Mean Contour Research Articles

The developing of prosody in infants: a longitudinal study over the first 16 months of infant life

Background The acquisition and development of prosodic aspects of vocal intonation are of special interest within the larger context of language acquisition. Method The present study explored the developmental trajectories of infant prosodic abilities from 4 to 16 months of life with an intensive time points assessment. Several aspects were considered: an acoustic analysis of infant vocal productions with specific software, the analyses of all the prosodic variables associated with the fundamental frequency (F0 mean, F0 range, and F0 final contours), the individual variability and the complexity of the vocal productions of the infants. Results The multi-level analysis evidenced specific prosodic developmental trajectories that differ for the different kind of vocal productions since the first months of life. Conclusions The findings suggest that in the second half of the first year of life infants show an intonational repertoire that may help manage interactions with their caregiver and that individual variability has to be taken in consideration when assessing infants’ prosody.

Cryptic Extensibility in von Willebrand Factor Revealed by Molecular Nanodissection.

Von Willebrand factor (VWF) is a multimer with a variable number of protomers, each of which is a head-to-head dimer of two multi-domain monomers. VWF responds to shear through the unfolding and extension of distinct domains, thereby mediating platelet adhesion and aggregation to the injured blood vessel wall. VWF's C1-6 segment uncoils and then the A2 domain unfolds and extends in a hierarchical and sequential manner. However, it is unclear whether there is any reservoir of further extensibility. Here, we explored the presence of cryptic extensibility in VWF by nanodissecting individual, pre-stretched multimers with atomic force microscopy (AFM). The AFM cantilever tip was pressed into the surface and moved in a direction perpendicular to the VWF axis. It was possible to pull out protein loops from VWF, which resulted in a mean contour length gain of 217 nm. In some cases, the loop became cleaved, and a gap was present along the contour. Frequently, small nodules appeared in the loops, indicating that parts of the nanodissected VWF segment remained folded. After analyzing the nodal structure, we conclude that the cryptic extensibility lies within the C1-6 and A1-3 regions. Cryptic extensibility may play a role in maintaining VWF's functionality in extreme shear conditions.

Investigation of the local apparent viscosity in a stirred tank with a shear-thinning fluid through particle image velocimetry

The local apparent viscosity in a cylindrical stirred tank was investigated using Particle Image Velocimetry in a shear-thinning fluid. Two impellers were studied, a Rushton turbine for validation and an axial impeller HTPG4, for three Reynolds numbers corresponding to the transition regime between laminar and turbulent flows. Mean and instantaneous velocity contours and profiles are analysed. Then, the different contributions to the mean total kinetic energy are investigated with the HTPG4 impeller. The sensitivity of the mean apparent viscosity to the method of determining the shear rate is discussed. This work shows that the mean apparent viscosity is highly sensitive to the method of determining the shear rate. Future work will focus on improving the measurement of the local shear rate.

4D-Precise: Learning-based 3D motion estimation and high temporal resolution 4DCT reconstruction from treatment 2D+t X-ray projections

Background and ObjectiveIn radiotherapy treatment planning, respiration-induced motion introduces uncertainty that, if not appropriately considered, could result in dose delivery problems. 4D cone-beam computed tomography (4D-CBCT) has been developed to provide imaging guidance by reconstructing a pseudo-motion sequence of CBCT volumes through binning projection data into breathing phases. However, it suffers from artefacts and erroneously characterizes the averaged breathing motion. Furthermore, conventional 4D-CBCT can only be generated post-hoc using the full sequence of kV projections after the treatment is complete, limiting its utility. Hence, our purpose is to develop a deep-learning motion model for estimating 3D+t CT images from treatment kV projection series. MethodsWe propose an end-to-end learning-based 3D motion modelling and 4DCT reconstruction model named 4D-Precise, abbreviated from Probabilistic reconstruction of image sequences from CBCT kV projections. The model estimates voxel-wise motion fields and simultaneously reconstructs a 3DCT volume at any arbitrary time point of the input projections by transforming a reference CT volume. Developing a Torch-DRR module, it enables end-to-end training by computing Digitally Reconstructed Radiographs (DRRs) in PyTorch. During training, DRRs with matching projection angles to the input kVs are automatically extracted from reconstructed volumes and their structural dissimilarity to inputs is penalised. We introduced a novel loss function to regulate spatio-temporal motion field variations across the CT scan, leveraging planning 4DCT for prior motion distribution estimation. ResultsThe model is trained patient-specifically using three kV scan series, each including over 1200 angular/temporal projections, and tested on three other scan series. Imaging data from five patients are analysed here. Also, the model is validated on a simulated paired 4DCT-DRR dataset created using the Surrogate Parametrised Respiratory Motion Modelling (SuPReMo). The results demonstrate that the reconstructed volumes by 4D-Precise closely resemble the ground-truth volumes in terms of Dice, volume similarity, mean contour distance, and Hausdorff distance, whereas 4D-Precise achieves smoother deformations and fewer negative Jacobian determinants compared to SuPReMo. ConclusionsUnlike conventional 4DCT reconstruction techniques that ignore breath inter-cycle motion variations, the proposed model computes both intra-cycle and inter-cycle motions. It represents motion over an extended timeframe, covering several minutes of kV scan series.

