Search

The ability to differentiate post-cancer from healthy tongue muscle coordination patterns is necessary for the advancement of speech motor control theories and for the development of therapeutic and rehabilitative strategies. A deep learning approach is presented to classify two groups using muscle coordination patterns from magnetic resonance imaging (MRI). The proposed method uses tagged-MRI to track the tongue's internal tissue points and atlas-driven non-negative matrix factorization to reduce the dimensionality of the deformation fields. A convolutional neural network is applied to the classification task yielding an accuracy of 96.90%, offering the potential to the development of therapeutic or rehabilitative strategies in speech-related disorders. [ABSTRACT FROM AUTHOR]

Muscle coordination patterns of lingual behaviors are synergies generated by deforming local muscle groups in a variety of ways. Functional units are functional muscle groups of local structural elements within the tongue that compress, expand, and move in a cohesive and consistent manner. Identifying the functional units using tagged-magnetic resonance imaging (MRI) sheds light on the mechanisms of normal and pathological muscle coordination patterns, yielding improvement in surgical planning, treatment, or rehabilitation procedures. In this paper, to mine this information, we propose a matrix factorization and probabilistic graphical model framework to produce building blocks and their associated weighting map using motion quantities extracted from tagged-MRI. Our tagged-MRI imaging and accurate voxel-level tracking provide previously unavailable internal tongue motion patterns, thus revealing the inner workings of the tongue during speech or other lingual behaviors. We then employ spectral clustering on the weighting map to identify the cohesive regions defined by the tongue motion that may involve multiple or undocumented regions. To evaluate our method, we perform a series of experiments. We first use two-dimensional images and synthetic data to demonstrate the accuracy of our method. We then use three-dimensional synthetic and in vivo tongue motion data using protrusion and simple speech tasks to identify subject-specific and data-driven functional units of the tongue in localized regions. [ABSTRACT FROM AUTHOR]

Tagged magnetic resonance imaging has been used for decades to observe and quantify motion and strain of deforming tissue. It is challenging to obtain 3-D motion estimates due to a tradeoff between image slice density and acquisition time. Typically, interpolation methods are used either to combine 2-D motion extracted from sparse slice acquisitions into 3-D motion or to construct a dense volume from sparse acquisitions before image registration methods are applied. This paper proposes a new phase-based 3-D motion estimation technique that first computes harmonic phase volumes from interpolated tagged slices and then matches them using an image registration framework. The approach uses several concepts from diffeomorphic image registration with a key novelty that defines a symmetric similarity metric on harmonic phase volumes from multiple orientations. The material property of harmonic phase solves the aperture problem of optical flow and intensity-based methods and is robust to tag fading. A harmonic magnitude volume is used in enforcing incompressibility in the tissue regions. The estimated motion fields are dense, incompressible, diffeomorphic, and inverse-consistent at a 3-D voxel level. The method was evaluated using simulated phantoms, human brain data in mild head accelerations, human tongue data during speech, and an open cardiac data set. The method shows comparable accuracy to three existing methods while demonstrating low computation time and robustness to tag fading and noise. [ABSTRACT FROM PUBLISHER]

Dynamic MRI has been widely used to track the motion of the tongue and measure its internal deformation during speech and swallowing. Accurate segmentation of the tongue is a prerequisite step to define the target boundary and constrain the tracking to tissue points within the tongue. Segmentation of 2D slices or 3D volumes is challenging because of the large number of slices and time frames involved in the segmentation, as well as the incorporation of numerous local deformations that occur throughout the tongue during motion. In this paper, we propose a semi-automatic approach to segment 3D dynamic MRI of the tongue. The algorithm steps include seeding a few slices at one time frame, propagating seeds to the same slices at different time frames using deformable registration, and random walker segmentation based on these seed positions. This method was validated on the tongue of five normal subjects carrying out the same speech task with multi-slice 2D dynamic cine-MR images obtained at three orthogonal orientations and 26 time frames. The resulting semi-automatic segmentations of a total of 130 volumes showed an average dice similarity coefficient (DSC) score of 0.92 with less segmented volume variability between time frames than in manual segmentations. [ABSTRACT FROM AUTHOR]

