI-net: cross-domain convolutional neural networks for reconstructing undersampled magnetic resonance images.

PURPOSE: To demonstrate accurate MR image reconstruction from undersampled k-space data using cross-domain convolutional neural networks (CNNs) METHODS: Cross-domain CNNs consist of 3 components: (1) a deep CNN operating on the k-space (KCNN), (2) a deep CNN operating on an image domain (ICNN), and (3) an interleaved data consistency operations. These components are alternately applied, and each CNN is trained to minimize the loss between the reconstructed and corresponding fully sampled k-spaces. The final reconstructed image is obtained by forward-propagating the undersampled k-space data through the entire network. RESULTS: Performances of K-net (KCNN with inverse Fourier transform), I-net (ICNN with interleaved data consistency), and various combinations of the 2 different networks were tested. The test results indicated that K-net and I-net have different advantages/disadvantages in terms of tissue-structure restoration. Consequently, the combination of K-net and I-net is superior to single-domain CNNs. Three MR data sets, the T2 fluid-attenuated inversion recovery (T2 FLAIR) set from the Alzheimer's Disease Neuroimaging Initiative and 2 data sets acquired at our local institute (T2 FLAIR and T1 weighted), were used to evaluate the performance of 7 conventional reconstruction algorithms and the proposed cross-domain CNNs, which hereafter is referred to as KIKI-net. KIKI-net outperforms conventional algorithms with mean improvements of 2.29 dB in peak SNR and 0.031 in structure similarity. CONCLUSION: KIKI-net exhibits superior performance over state-of-the-art conventional algorithms in terms of restoring tissue structures and removing aliasing artifacts. The results demonstrate that KIKI-net is applicable up to a reduction factor of 3 to 4 based on variable-density Cartesian undersampling.

Small-scale noise-like moiré pattern caused by detector sensitivity inhomogeneity in computed tomography.

We report a new type of moiré pattern caused by inhomogeneous detector sensitivity in computed tomography. Defects in one or a few detector bins or miscalibrated detectors induce well-known ring artifacts. When detector sensitivity is not homogenous over all detector bins, these ring artifacts occur everywhere as distributed rings in reconstructed images and may cause a moiré pattern when combined with insufficient view sampling, which induces a noise-like pattern or a subtle texture in the reconstructed images. Complete correction of the inhomogeneity in detectors can remove the pattern and improve image quality. This paper describes several properties of moiré patterns caused by detector sensitivity inhomogeneity.

High-SNR multiple T2 (*)-contrast magnetic resonance imaging using a robust denoising method based on tissue characteristics.

PURPOSE: To develop an effective method that can suppress noise in successive multiecho T2 (*)-weighted magnetic resonance (MR) brain images while preventing filtering artifacts. MATERIALS AND METHODS: For the simulation experiments, we used multiple T2 -weighted images of an anatomical brain phantom. For in vivo experiments, successive multiecho MR brain images were acquired from five healthy subjects using a multiecho gradient-recalled-echo (MGRE) sequence with a 3T MRI system. Our denoising method is a nonlinear filter whose filtering weights are determined by tissue characteristics among pixels. The similarity of the tissue characteristics is measured based on the l2 -difference between two temporal decay signals. Both numerical and subjective evaluations were performed in order to compare the effectiveness of our denoising method with those of conventional filters, including Gaussian low-pass filter (LPF), anisotropic diffusion filter (ADF), and bilateral filter. Root-mean-square error (RMSE), signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were used in the numerical evaluation. Five observers, including one radiologist, assessed the image quality and rated subjective scores in the subjective evaluation. RESULTS: Our denoising method significantly improves RMSE, SNR, and CNR of numerical phantom images, and CNR of in vivo brain images in comparison with conventional filters (P < 0.005). It also receives the highest scores for structure conspicuity (8.2 to 9.4 out of 10) and naturalness (9.2 to 9.8 out of 10) among the conventional filters in the subjective evaluation. CONCLUSION: This study demonstrates that high-SNR multiple T2 (*)-contrast MR images can be obtained using our denoising method based on tissue characteristics without noticeable artifacts. Evidence level: 2 J. MAGN. RESON. IMAGING 2017;45:1835-1845.

Improved estimation of myelin water fraction using complex model fitting.

In gradient echo (GRE) imaging, three compartment water modeling (myelin water, axonal water and extracellular water) in white matter has been demonstrated to show different frequency shifts that depend on the relative orientation of fibers and the B0 field. This finding suggests that in GRE-based myelin water imaging, a signal model may need to incorporate frequency offset terms and become a complex-valued model. In the current study, three different signal models and fitting approaches (a magnitude model fitted to magnitude data, a complex model fitted to magnitude data, and a complex model fitted to complex data) were investigated to address the reliability of each model in the estimation of the myelin water signal. For the complex model fitted to complex data, a new fitting approach that does not require background phase removal was proposed. When the three models were compared, the results from the new complex model fitting showed the most stable parameter estimation.