A two-step deep learning method for 3DCT-2DUS kidney registration during breathing

This work proposed KidneyRegNet, a novel deep registration pipeline for 3D CT and 2D U/S kidney scans of free breathing, which comprises a feature network, and a 3D–2D CNN-based registration network. The feature network has handcrafted texture feature layers to reduce the semantic gap. The registration network is an encoder-decoder structure with loss of feature-image-motion (FIM), which enables hierarchical regression at decoder layers and avoids multiple network concatenation. It was first pretrained with a retrospective dataset cum training data generation strategy and then adapted to specific patient data under unsupervised one-cycle transfer learning in onsite applications. The experiment was performed on 132 U/S sequences, 39 multiple-phase CT and 210 public single-phase CT images, and 25 pairs of CT and U/S sequences. This resulted in a mean contour distance (MCD) of 0.94 mm between kidneys on CT and U/S images and MCD of 1.15 mm on CT and reference CT images. Datasets with small transformations resulted in MCDs of 0.82 and 1.02 mm, respectively. Large transformations resulted in MCDs of 1.10 and 1.28 mm, respectively. This work addressed difficulties in 3DCT-2DUS kidney registration during free breathing via novel network structures and training strategies.

Fully Automatic initialization and segmentation of left and right ventricles for large-scale cardiac MRI using a deeply supervised network and 3D-ASM

Background and objectiveThe sheer volume of data generated by population imaging studies is unparalleled by current capabilities to extract objective and quantitative cardiac phenotypes; subjective and time-consuming manual image analysis remains the gold standard. Automated image analytics to compute quantitative imaging biomarkers of cardiac function are desperately needed. Data volumes and their variability pose a challenge to most state-of-the-art methods for endo- and epicardial contours, which lack robustness when applied to very large datasets. Our aim is to develop an analysis pipeline for the automatic quantification of cardiac function from cine magnetic resonance imaging data. MethodThis work adopt 4,638 cardiac MRI cases coming from UK Biobank with ground truth available for left and RV contours. A hybrid and robust algorithm is proposed to improve the accuracy of automatic left and right ventricle segmentation by harnessing the localization accuracy of deep learning and the morphological accuracy of 3D-ASM (three-dimensional active shape models). The contributions of this paper are three-fold. First, a fully automatic method is proposed for left and right ventricle initialization and cardiac MRI segmentation by taking full advantage of spatiotemporal constraint. Second, a deeply supervised network is introduced to train and segment the heart. Third, the 3D-ASM image search procedure is improved by combining image intensity models with convolutional neural network (CNN) derived distance maps improving endo- and epicardial edge localization. ResultsThe proposed architecture outperformed the state of the art for cardiac MRI segmentation from UK Biobank. The statistics of RV landmarks detection errors for Triscuspid valve and RV apex are 4.17 mm and 5.58 mm separately. The overlap metric, mean contour distance, Hausdorff distance and cardiac functional parameters are calculated for the LV (Left Ventricle) and RV (Right Ventricle) contour segmentation. Bland–Altman analysis for clinical parameters shows that the results from our automated image analysis pipelines are in good agreement with results from expert manual analysis. ConclusionsOur hybrid scheme combines deep learning and statistical shape modeling for automatic segmentation of the LV/RV from cardiac MRI datasets is effective and robust and can compute cardiac functional indexes from population imaging.