Magnetic resonance imaging has been widely used in speech production research. Often only one image stack (sagittal, axial, or coronal) is used for vocal tract modeling. As a result, complementary information from other available stacks is not utilized. To overcome this, a recently developed super-resolution technique was applied to integrate three orthogonal low-resolution stacks into one isotropic volume. The results on vowels show that the super-resolution volume produces better vocal tract visualization than any of the low-resolution stacks. Its derived area functions generally produce formant predictions closer to the ground truth, particularly for those formants sensitive to area perturbations at constrictions. [ABSTRACT FROM AUTHOR]

Magnetic resonance images of the tongue have been used in both clinical studies and scientific research to reveal tongue structure. In order to extract different features of the tongue and its relation to the vocal tract, it is beneficial to acquire three orthogonal image volumes—e.g., axial, sagittal, and coronal volumes. In order to maintain both low noise and high visual detail and minimize the blurred effect due to involuntary motion artifacts, each set of images is acquired with an in-plane resolution that is much better than the through-plane resolution. As a result, any one dataset, by itself, is not ideal for automatic volumetric analyses such as segmentation, registration, and atlas building or even for visualization when oblique slices are required. This paper presents a method of superresolution volume reconstruction of the tongue that generates an isotropic image volume using the three orthogonal image volumes. The method uses preprocessing steps that include registration and intensity matching and a data combination approach with the edge-preserving property carried out by Markov random field optimization. The performance of the proposed method was demonstrated on 15 clinical datasets, preserving anatomical details and yielding superior results when compared with different reconstruction methods as visually and quantitatively assessed. [ABSTRACT FROM PUBLISHER]

Purpose: The authors aimed to develop an image-based registration scheme to detect and correct patient motion in stress and rest cardiac positron emission tomography (PET)/CT images. The patient motion correction was of primary interest and the effects of patient motion with the use of flurpiridaz F 18 and 82Rb were demonstrated. Methods: The authors evaluated stress/rest PET myocardial perfusion imaging datasets in 30 patients (60 datasets in total, 21 male and 9 female) using a new perfusion agent (flurpiridaz F 18) (n = 16) and 82Rb (n = 14), acquired on a Siemens Biograph-64 scanner in list mode. Stress and rest images were reconstructed into 4 (82Rb) or 10 (flurpiridaz F 18) dynamic frames (60 s each) using standard reconstruction (2D attenuation weighted ordered subsets expectation maximization). Patient motion correction was achieved by an image-based registration scheme optimizing a cost function using modified normalized cross-correlation that combined global and local features. For comparison, visual scoring of motion was performed on the scale of 0 to 2 (no motion, moderate motion, and large motion) by two experienced observers. Results: The proposed registration technique had a 93% success rate in removing left ventricular motion, as visually assessed. The maximum detected motion extent for stress and rest were 5.2 mm and 4.9 mm for flurpiridaz F 18 perfusion and 3.0 mm and 4.3 mm for 82Rb perfusion studies, respectively. Motion extent (maximum frame-to-frame displacement) obtained for stress and rest were (2.2 ± 1.1, 1.4 ± 0.7, 1.9 ± 1.3) mm and (2.0 ± 1.1, 1.2 ±0 .9, 1.9 ± 0.9) mm for flurpiridaz F 18 perfusion studies and (1.9 ± 0.7, 0.7 ± 0.6, 1.3 ± 0.6) mm and (2.0 ± 0.9, 0.6 ± 0.4, 1.2 ± 1.2) mm for 82Rb perfusion studies, respectively. A visually detectable patient motion threshold was established to be ≥2.2 mm, corresponding to visual user scores of 1 and 2. After motion correction, the average increases in contrast-to-noise ratio (CNR) from all frames for larger than the motion threshold were 16.2% in stress flurpiridaz F 18 and 12.2% in rest flurpiridaz F 18 studies. The average increases in CNR were 4.6% in stress 82Rb studies and 4.3% in rest 82Rb studies. Conclusions: Fully automatic motion correction of dynamic PET frames can be performed accurately, potentially allowing improved image quantification of cardiac PET data. [ABSTRACT FROM AUTHOR]