Susceptibility map-weighted imaging (SMWI) for neuroimaging.

PURPOSE: To propose a susceptibility map-weighted imaging (SMWI) method by combining a magnitude image with a quantitative susceptibility mapping (QSM) -based weighting factor thereby providing an alternative contrast compared with magnitude image, susceptibility-weighted imaging, and QSM. METHODS: A three-dimensional multi-echo gradient echo sequence is used to obtain the data. The QSM was transformed to a susceptibility mask that varies in amplitude between zero and unity. This mask was multiplied several times with the original magnitude image to create alternative contrasts between tissues with different susceptibilities. A temporal domain denoising method to enhance the signal-to-noise ratio was further applied. Optimal reconstruction processes of the SMWI were determined from simulations. RESULTS: Temporal domain denoising enhanced the signal-to-noise ratio, especially at late echoes without spatial artifacts. From phantom simulations, the optimal number of multiplication and threshold values was chosen. Reconstructed SMWI created different contrasts based on its weighting factors made from paramagnetic or diamagnetic susceptibility tissue and provided an excellent delineation of microhemorrhage without blooming artifacts typically caused by the nonlocal property of phase. CONCLUSION: SMWI presents an alternative contrast for susceptibility-based imaging. The validity of this method was demonstrated using in vivo data. This proposed method together with denoising allows high-quality reconstruction of susceptibility-weighted image of human brain in vivo.

Ring artifact correction using detector line-ratios in computed tomography.

Ring artifacts in computed tomography (CT) images degrade image quality and obscure the true shapes of objects. While several correction methods have been developed, their performances are often task-dependent and not generally applicable. Here, we propose a novel method to reduce ring artifacts by calculating the ratio of adjacent detector elements in the projection data, termed the line-ratio. Our method estimates the sensitivity of each detector element and equalizes them in sinogram space. As a result, the stripe pattern can be effectively removed from sinogram data, thereby also removing ring artifacts from the reconstructed CT image. Numerical simulations were performed to evaluate and compare the performance of our method with that of conventional methods. We also tested our method experimentally and demonstrated that our method has superior performance to other methods.

Monolayer, open-mesh, pristine PEDOT:PSS-based conformal brain implants for fully MRI-compatible neural interfaces.

Understanding brain function is essential for advancing our comprehension of human cognition, behavior, and neurological disorders. Magnetic resonance imaging (MRI) stands out as a powerful tool for exploring brain function, providing detailed insights into its structure and physiology. Combining MRI technology with electrophysiological recording system can enhance the comprehension of brain functionality through synergistic effects. However, the integration of neural implants with MRI technology presents challenges because of its strong electromagnetic (EM) energy during MRI scans. Therefore, MRI-compatible neural implants should facilitate detailed investigation of neural activities and brain functions in real-time in high resolution, without compromising patient safety and imaging quality. Here, we introduce the fully MRI-compatible monolayer open-mesh pristine PEDOT:PSS neural interface. This approach addresses the challenges encountered while using traditional metal-based electrodes in the MRI environment such as induced heat or imaging artifacts. PEDOT:PSS has a diamagnetic property with low electrical conductivity and negative magnetic susceptibility similar to human tissues. Furthermore, by adopting the optimized open-mesh structure, the induced currents generated by EM energy are significantly diminished, leading to optimized MRI compatibility. Through simulations and experiments, our PEDOT:PSS-based open-mesh electrodes showed improved performance in reducing heat generation and eliminating imaging artifacts in an MRI environment. The electrophysiological recording capability was also validated by measuring the local field potential (LFP) from the somatosensory cortex with an in vivo experiment. The development of neural implants with maximized MRI compatibility indicates the possibility of potential tools for future neural diagnostics.

Improvement of the SNR and resolution of susceptibility-weighted venography by model-based multi-echo denoising.

The vein structures of the brain are important for understanding brain function and structure, especially when functional magnetic resonance imaging (fMRI) is utilized, as fMRI is based on changes in the blood-oxygen-level-dependent (BOLD) signal, which is directly related to veins. The aim of the present study was to develop an effective method to produce high signal-to-noise-ratio (SNR) and high-resolution multi-contrast susceptibility-weighted (SW) images of vein structures from 3T magnetic resonance (MR) scanners using multi-gradient-echo MR acquisition and a successive denoising process for both magnitude and phase data. Successive multi-echo MR images were acquired at multiple time points using a multigradient-recalled echo sequence at 3T, and noise in the magnitude and phase data was effectively suppressed using model-based denoising methods. A T(2)* relaxation model was used to denoise the magnitude data and a linear phase model was used to denoise the phase data. SW venography images were obtained from the denoised MR data and compared with conventional SW venography. To evaluate the performance of our denoising methods, we conducted numerical simulation studies and compared the mean-squared-error (MSE), SNR, and contrast-to-noise ratio (CNR) that we obtained using our procedure with those obtained using conventional denoising methods. In addition, images were inspected visually. Numerical simulations showed that our proposed model-based denoising methods were the most effective at suppressing noise. In vivo experiments also showed a substantial increase in the SNR of the phase mask obtained using the proposed denoising process (twice that of the conventional GRE-based phase mask). The T(2)* relaxation model method improved the SNR of the magnitude image (1.17-1.35 times that of the GRE-based magnitude image). Noise suppression of both magnitude and phase data using our proposed method resulted in an overall increase in the SNR and CNR in the final SW venography (1.1-1.5-fold and 1.96-fold higher SNR and CNR, respectively, than that of the GRE-based SW venography). We demonstrated that high SNR and high-resolution SW venograms can be obtained using multi-echo gradient-recalled acquisition and successive model-based denoising of both magnitude and phase data.