Fuzzy logic optimization with regression analysis on EDM machining parameters of Si3N4-TiN ceramic composites

Electro discharge machining (EDM) is a cycle for molding tough materials and framing profound contour formed openings by warm disintegration in all sort of electrically conductive materials. The goal of the venture to be concentrating because of working parameters of EDM for machining of silicon nitride-titanium nitride in the machining qualities with copper electrode, for example input Spark on time (Son), current (Ip), Spark off time (Soff), spark gap and dielectric pressure on the metal removal rate (MRR) and Electrode Wear Rate (EWR) were analyzed. Subsequently, using Taguchi analysis of various plots like Mean effect plots, Interaction plots, and contour plots, performance characteristics are looked at in relation to multiple process factors. Fuzzy logic and Regression analysis is utilized to combine various reactions into a solitary trademark record known as the Multi Response Performance Index (MRPI).The trial and anticipated qualities were in a decent programming instrument for discovering the MRPI esteem. For numerous performance aspects, such as material removal rate, electrode wear rate and so on, the optimal process parameter combination was established using fuzzy logic analysis. The key process factors, which included spark off time and current, were found using an ANOVA based on a fuzzy algorithm. Topography on machined surface and cross-sectional view of conductive Si3N4-TiN composite and surface characteristics of machined electrode is examined by SEM analysis and identified the best hole surface and worst hole surface. Sensitivity analysis is being utilized to determine how much the input values, such as Ip, Son and Soff, will need to alter in order to get the desired, optimal result. In the complexity analysis, each constraint of the machine, composite and process is addressed. Future researches might look into various electrodes to assess geometrical tolerances including angularity, parallelism, total run out, flatness, straightness, concentricity, and line profile employing other optimization methodologies to achieve the best outcome. The findings of the confirmatory experiment have been established, indicating that it may be feasible to successfully strengthen the spark eroding technique.

Left ventricular myocardial motion tracking in cardiac cine magnetic resonance images based on a biomechanical model.

Cardiac cine magnetic resonance (CCMR) imaging plays an important role in the clinical cardiovascular disease (CVD) examination and evaluation. To accurately reconstruct the displacement field and describe the motion of the left ventricular myocardium (LVM), this study proposes and tests a new approach for tracking myocardial motion of the left ventricle based on a biomechanical model. CCMR imaging data acquired from 103 patients are enrolled, including two simulated and 101 clinical data. A non-rigid image registration method with a combination of a thin-plate spline function and random sample consensus is used to recover the observed displacement field of LVM. Next, a biomechanical model and a material matrix are introduced to solve the dense displacement field of LVM using a finite element framework. Then, the tracking precision and error of results for the two groups are analyzed. Displacement results of the simulated data show correlation coefficient≥0.876 and mean square error≤0.159, while displacement results of the clinical data show Dice≥0.97 and mean contour distance≤0.464. Additionally, the strain results show correlation coefficient≥0.717. This study demonstrates that the proposed new method enables to accurately track the motion of the LVM and evaluate strain, which has clinical auxiliary value in the diagnosis of CVD.

Image Quality Metrics, Personality Traits, and Subjective Evaluation of Indoor Environment Images

Adaptive lighting systems can be designed to detect the spatial characteristics of the visual environment and adjust the light output to increase visual comfort and performance. Such systems would require computational metrics to estimate occupants’ visual perception of indoor environments. This paper describes an experimental study to investigate the relationship between the perceived quality of indoor environments, personality, and computational image quality metrics. Forty participants evaluated the visual preference, clarity, complexity, and colorfulness of 50 images of indoor environments. Twelve image quality metrics (maximum local variation (MLV), spatial frequency slope (α), BRISQUE, entropy (S), ITU spatial information (SI), visual complexity (Rspt), colorfulness (M), root mean square (RMS) contrast, Euler, energy (E), contour, and fractal dimension) were used to estimate participants’ subjective evaluations. While visual clarity, visual complexity, and colorfulness could be estimated using at least one metric, none of the metrics could estimate visual preference. The results indicate that perceived colorfulness is highly correlated with perceived clarity and complexity. Personality traits tested by the 10-item personality inventory (TIPI) did not impact the subjective evaluations of the indoor environmental images. Future studies will explore the impact of target and background luminance on the perceived quality of indoor images.