Purpose: Coronary CT angiography (CCTA) is a high-resolution three-dimensional imaging technique for the evaluation of coronary arteries in suspected or confirmed coronary artery disease (CAD). Coregistration of serial CCTA scans would allow precise superimposition of images obtained at two different points in time, which could aid in recognition of subtle changes and precise monitoring of coronary plaque progression or regression. To this end, the authors aimed at developing a fully automatic nonlinear volume coregistration for longitudinal CCTA scan pairs. Methods: The algorithm combines global displacement and local deformation using nonlinear volume coregistration with a volume-preserving constraint. Histogram matching of intensities between two serial scans is performed prior to nonlinear coregistration with dense nonparametric local deformation in which sum of squared differences is used as a similarity measure. The approximate segmentation of coronary arteries obtained from commercially available software provides initial anatomical landmarks for the coregistration algorithm that help localize and emphasize the structure of interest. To avoid possible bias caused by incorrect segmentation, the authors convolve the Gaussian kernel with the segmented binary coronary tree mask and define an extended weighted region of interest. A multiresolution approach is employed to represent coarse-to-fine details of both volumes and the energy function is optimized using a gradient descent method. The authors applied the algorithm in ten paired CCTA datasets (20 scans in total) obtained within 10.7±5.7 months from each other on a dual source CT scanner to monitor progression of CAD. Results: Serial CCTA coregistration was successful in 9/10 cases as visually confirmed. The global displacement and local deformation of target registration error obtained from four anatomical landmarks were 2.22±1.15 and 1.56±0.74 mm, respectively, and the inverse consistency error of local deformation was 0.14±0.06 mm. The observer variability between two expert observers was 1.31±0.91 mm. Conclusions: The proposed coregistration algorithm demonstrates potential to accurately register serial CCTA scans, which may allow direct comparison of calcified and noncalcified atherosclerotic plaque changes between the two scans. [ABSTRACT FROM AUTHOR]

• We presented a method to identify both common and subject-specific functional units within a material coordinate system from MRI. • We proposed to convert non-negative matrix factorization with sparse and manifold regularizations into modular structures. • We provided validation to demonstrate superior performance over the comparison methods on both simulated and in vivo tongue data. [Display omitted] Intelligible speech is produced by creating varying internal local muscle groupings—i.e., functional units—that are generated in a systematic and coordinated manner. There are two major challenges in characterizing and analyzing functional units. First, due to the complex and convoluted nature of tongue structure and function, it is of great importance to develop a method that can accurately decode complex muscle coordination patterns during speech. Second, it is challenging to keep identified functional units across subjects comparable due to their substantial variability. In this work, to address these challenges, we develop a new deep learning framework to identify common and subject-specific functional units of tongue motion during speech. Our framework hinges on joint deep graph-regularized sparse non-negative matrix factorization (NMF) using motion quantities derived from displacements by tagged Magnetic Resonance Imaging. More specifically, we transform NMF with sparse and graph regularizations into modular architectures akin to deep neural networks by means of unfolding the Iterative Shrinkage-Thresholding Algorithm to learn interpretable building blocks and associated weighting map. We then apply spectral clustering to common and subject-specific weighting maps from which we jointly determine the common and subject-specific functional units. Experiments carried out with simulated datasets show that the proposed method achieved on par or better clustering performance over the comparison methods.Experiments carried out with in vivo tongue motion data show that the proposed method can determine the common and subject-specific functional units with increased interpretability and decreased size variability. [ABSTRACT FROM AUTHOR]

This study investigates mechanical cooperation among tongue muscles. Five volunteers were imaged using tagged magnetic resonance imaging to quantify spatiotemporal kinematics while speaking. Waveforms of strain in the line of action of fibers (SLAF) were estimated by projecting strain tensors onto a model of fiber directionality. SLAF waveforms were temporally aligned to determine consistency across subjects and correlation across muscles. The cohort exhibited consistent patterns of SLAF, and muscular extension-contraction was correlated. Volume-preserving tongue movement in speech generation can be achieved through multiple paths, but the study reveals similarities in motion patterns and muscular action—despite anatomical (and other) dissimilarities. [ABSTRACT FROM AUTHOR]

