Search

Abstract Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease and the traditional variables extracted from computed tomography (CT) images may not be sufficient to describe all the topological features of lung tissues in COPD patients. We employed an unsupervised three-dimensional (3D) convolutional autoencoder (CAE)-feature constructor (FC) deep learning network to learn from CT data and derive tissue pattern-clusters jointly. We then applied exploratory factor analysis (EFA) to discover the unobserved latent traits (factors) among pattern-clusters. CT images at total lung capacity (TLC) and residual volume (RV) of 541 former smokers and 59 healthy non-smokers from the cohort of the SubPopulations and Intermediate Outcome Measures in the COPD Study (SPIROMICS) were analyzed. TLC and RV images were registered to calculate the Jacobian (determinant) values for all the voxels in TLC images. 3D Regions of interest (ROIs) with two data channels of CT intensity and Jacobian value were randomly extracted from training images and were fed to the 3D CAE-FC model. 80 pattern-clusters and 7 factors were identified. Factor scores computed for individual subjects were able to predict spirometry-measured pulmonary functions. Two factors which correlated with various emphysema subtypes, parametric response mapping (PRM) metrics, airway variants, and airway tree to lung volume ratio were discriminants of patients across all severity stages. Our findings suggest the potential of developing factor-based surrogate markers for new COPD phenotypes.

Abstract Fully CMR-guided electrophysiological interventions (EP-CMR) have recently been introduced but data on the optimal CMR imaging protocol are scarce. This study determined the clinical utility of 3D non-selective whole heart steady-state free precession imaging using compressed SENSE (nsWHcs) for automatic segmentation of cardiac cavities as the basis for targeted catheter navigation during EP-CMR cavo-tricuspid isthmus ablation. Fourty-two consecutive patients with isthmus-dependent right atrial flutter underwent EP-CMR radiofrequency ablations. nsWHcs succeeded in all patients (nominal scan duration, 98 ± 10 s); automatic segmentation/generation of surface meshes of right-sided cavities exhibited short computation times (16 ± 3 s) with correct delineation of right atrium, right ventricle, tricuspid annulus and coronary sinus ostium in 100%, 100%, 100% and 95%, respectively. Point-by-point ablation adhered to the predefined isthmus line in 62% of patients (26/42); activation mapping confirmed complete bidirectional isthmus block (conduction time difference, 136 ± 28 ms). nsWHcs ensured automatic and reliable 3D segmentation of targeted endoluminal cavities, multiplanar reformatting and image fusion (e.g. activation time measurements) and represented the basis for precise real-time active catheter navigation during EP-CMR ablations of isthmus-dependent right atrial flutter. Hence, nsWHcs can be considered a key component in order to advance EP-CMR towards the ultimate goal of targeted substrate-based ablation procedures.

Abstract Three-dimensional (3D) images provide a comprehensive view of material microstructures, enabling numerical simulations unachievable with two-dimensional (2D) imaging alone. However, obtaining these 3D images can be costly and constrained by resolution limitations. We introduce a novel method capable of generating large-scale 3D images of material microstructures, such as metal or rock, from a single 2D image. Our approach circumvents the need for 3D image data while offering a cost-effective, high-resolution alternative to existing imaging techniques. Our method combines a denoising diffusion probabilistic model with a generative adversarial network framework. To compensate for the lack of 3D training data, we implement chain sampling, a technique that utilizes the 3D intermediate outputs obtained by reversing the diffusion process. During the training phase, these intermediate outputs are guided by a 2D discriminator. This technique facilitates our method’s ability to gradually generate 3D images that accurately capture the geometric properties and statistical characteristics of the original 2D input. This study features a comparative analysis of the 3D images generated by our method, SliceGAN (the current state-of-the-art method), and actual 3D micro-CT images, spanning a diverse set of rock and metal types. The results shown an improvement of up to three times in the Frechet inception distance score, a typical metric for evaluating the performance of image generative models, and enhanced accuracy in derived properties compared to SliceGAN. The potential of our method to produce high-resolution and statistically representative 3D images paves the way for new applications in material characterization and analysis domains.