Study of wind loads on rectangular plan tall building under interference condition

The change in distribution of wind pressure due to alteration of wind flow pattern caused by the existence of an interfering building is explored in this study on a principal tall building having cross-section of rectangular shape. Interfering building having same cross-sectional shape as the principal building but a variation in height is placed in-front of the principal building, with across-wind lengths of two buildings being in-line with each other. Keeping the principal building fixed in position, the spacing between two buildings is varied by changing the position of interfering building, so as to understand the effect of spacing and height variation upon the wind loads and pressures. Two independent type of measurements are undertaken for varying interfering building height: (1). Force Measurement (2). Pressure Measurement. For purpose of validation, the results obtained from two independent measurements mentioned are compared later. The results obtained from force measurements are expressed in the form of two-dimensional X-Y plots, while that for pressure measurement are depicted in the form of mean and RMS wind pressure coefficient (CP and CP') contour plots on all vertical faces of the principal building. Force measurement results show that the interference effect subsides as the interfering building height is reduced and spacing is increased. However, from pressure contours it is evident that when the interfering building is close to principal building pressure force on the windward face reduces while the suction on the leeward face increases with the reduction in interfering building height. Negative drag is also observed for few of the cases.

The real-time interaction model for transient mode investigations of a dual-piston free-piston engine generator

A free-piston engine generator is being developed worldwide as a novel means of electrical power generator for light vehicle application, which can provide a complementary solution towards reducing carbon dioxide emissions of road vehicles in general. However, the absence of a crank-slider mechanism in free-piston engines resulted in poor transient operations. Transient real-time modelling and simulation can provide the interaction between in-cylinder combustion with dynamics of piston motion, which has not been fully explored previously. This paper presents a real-time interaction model between dynamics of the piston motion and thermodynamics of the in-cylinder combustion of a two-stroke spark ignition dual-piston free-piston engine generator for transient operation investigations. The simulation model was developed based on the working prototype and comprised of zero- and one-dimensional sub-models interacted in real-time governed by a single timestep. The study focuses on three critical transient modes: motoring, starting and generating for the transient performance investigation. A series of experimental results during motoring was used for validating the simulation model, which showed good agreement between simulation and experiment results with 2–5% errors. The targeted brake thermal efficiency is around 20–30% at 50–60 Hz engine speed which has been shown achievable. Transient speeds of 10 and 25 Hz produced higher combustion pressure around the top dead centre but suffered pumping loss at the end of the expansion stroke. The lower peak pressures at 50 and 60 Hz have contributed to the lower brake mean effective pressure values. Maximum brake thermal efficiency of 25.8% lies in the mid-range of compression ratios between 10:1 to 15:1 while at the cyclic speeds of 50 Hz to 60 Hz. Maximum brake mean effective pressure contour occurs between 5 and 12.5 compression ratios and corresponding speeds of 10 to 30 Hz. It was observed that the excess energy from combustion results in piston overshoot condition while insufficient energy prevents the piston from achieving the targeted stroke and compression ratio. Knock events were observed for high compression ratio cases regardless of engine speed. Predictive stroke control and knock detection capability are the main contribution of the real-time interaction model presented in this paper for realistic transient operation investigation and performance prediction of the dual-piston free-piston engine generator.

Molecular clustering based on gene set expression and its relationship with prognosis in patients with lung adenocarcinoma.

BackgroundLung adenocarcinoma (LUAD) is a subtype of lung cancer with high morbidity and mortality. While genotyping is an important determinant for the prognosis of LUAD patients, there is a paucity of studies on gene set-based expression (GSE) typing for LUAD. This current study used GSE methodology to perform gene typing of LUAD patients.MethodsClinical and genomic information of the LUAD patients were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Patients with LUAD were clustered into different molecular subtypes depending on the clinical and gene set expression characteristics. The survival rate and silhouette widths were compared between each molecular subtype. Differences in survival rate between gene sets were analyzed using Kaplan-Meier survival curves. Cox regression and Lasso regression were used to establish the prognostic gene set model based on the TCGA database, and the results were validated using the GEO dataset.ResultsA total of 10 hub genes were finally identified and clustered into 3 subtypes with a mean contour width of 0.96. There were significant differences in survival rates among the 3 subtypes (P<0.05). Gene Ontology (GO) analysis indicated that the related biological processes (BP) were mainly involved in regulation of cell cycle, mitotic cell cycle phase transition, and proteasome-mediated ubiquitin-dependent protein catabolic process. The cellular components (CC) were related to the spindle, chromosomal region, and midbody. Molecular function (MF) mainly focused on ubiquitin-like protein ligase binding, translation regulator activity, and oxidation activity. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showed that the main pathways included the Epstein Barr virus infection pathway of neurogeneration, the p53 signaling pathway, and the proteome pathways. In addition, the protein-protein interaction network was analyzed using the STRING and Cytospace software, and the top 9 hub genes identified were KIF2C, DLGAP5, KIF20A, PSMC1, PSMD1, PSMB7, SNAI2, FGF13, and BMP2.ConclusionsPatients with LUAD can be clustered into three subtypes based on the expression of gene sets. These findings contribute to understanding the pathogenesis and molecular mechanisms in LUAD, and may lead to potential individualized pharmacogenetic therapy for patients with LUAD.