Thus far, there have been no reported specific rules for systematically determining the appropriate augmented sample size to optimize model performance when conducting data augmentation. In this paper, we report on the feasibility of synthetic data augmentation using generative adversarial networks (GAN) by proposing an automation pipeline to find the optimal multiple of data augmentation to achieve the best deep learning-based diagnostic performance in a limited dataset. We used Waters' view radiographs for patients diagnosed with chronic sinusitis to demonstrate the method developed herein. We demonstrate that our approach produces significantly better diagnostic performance parameters than models trained using conventional data augmentation. The deep learning method proposed in this study could be implemented to assist radiologists in improving their diagnosis. Researchers and industry workers could overcome the lack of training data by employing our proposed automation pipeline approach in GAN-based synthetic data augmentation. This is anticipated to provide new means to overcome the shortage of graphic data for algorithm training. [ABSTRACT FROM AUTHOR]

• Deep-learning based GTV delineation modeling reader variability can predict confidence levels in GTV contours. • Learning of GTV confidence levels can be performed with a modified U-Net with an ordinal loss combined with a training strategy using discretization of target GTV confidence levels. • The agreement between CTV drawn from human and predicted GTV contours is even higher than the agreement between the GTV contours, approaching 90% Dice score. The delineation of the gross tumor volume (GTV) is a critical step for radiation therapy treatment planning. The delineation procedure is typically performed manually which exposes two major issues: cost and reproducibility. Delineation is a time-consuming process that is subject to inter- and intra-observer variability. While methods have been proposed to predict GTV contours, typical approaches ignore variability and therefore fail to utilize the valuable confidence information offered by multiple contours. In this work we propose an automatic GTV contouring method for soft-tissue sarcomas from X-ray computed tomography (CT) images, using deep learning by integrating inter- and intra-observer variability in the learned model. Sixty-eight patients with soft tissue and bone sarcomas were considered in this evaluation, all underwent pre-operative CT imaging used to perform GTV delineation. Four radiation oncologists and radiologists performed three contouring trials each for all patients. We quantify variability by defining confidence levels based on the frequency of inclusion of a given voxel into the GTV and use a deep convolutional neural network to learn GTV confidence maps. Results were compared to confidence maps from the four readers as well as ground-truth consensus contours established jointly by all readers. The resulting continuous Dice score between predicted and true confidence maps was 87% and the Hausdorff distance was 14 mm. Results demonstrate the ability of the proposed method to predict accurate contours while utilizing variability and as such it can be used to improve clinical workflow. [ABSTRACT FROM AUTHOR]

The purpose of this study was to identify aspects of impaired tongue motor performance that limit the ability to produce distinct speech sounds and contribute to reduced speech intelligibility in individuals with dysarthria secondary to amyotrophic lateral sclerosis (ALS). We analyzed simultaneously recorded tongue kinematic and acoustic data from 22 subjects during three target words (cat, dog, and took). The subjects included 11 participants with ALS and 11 healthy controls from the X-ray microbeam dysarthria database (Westbury, 1994). Novel measures were derived based on the range and speed of relative movement between two quasi-independent regions of the tongue – blade and dorsum – to characterize the global pattern of tongue dynamics. These "whole tongue" measures, along with the range and speed of single tongue regions, were compared across words, groups (ALS vs. control), and measure types (whole tongue vs. tongue blade vs. tongue dorsum). Reduced range and speed of both global and regional tongue movements were found in participants with ALS relative to healthy controls, reflecting impaired tongue motor performance in ALS. The extent of impairment, however, varied across words and measure types. Compared with the regional tongue measures, the whole tongue measures showed more consistent disease-related changes across the target words and were more robust predictors of speech intelligibility. Furthermore, these whole tongue measures were correlated with various word-specific acoustic features associated with intelligibility decline in ALS, suggesting that impaired tongue movement likely contributes to reduced phonetic distinctiveness of both vowels and consonants that underlie speech intelligibility decline in ALS. [ABSTRACT FROM AUTHOR]

