Search

Dynamic magnetic resonance imaging has emerged as a powerful modality for investigating upper-airway function during speech production. Analyzing the changes in the vocal tract airspace, including the position of soft-tissue articulators (e.g., the tongue and velum), enhances our understanding of speech production. The advent of various fast speech MRI protocols based on sparse sampling and constrained reconstruction has led to the creation of dynamic speech MRI datasets on the order of 80–100 image frames/second. In this paper, we propose a stacked transfer learning U-NET model to segment the deforming vocal tract in 2D mid-sagittal slices of dynamic speech MRI. Our approach leverages (a) low- and mid-level features and (b) high-level features. The low- and mid-level features are derived from models pre-trained on labeled open-source brain tumor MR and lung CT datasets, and an in-house airway labeled dataset. The high-level features are derived from labeled protocol-specific MR images. The applicability of our approach to segmenting dynamic datasets is demonstrated in data acquired from three fast speech MRI protocols: Protocol 1: 3 T-based radial acquisition scheme coupled with a non-linear temporal regularizer, where speakers were producing French speech tokens; Protocol 2: 1.5 T-based uniform density spiral acquisition scheme coupled with a temporal finite difference (FD) sparsity regularization, where speakers were producing fluent speech tokens in English, and Protocol 3: 3 T-based variable density spiral acquisition scheme coupled with manifold regularization, where speakers were producing various speech tokens from the International Phonetic Alphabetic (IPA). Segments from our approach were compared to those from an expert human user (a vocologist), and the conventional U-NET model without transfer learning. Segmentations from a second expert human user (a radiologist) were used as ground truth. Evaluations were performed using the quantitative DICE similarity metric, the Hausdorff distance metric, and segmentation count metric. This approach was successfully adapted to different speech MRI protocols with only a handful of protocol-specific images (e.g., of the order of 20 images), and provided accurate segmentations similar to those of an expert human. [ABSTRACT FROM AUTHOR]

The main focus of this work is to introduce a single free-breathing and ungated imaging protocol to jointly estimate cardiac function and myocardial T 1 maps. We reconstruct a time series of images corresponding to k-space data from a free-breathing and ungated inversion recovery gradient echo sequence using a manifold algorithm. We model each image in the time series as a non-linear function of three variables: cardiac and respiratory phases and inversion time. The non-linear function is realized using a convolutional neural networks (CNN) generator, while the CNN parameters, as well as the phase information, are estimated from the measured k-t space data. We use a dense conditional auto-encoder to estimate the cardiac and respiratory phases from the central multi-channel k-space samples acquired at each frame. The latent vectors of the auto-encoder are constrained to be bandlimited functions with appropriate frequency bands, which enables the disentanglement of the latent vectors into cardiac and respiratory phases, even when the data are acquired with intermittent inversion pulses. Once the phases are estimated, we pose the image recovery as the learning of the parameters of the CNN generator from the measured k-t space data. The learned CNN generator is used to generate synthetic data on demand by feeding it with appropriate latent vectors. The proposed approach capitalizes on the synergies between cine MRI and T 1 mapping to reduce the scan time and improve patient comfort. The framework also enables the generation of synthetic breath-held cine movies with different inversion contrasts, which improves the visualization of the myocardium. In addition, the approach also enables the estimation of the T 1 maps with specific phases, which is challenging with breath-held approaches. [ABSTRACT FROM AUTHOR]

Simple Summary: Prior radiomic studies have addressed a two-class tumor classification problem (glioblastoma (GBM) versus primary CNS lymphoma (PCNSL) or GBM versus metastasis). However, this approach is prone to bias and excludes other common brain tumor types. We addressed a real-life clinical problem by including the three most common brain tumor types (GBM, PCNSL, and metastasis). We investigated two key issues using different MRI sequence combinations: performance variation based on tumor subregions (necrotic, enhancing, edema and combined enhancing, and necrotic masks), and performance metrics based on the chosen classifier model/feature selection combination. Our study provides evidence that radiomics-based three-class tumor differentiation is feasible, and that embedded models perform better than those with a priori feature selection. We found that T1 contrast enhanced is the single best sequence with comparable performance to that of multiparametric MRI, and model performance varies based on tumor subregion and the combination of model/feature selection methods. Prior radiomics studies have focused on two-class brain tumor classification, which limits generalizability. The performance of radiomics in differentiating the three most common malignant brain tumors (glioblastoma (GBM), primary central nervous system lymphoma (PCNSL), and metastatic disease) is assessed; factors affecting the model performance and usefulness of a single sequence versus multiparametric MRI (MP-MRI) remain largely unaddressed. This retrospective study included 253 patients (120 metastatic (lung and brain), 40 PCNSL, and 93 GBM). Radiomic features were extracted for whole a tumor mask (enhancing plus necrotic) and an edema mask (first pipeline), as well as for separate enhancing and necrotic and edema masks (second pipeline). Model performance was evaluated using MP-MRI, individual sequences, and the T1 contrast enhanced (T1-CE) sequence without the edema mask across 45 model/feature selection combinations. The second pipeline showed significantly high performance across all combinations (Brier score: 0.311–0.325). GBRM fit using the full feature set from the T1-CE sequence was the best model. The majority of the top models were built using a full feature set and inbuilt feature selection. No significant difference was seen between the top-performing models for MP-MRI (AUC 0.910) and T1-CE sequence with (AUC 0.908) and without edema masks (AUC 0.894). T1-CE is the single best sequence with comparable performance to that of multiparametric MRI (MP-MRI). Model performance varies based on tumor subregion and the combination of model/feature selection methods. [ABSTRACT FROM AUTHOR]

The role of reliable, non-invasive imaging-based recognition of pulmonary hypertension (PH) remains a diagnostic challenge. The aim of the current pilot radiomics study was to assess the diagnostic performance of cardiac MRI (cMRI)-based texture features to accurately predict PH. The study involved IRB-approved retrospective analysis of cMRIs from 72 patients (42 PH and 30 healthy controls) for the primary analysis. A subgroup analysis was performed including patients from the PH group with left ventricle ejection fraction ≥ 50%. Texture features were generated from mid-left ventricle myocardium using balanced steady-state free precession (bSSFP) cine short-axis imaging. Forty-five different combinations of classifier models and feature selection techniques were evaluated. Model performance was assessed using receiver operating characteristic curves. A multilayer perceptron model fitting using full feature sets was the best classifier model for both the primary analysis (AUC 0.862, accuracy 78%) and the subgroup analysis (AUC 0.918, accuracy 80%). Model performance demonstrated considerable variation between the models (AUC 0.523–0.918) based on the chosen model–feature selection combination. Cardiac MRI-based radiomics recognition of PH using texture features is feasible, even with preserved left ventricular ejection fractions. [ABSTRACT FROM AUTHOR]