Evaluation of Auto-Contouring and Dose Distributions for Online Adaptive Radiation Therapy of Patients With Locally Advanced Lung Cancers

PurposeRetrospective studies were performed to evaluate the accuracy of automatically mapped structures and dosimetric consequences of daily online adaptive radiation therapy (ART) for lung cancer treatments. Methods and MaterialsTen patients with locally advanced lung cancer (prescription = 2 Gy × 30) with 297 fractions of treatment were selected for this retrospective study on a research emulator (Ethos, Varian Medical Systems). All adaptive treatments were simulated twice: automatic-ART (A-ART), automatic contours were used without modification, and supervised-ART (S-ART), automatic contours were modified manually by physicians and physicists. Dosimetric results were analyzed by relating supervised scheduled (S-SCH) dose (initial baseline reference plan delivered on daily anatomy and supervised contour correction without any adaptation), A-ART and S-ART to the initial baseline reference dose. ResultsTwo hundred ninety (of 297) fractions were analyzed. Comparing target volumes between A-ART and S-ART, dice similarity coefficient was 0.93 ± 0.05, mean contour distance was 1.5 ± 1.2 mm, and Hausdorff distance was 4.0 ± 2.3 mm. Analysis of daily results over 290 fractions of treatment showed that average target coverage improved from 0.96 ± 0.04 (S-SCH) to 1.00 ± 0.02 (A-ART) and 1.02 ± 0.04 (S-ART); average upper dose constraint was reduced from 1.01 ± 0.11 (S-SCH) to 0.94 ± 0.10 (A-ART) and 0.93 ± 0.12 (S-ART). A-ART and S-ART improved planning target volume minimum doses by 4.85 ± 3.03 Gy (P = .049) and 4.46 ± 8.99 Gy (P = .058), respectively. Statistical analysis shows that A-ART and S-ART significantly improved cumulative target dose by 0.033 ± 0.087 (P = .002) and 0.032 ± 0.086 (P = .003) and reduced upper constraints by 0.033 ± 0.072 (P < .001) and 0.032 ± 0.072 (P < .001) relative to S-SCH dose results, respectively. ConclusionsAccuracy of Ethos automatic contouring for lung cancer is considered clinically acceptable. The online adaptive radiation therapy improves target coverage and spares organs-at-risk significantly.