Magnetic resonance (MR) imaging is becoming an established tool in capturing articulatory and physiological motion of the structures and muscles throughout the vocal tract and enabling visual and quantitative assessment of real-time speech activities. Although motion capture speed has been regularly improved by the continual developments in high-speed MR technology, quantitative analysis of multi-subject group data remains challenging due to variations in speaking rate and imaging time among different subjects. In this paper, a workflow of post-processing methods that matches different MR image datasets within a study group is proposed. Each subject's recorded audio waveform during speech is used to extract temporal domain information and generate temporal alignment mappings from their matching pattern. The corresponding image data are resampled by deformable registration and interpolation of the deformation fields, achieving inter-subject temporal alignment between image sequences. A four-dimensional dynamic MR speech atlas is constructed using aligned volumes from four human subjects. Similarity tests between subject and target domains using the squared error, cross correlation, and mutual information measures all show an overall score increase after spatiotemporal alignment. The amount of image variability in atlas construction is reduced, indicating a quality increase in the multi-subject data for groupwise quantitative analysis. [ABSTRACT FROM AUTHOR]

• GTV delineation of sarcoma is reader-dependent and subject to intra and inter-reader variability. • CT + MR + PET decreases inter- and intra-observer variability compared to CT + MR only. • Adding pre-existing PET images to delineation protocol reduces variability at no additional cost. • Adding pre-existing PET images to delineation protocol improves reproducibility. Accurate gross tumor volume (GTV) delineation is a critical step in radiation therapy treatment planning. However, it is reader dependent and thus susceptible to intra- and inter-reader variability. GTV delineation of soft tissue sarcoma (STS) often relies on CT and MR images. This study investigates the potential role of 18F-FDG PET in reducing intra- and inter-reader variability thereby improving reproducibility of GTV delineation in STS, without incurring additional costs or radiation exposure. Three readers performed independent GTV delineation of 61 patients with STS using first CT and MR followed by CT, MR, and 18F-FDG PET images. Each reader performed a total of six delineation trials, three trials per imaging modality group. Dice Similarity Coefficient (DSC) score and Hausdorff distance (HD) were used to assess both intra- and inter-reader variability using generated simultaneous truth and performance level estimation (STAPLE) GTVs as ground truth. Statistical analysis was performed using a Wilcoxon signed-ranked test. There was a statistically significant decrease in both intra- and inter-reader variability in GTV delineation using CT, MR 18F-FDG PET images vs. CT and MR images. This was translated by an increase in the DSC score and a decrease in the HD for GTVs drawn from CT, MR and 18F-FDG PET images vs. GTVs drawn from CT and MR for all readers and across all three trials. Incorporation of 18F-FDG PET into CT and MR images decreased intra- and inter-reader variability and subsequently increased reproducibility of GTV delineation in STS. [ABSTRACT FROM AUTHOR]

Magnetic resonance (MR) images with both high resolutions and high signal-to-noise ratios (SNRs) are desired in many clinical and research applications. However, acquiring such images takes a long time, which is both costly and susceptible to motion artifacts. Acquiring MR images with good in-plane resolution and poor through-plane resolution is a common strategy that saves imaging time, preserves SNR, and provides one viewpoint with good resolution in two directions. Unfortunately, this strategy also creates orthogonal viewpoints that have poor resolution in one direction and, for 2D MR acquisition protocols, also creates aliasing artifacts. A deep learning approach called SMORE that carries out both anti-aliasing and super-resolution on these types of acquisitions using no external atlas or exemplars has been previously reported but not extensively validated. This paper reviews the SMORE algorithm and then demonstrates its performance in four applications with the goal to demonstrate its potential for use in both research and clinical scenarios. It is first shown to improve the visualization of brain white matter lesions in FLAIR images acquired from multiple sclerosis patients. Then it is shown to improve the visualization of scarring in cardiac left ventricular remodeling after myocardial infarction. Third, its performance on multi-view images of the tongue is demonstrated and finally it is shown to improve performance in parcellation of the brain ventricular system. Both visual and selected quantitative metrics of resolution enhancement are demonstrated. [ABSTRACT FROM AUTHOR]