A tissue-relaxation-dependent neighboring method for robust mapping of the myelin water fraction.

Quantitative assessment of the myelin content in white matter (WM) using MRI has become a useful tool for investigating myelin-related diseases, such as multiple sclerosis (MS). Myelin water fraction (MWF) maps can be estimated pixel-by-pixel by a determination of the T2 or T2* spectrum from signal decay measurements at each individual image pixel. However, detection of parameters from the measured decay curve, assuming a combination of smooth multi-exponential curves, results in a nonlinear and seriously ill-posed problem. In this paper, we propose a new method to obtain a stable MWF map robust to the presence of noise while sustaining sufficient resolution, which uses weighted combinations of measured decay signals in a spatially independent neighborhood to combine tissues with similar relaxation parameters. To determine optimal weighting factors, we define a spatially independent neighborhood for each pixel and a distance with respect to decay rates that effectively includes pixels with similar decay characteristics, and which therefore have similar relaxation parameters. We recover the MWF values by using optimally weighted decay curves. We use numerical simulations and in vitro and in vivo experimental brain data scanned with a multi-gradient-echo sequence to demonstrate the feasibility of our proposed algorithm and to highlight its advantages compared to the conventional method.

Development and validation of a hybrid deep learning-machine learning approach for severity assessment of COVID-19 and other pneumonias.

The Coronavirus Disease 2019 (COVID-19) is transitioning into the endemic phase. Nonetheless, it is crucial to remain mindful that pandemics related to infectious respiratory diseases (IRDs) can emerge unpredictably. Therefore, we aimed to develop and validate a severity assessment model for IRDs, including COVID-19, influenza, and novel influenza, using CT images on a multi-centre data set. Of the 805 COVID-19 patients collected from a single centre, 649 were used for training and 156 were used for internal validation (D1). Additionally, three external validation sets were obtained from 7 cohorts: 1138 patients with COVID-19 (D2), and 233 patients with influenza and novel influenza (D3). A hybrid model, referred to as Hybrid-DDM, was constructed by combining two deep learning models and a machine learning model. Across datasets D1, D2, and D3, the Hybrid-DDM exhibited significantly improved performance compared to the baseline model. The areas under the receiver operating curves (AUCs) were 0.830 versus 0.767 (p = 0.036) in D1, 0.801 versus 0.753 (p < 0.001) in D2, and 0.774 versus 0.668 (p < 0.001) in D3. This study indicates that the Hybrid-DDM model, trained using COVID-19 patient data, is effective and can also be applicable to patients with other types of viral pneumonia.

High-quality multiple T(2)(*) contrast MR images from low-quality multi-echo images using temporal-domain denoising methods.

PURPOSE: The aim of this study was to develop an effective postprocessing method to increase the signal-to-noise ratio in successive multi-echo magnetic resonance (MR) images acquired at multiple time points and generate high-quality multiple T(2)(*) contrast images from low-quality multi-echo images. METHODS: Successive multi-echo MR images were acquired at multiple time points using a multigradient-recalled echo sequence at 3T and rearranged so that each pixel in the images had its own decay signal in the temporal-domain. Two different denonising approaches were implemented in the temporal-domain: (1) In a filtering approach, conventional low-pass filter, median filter, and anisotropic diffusion filter were applied to the decay signals to reduce random noise; (2) In a model-based approach, a non-negative least squares algorithm was applied for fitting to MR relaxation model for decay signals. Numerical simulations and in vivo experiments were conducted. The denoised images were compared to each other by visual inspection and analysis of mean square error (MSE) and contrast-to-noise ratio (CNR) on several regions of interest. RESULTS: Our proposed method suppressed noise in each multi-echo images without introducing spatial artifacts. This was a natural consequence of the proposed denoising process, which was performed in the temporal-domain and did not use any cross-pixel operation. MSEs decreased by a factor of 5.4-7.9 and CNRs increased by a factor of 5 in simulation studies. The results were consistent with the in vivo findings. Random noise in the images was effectively reduced and high-quality multiple T(2)(*) contrast images were obtained. CONCLUSIONS: This study demonstrated that denoising methods in the temporal-domain can effectively suppress noise in the spatial domain, and increase signal-to-noise ratio (SNR) for each image of different T(2)(*) weights at multiple time points, resulting in multiple high-quality T(2)(*) contrast images.