Cortical Morphological Changes in Congenital Amusia: Surface-Based Analyses

Background: Congenital amusia (CA) is a rare disorder characterized by deficits in pitch perception, and many structural and functional magnetic resonance imaging studies have been conducted to better understand its neural bases. However, a structural magnetic resonance imaging analysis using a surface-based morphology method to identify regions with cortical features abnormalities at the vertex-based level has not yet been performed.Methods: Fifteen participants with CA and 13 healthy controls underwent structural magnetic resonance imaging. A surface-based morphology method was used to identify anatomical abnormalities. Then, the surface parameters' mean value of the identified clusters with statistically significant between-group differences were extracted and compared. Finally, Pearson's correlation analysis was used to assess the correlation between the Montreal Battery of Evaluation of Amusia (MBEA) scores and surface parameters.Results: The CA group had significantly lower MBEA scores than the healthy controls (p = 0.000). The CA group exhibited a significant higher fractal dimension in the right caudal middle frontal gyrus and a lower sulcal depth in the right pars triangularis gyrus (p < 0.05; false discovery rate-corrected at the cluster level) compared to healthy controls. There were negative correlations between the mean fractal dimension values in the right caudal middle frontal gyrus and MBEA score, including the mean MBEA score (r = −0.5398, p = 0.0030), scale score (r = −0.5712, p = 0.0015), contour score (r = −0.4662, p = 0.0124), interval score (r = −0.4564, p = 0.0146), rhythmic score (r = −0.5133, p = 0.0052), meter score (r = −0.3937, p = 0.0382), and memory score (r = −0.3879, p = 0.0414). There was a significant positive correlation between the mean sulcal depth in the right pars triangularis gyrus and the MBEA score, including the mean score (r = 0.5130, p = 0.0052), scale score (r = 0.5328, p = 0.0035), interval score (r = 0.4059, p = 0.0321), rhythmic score (r = 0.5733, p = 0.0014), meter score (r = 0.5061, p = 0.0060), and memory score (r = 0.4001, p = 0.0349).Conclusion: Individuals with CA exhibit cortical morphological changes in the right hemisphere. These findings may indicate that the neural basis of speech perception and memory impairments in individuals with CA is associated with abnormalities in the right pars triangularis gyrus and middle frontal gyrus, and that these cortical abnormalities may be a neural marker of CA.

Ultrasound assessment of abdominal adipose panniculus in patients treated with a single session of cryolipolysis in a clinical setting.

ObjectiveTo assess the effectiveness and safety of cryolipolysis treatment (four‐to‐six cycles in one session) for reducing abdominal fat.MethodsRetrospective study conducted on consecutive healthy women who underwent one session of cryolipolysis for abdominal fat reduction. Ultrasound images (USI) were acquired at baseline and at month 6 after treatment. A patient‐tailored approach based on adipose tissue depth and architecture was selected.ResultsThirty women were included. The mean age was 51.4 ± 3.4 years, and 27 (90%) were menopausal. At month 6, the mean abdominal fat layer thickness reduction was 46.6% (41.2%–51.9%). The mean contour was significantly reduced from 84.3, 99.2, 90.6, and 97.1 cm to 81.0, 93.6, 85.8, and 92.2 cm in the infracostal, supraumbilical, umbilical, and ischiopubic regions, respectively; p < 0.0001 each. There were no significant changes in weight, body mass index, fat mass, or lean mass throughout the study. After adjusting for smoking, exercise, hypothyroidism, and type of fat, there was a significant reduction in abdominal fat layer thickness (−4.5 ± 0.9 mm, p < 0.0001) and contour measurements in infracostal (−3.3 ± 2.4 mm, p = 0.0317); supraumbilical (−4.8 ± 2.1 mm, p = 0.0254); umbilical (−5.6 ± 2.3, p = 0.0161); and ischiopubic (−4.9 ± 1.8 mm, p = 0.0080). The qualitative analysis of the USI suggested the appearance of numerous bands of ordered and structured collagen fibers.ConclusionsFour‐to‐six cycles of cryolipolysis, administered in one session, seemed to be a safe and effective procedure for reducing localized fat in the abdominal region in this sample. Additionally, ultrasound images suggested that cryolipolysis was associated with a skin quality improvement and neocollagenesis.

Deformable registration of lateral cephalogram and cone-beam computed tomography image.

This study aimed to design and evaluate a novel method for the registration of 2D lateral cephalograms and 3D craniofacial cone-beam computed tomography (CBCT) images, providing patient-specific 3D structures from a 2D lateral cephalogram without additional radiationexposure. We developed a cross-modal deformable registration model based on a deep convolutional neural network. Our approach took advantage of a low-dimensional deformation field encoding and an iterative feedback scheme to infer coarse-to-fine volumetric deformations. In particular, we constructed a statistical subspace of deformation fields and parameterized the nonlinear mapping function from an image pair, consisting of the target 2D lateral cephalogram and the reference volumetric CBCT, to a latent encoding of the deformation field. Instead of the one-shot registration by the learned mapping function, a feedback scheme was introduced to progressively update the reference volumetric image and to infer coarse-to-fine deformations fields, accounting for the shape variations of anatomical structures. A total of 220 clinically obtained CBCTs were used to train and validate the proposed model, among which 120 CBCTs were used to generate a training dataset with 24k paired synthetic lateral cephalograms and CBCTs. The proposed approach was evaluated on the deformable 2D-3D registration of clinically obtained lateral cephalograms and CBCTs from growing and adult orthodonticpatients. Strong structural consistencies were observed between the deformed CBCT and the target lateral cephalogram in all criteria. The proposed method achieved state-of-the-art performances with the mean contour deviation of 0.41 0.12 mm on the anterior cranial base, 0.48 0.17 mm on the mandible, and 0.35 0.08 mm on the maxilla, respectively. The mean surface mesh ranged from 0.78 to 0.97 mm on various craniofacial structures, and the LREs ranged from 0.83 to 1.24 mm on the growing datasets regarding 14 landmarks. The proposed iterative feedback scheme handled the structural details and improved the registration. The resultant deformed volumetric image was consistent with the target lateral cephalogram in both 2D projective planes and 3D volumetric space regarding the multicategory craniofacialstructures. The results suggest that the deep learning-based 2D-3D registration model enables the deformable alignment of 2D lateral cephalograms and CBCTs and estimates patient-specific 3D craniofacial structures.