Purpose: Intrinsic and extrinsic tongue muscles in healthy and diseased populations vary both in their intra- and intersubject behaviors during speech. Identifying coordination patterns among various tongue muscles can provide insights into speech motor control and help in developing new therapeutic and rehabilitative strategies. Method: We present a method to analyze multisubject tongue muscle correlation using motion patterns in speech sound production. Motion of muscles is captured using tagged magnetic resonance imaging and computed using a phasebased deformation extraction algorithm. After being assembled in a common atlas space, motions from multiple subjects are extracted at each individual muscle location based on a manually labeled mask using high-resolution magnetic resonance imaging and a vocal tract atlas. Motion correlation between each muscle pair is computed within each labeled region. The analysis is performed on a population of 16 control subjects and 3 post-partial glossectomy patients. Results: The floor-of-mouth (FOM) muscles show reduced correlation comparing to the internal tongue muscles. Patients present a higher amount of overall correlation between all muscles and exercise en bloc movements. Conclusions: Correlation matrices in the atlas space show the coordination of tongue muscles in speech sound production. The FOM muscles are weakly correlated with the internal tongue muscles. Patients tend to use FOM muscles more than controls to compensate for their postsurgery function loss. [ABSTRACT FROM AUTHOR]

Amyotrophic Lateral Sclerosis (ALS) is a neurological disorder, which impairs tongue function for speech and swallowing. A widely used Diffusion Tensor Imaging (DTI) analysis pipeline is employed for quantifying differences in tongue fiber myoarchitecture between controls and ALS patients. This pipeline uses both high-resolution magnetic resonance imaging (hMRI) and DTI. hMRI is used to delineate tongue muscles, while DTI provides indices to reveal fiber connectivity within and between muscles. The preliminary results using five controls and two patients show quantitative differences between the groups. This work has the potential to provide insights into the detrimental effects of ALS on speech and swallowing. [ABSTRACT FROM AUTHOR]

Self-training is an important class of unsupervised domain adaptation (UDA) approaches that are used to mitigate the problem of domain shift, when applying knowledge learned from a labeled source domain to unlabeled and heterogeneous target domains. While self-training-based UDA has shown considerable promise on discriminative tasks, including classification and segmentation, through reliable pseudo-label filtering based on the maximum softmax probability, there is a paucity of prior work on self-training-based UDA for generative tasks, including image modality translation. To fill this gap, in this work, we seek to develop a generative self-training (GST) framework for domain adaptive image translation with continuous value prediction and regression objectives. Specifically, we quantify both aleatoric and epistemic uncertainties within our GST using variational Bayes learning to measure the reliability of synthesized data. We also introduce a self-attention scheme that de-emphasizes the background region to prevent it from dominating the training process. The adaptation is then carried out by an alternating optimization scheme with target domain supervision that focuses attention on the regions with reliable pseudo-labels. We evaluated our framework on two cross-scanner/center, inter-subject translation tasks, including tagged-to-cine magnetic resonance (MR) image translation and T1-weighted MR-to-fractional anisotropy translation. Extensive validations with unpaired target domain data showed that our GST yielded superior synthesis performance in comparison to adversarial training UDA methods. [ABSTRACT FROM AUTHOR]

Purpose We have recently reported a method for measuring rest-stress myocardial blood flow ( MBF) using a single, relatively short, PET scan session. The method requires two IV tracer injections, one to initiate rest imaging and one at peak stress. We previously validated absolute flow quantitation in ml/min/cc for standard bull's eye, segmental analysis. In this work, we extend the method for fast computation of rest-stress MBF parametric images. Methods We provide an analytic solution to the single-scan rest-stress flow model which is then solved using a two-dimensional table lookup method ( LM). Simulations were performed to compare the accuracy and precision of the lookup method with the original nonlinear method ( NLM). Then the method was applied to 16 single scan rest/stress measurements made in 12 pigs: seven studied after infarction of the left anterior descending artery ( LAD) territory, and nine imaged in the native state. Parametric maps of rest and stress MBF as well as maps of left ( f LV) and right ( f RV) ventricular spill-over fractions were generated. Regions of interest ( ROIs) for 17 myocardial segments were defined in bull's eye fashion on the parametric maps. The mean of each ROI was then compared to the rest ( K 1r) and stress ( K 1s) MBF estimates obtained from fitting the 17 regional TACs with the NLM. Results In simulation, the LM performed as well as the NLM in terms of precision and accuracy. The simulation did not show that bias was introduced by the use of a predefined two-dimensional lookup table. In experimental data, parametric maps demonstrated good statistical quality and the LM was computationally much more efficient than the original NLM. Very good agreement was obtained between the mean MBF calculated on the parametric maps for each of the 17 ROIs and the regional MBF values estimated by the NLM ( K 1map LM = 1.019 × K 1 ROI NLM + 0.019, R2 = 0.986; mean difference = 0.034 ± 0.036 mL/min/cc). Conclusions We developed a table lookup method for fast computation of parametric imaging of rest and stress MBF. Our results show the feasibility of obtaining good quality MBF maps using modest computational resources, thus demonstrating that the method can be applied in a clinical environment to obtain full quantitative MBF information. [ABSTRACT FROM AUTHOR]