Symmetric mean and directional contour pattern for texture classification

In this letter, we propose a simple yet effective texture descriptor, symmetric mean and directional contour pattern (SMDCP), for texture classification. In particular, first the robust symmetric mean pattern (RSMP) that extracts the sign and amplitude information of the local difference through the neighbourhood average in a new scheme of encoding to further enhance the robustness to noise is constructed. Then a local directional and contour pattern (LDCP) to represent the contour information and direction information of adjacent sampling points is extracted. By concatenating the RSMP and LDCP, a robust and effective texture descriptor (SMDCP) for classification is built. Experimental results reveal that the proposed method obtains significant performance and high discrimination in comparison with 10 representative approaches.

Bézier curves as a total approach to measure the upper lid contour: redefining clinical outcomes in palpebral surgery

AimsTo define and quantify the upper lid contour by adapting Bézier curves with a newly developed software in normal subjects, assessing their reliability.MethodsFifty eyes of 50 healthy patients with no...

Adoption of Snake Variable Model-Based Method in Segmentation and Quantitative Calculation of Cardiac Ultrasound Medical Images.

This paper intends to explore the effect of the enhanced snake variable model in the segmentation of cardiac ultrasound images and its adoption in quantitative measurement of cardiac cavity. First, the basic principles of the traditional snake model and the gradient vector flow (GVF) snake model are explained. Then, an ellipsoid model is constructed to obtain the initial contour of the heart based on the three-dimensional volume of cardiac ultrasound image, and a discretized triangular mesh model is generated. Finally, the vortical gradient vector flow (VGVF) external force field is introduced and combined with the greedy algorithm to process the deformation of the initial ellipsoid contour of the heart. The segmentation effect is quantitatively evaluated regarding the area overlap rate (AOR) and the mean contour distance (MCD). The results show that the VGVF snake model can segment the deep recessed area of the “U-shaped map” in contrast to the traditional snake model and the GVF snake model. After being applied to ultrasonic image segmentation, the VGVF snake model obtains the segmentation result that is close to the doctor's manual segmentation result, and the average AOR and MCD are 97.4% and 3.2, respectively. The quantitative evaluation of the cardiac cavity is carried out based on the segmentation results, and the measurement of the volume change of the left ventricle within a cardiac cycle is realized. To sum up, VGVF snake model is superior to the traditional snake and GVF snake models in terms of ultrasonic image segmentation, which realizes the three-dimensional segmentation and quantitative calculation of the cardiac cavity.

Singing Transcription from Polyphonic Music Using Melody Contour Filtering

Automatic singing transcription and analysis from polyphonic music records are essential in a number of indexing techniques for computational auditory scenes. To obtain a note-level sequence in this work, we divide the singing transcription task into two subtasks: melody extraction and note transcription. We construct a salience function in terms of harmonic and rhythmic similarity and a measurement of spectral balance. Central to our proposed method is the measurement of melody contours, which are calculated using edge searching based on their continuity properties. We calculate the mean contour salience by separating melody analysis from the adjacent breakpoint connective strength matrix, and we select the final melody contour to determine MIDI notes. This unique method, combining audio signals with image edge analysis, provides a more interpretable analysis platform for continuous singing signals. Experimental analysis using Music Information Retrieval Evaluation Exchange (MIREX) datasets shows that our technique achieves promising results both for audio melody extraction and polyphonic singing transcription.