Unsupervised domain adaptation (UDA) has been a vital protocol for migrating information learned from a labeled source domain to facilitate the implementation in an unlabeled heterogeneous target domain. Although UDA is typically jointly trained on data from both domains, accessing the labeled source domain data is often restricted, due to concerns over patient data privacy or intellectual property. To sidestep this, we propose "off-the-shelf (OS)" UDA (OSUDA), aimed at image segmentation, by adapting an OS segmentor trained in a source domain to a target domain, in the absence of source domain data in adaptation. Toward this goal, we aim to develop a novel batch-wise normalization (BN) statistics adaptation framework. In particular, we gradually adapt the domain-specific low-order BN statistics, e.g., mean and variance, through an exponential momentum decay strategy, while explicitly enforcing the consistency of the domain shareable high-order BN statistics, e.g., scaling and shifting factors, via our optimization objective. We also adaptively quantify the channel-wise transferability to gauge the importance of each channel, via both low-order statistics divergence and a scaling factor. Furthermore, we incorporate unsupervised self-entropy minimization into our framework to boost performance alongside a novel queued, memory-consistent self-training strategy to utilize the reliable pseudo label for stable and efficient unsupervised adaptation. We evaluated our OSUDA-based framework on both cross-modality and cross-subtype brain tumor segmentation and cardiac MR to CT segmentation tasks. Our experimental results showed that our memory consistent OSUDA performs better than existing source-relaxed UDA methods and yields similar performance to UDA methods with source data. • A novel source-free unsupervised domain adaptation (UDA) segmentation framework • Systematically exploration of domain-specific/shareable batch normalization statistics • A queued memory consistent self-training for stable and efficient progression • Evaluation on cross-modality/subtype brain tumor and cardiac MR to CT segmentation [ABSTRACT FROM AUTHOR]

Imaging and quantification of tongue anatomy is helpful in surgical planning, post-operative rehabilitation of tongue cancer patients, and studying of how humans adapt and learn new strategies for breathing, swallowing and speaking to compensate for changes in function caused by disease, medical interventions or aging. In vivo acquisition of high-resolution three-dimensional (3D) magnetic resonance (MR) images with clearly visible tongue muscles is currently not feasible because of breathing and involuntary swallowing motions that occur over lengthy imaging times. However, recent advances in image reconstruction now allow the generation of super-resolution 3D MR images from sets of orthogonal images, acquired at a high in-plane resolution and combined using super-resolution techniques. This paper presents, to the best of our knowledge, the first attempt towards automatic tongue muscle segmentation from MR images. We devised a database of ten super-resolution 3D MR images, in which the genioglossus and inferior longitudinalis tongue muscles were manually segmented and annotated with landmarks. We demonstrate the feasibility of segmenting the muscles of interest automatically by applying the landmark-based game-theoretic framework (GTF), where a landmark detector based on Haar-like features and an optimal assignment-based shape representation were integrated. The obtained segmentation results were validated against an independent manual segmentation performed by a second observer, as well as against B-splines and demons atlasing approaches. The segmentation performance resulted in mean Dice coefficients of 85.3%, 81.8%, 78.8% and 75.8% for the second observer, GTF, B-splines atlasing and demons atlasing, respectively. The obtained level of segmentation accuracy indicates that computerized tongue muscle segmentation may be used in surgical planning and treatment outcome analysis of tongue cancer patients, and in studies of normal subjects and subjects with speech and swallowing problems. [ABSTRACT FROM AUTHOR]