Abstract In this study, we aimed to create an 11-year hypertension risk prediction model using data from the Trøndelag Health (HUNT) Study in Norway, involving 17 852 individuals (20–85 years; 38% male; 24% incidence rate) with blood pressure (BP) below the hypertension threshold at baseline (1995–1997). We assessed 18 clinical, behavioral, and socioeconomic features, employing machine learning models such as eXtreme Gradient Boosting (XGBoost), Elastic regression, K-Nearest Neighbor, Support Vector Machines (SVM) and Random Forest. For comparison, we used logistic regression and a decision rule as reference models and validated six external models, with focus on the Framingham risk model. The top-performing models consistently included XGBoost, Elastic regression and SVM. These models efficiently identified hypertension risk, even among individuals with optimal baseline BP (

Abstract The inverse design of novel molecules with a desirable optoelectronic property requires consideration of the vast chemical spaces associated with varying chemical composition and molecular size. First principles-based property predictions have become increasingly helpful for assisting the selection of promising candidate chemical species for subsequent experimental validation. However, a brute-force computational screening of the entire chemical space is decidedly impossible. To alleviate the computational burden and accelerate rational molecular design, we here present an iterative deep learning workflow that combines (i) the density-functional tight-binding method for dynamic generation of property training data, (ii) a graph convolutional neural network surrogate model for rapid and reliable predictions of chemical and physical properties, and (iii) a masked language model. As proof of principle, we employ our workflow in the iterative generation of novel molecules with a target energy gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO).

Abstract Terpenoids are essential for plant growth, development, defense, and adaptation mechanisms. Psidium cattleyanum (Myrtaceae) is a fleshy fruit tree species endemics from Atlantic Forest, known for its pleasant fragrance and sweet taste, attributed to terpenoids in its leaves and fruits. In this study, we conducted genome-wide identification, evolutionary and expression analyses of the terpene synthase gene (TPS) family in P. cattleyanum red guava (var. cattleyanum), and yellow guava (var. lucidum Hort.) morphotypes. We identified 32 full-length TPS in red guava (RedTPS) and 30 in yellow guava (YlwTPS). We showed different expression patterns of TPS paralogous in the two morphotypes, suggesting the existence of distinct gene regulation mechanisms and their influence on the final essential oil content in both morphotypes. Moreover, the oil profile of red guava was dominated by 1,8-cineole and linalool and yellow guava was enriched in α-pinene, coincident in proportion to TPS-b1 genes, which encode enzymes that produce cyclic monoterpenes, suggesting a lineage-specific subfamily expansion of this family. Finally, we identified amino acid residues near the catalytic center and functional areas under positive selection. Our findings provide valuable insights into the terpene biosynthesis in a Neotropical Myrtaceae species and their potential involvement in adaptation mechanisms.

Abstract Obtaining an accurate segmentation of images obtained by computed microtomography (micro-CT) techniques is a non-trivial process due to the wide range of noise types and artifacts present in these images. Current methodologies are often time-consuming, sensitive to noise and artifacts, and require skilled people to give accurate results. Motivated by the rapid advancement of deep learning-based segmentation techniques in recent years, we have developed a tool that aims to fully automate the segmentation process in one step, without the need for any extra image processing steps such as noise filtering or artifact removal. To get a general model, we train our network using a dataset made of high-quality three-dimensional micro-CT images from different scanners, rock types, and resolutions. In addition, we use a domain-specific augmented training pipeline with various types of noise, synthetic artifacts, and image transformation/distortion. For validation, we use a synthetic dataset to measure accuracy and analyze noise/artifact sensitivity. The results show a robust and accurate segmentation performance for the most common types of noises present in real micro-CT images. We also compared the segmentation of our method and five expert users, using commercial and open software packages on real rock images. We found that most of the current tools fail to reduce the impact of local and global noises and artifacts. We quantified the variation on human-assisted segmentation results in terms of physical properties and observed a large variation. In comparison, the new method is more robust to local noises and artifacts, outperforming the human segmentation and giving consistent results. Finally, we compared the porosity of our model segmented images with experimental porosity measured in the laboratory for ten different untrained samples, finding very encouraging results.

Abstract Atrial fibrillation (AF) leads to remodeling characterized by changes in both size and shape of the left atrium (LA). Here we aimed to study the effect of hypertrophic cardiomyopathy (HCM) on the pattern of LA remodeling in AF-patients. HCM-patients (n = 23) undergoing AF ablation (2009–2012) were matched and compared with 125 Non-HCM patients from our prospective registry. Pre-procedural CT data were analyzed (EnSite Verismo, SJM, MN) to determine the maximal sagittal (anterior–posterior, AP), coronal (superior-inferior, SI and transversal, TV) dimensions and the sphericity index (LAS). Volume (LAV) was rendered after appendage (LAA) and pulmonary vein (PV) exclusion. A cutting plane, between PV ostia/LAA and parallel to the posterior wall, divided LAV into anterior- (LA­A) and posterior-LA (LA­P) parts. The ratio LA-A/LAV was defined as asymmetry index (ASI). HCM patients had a wider inter-ventricular septum and a smaller LV than Non-HCM patients. LA volume (LAV 166 ± 72 vs. 130 ± 36 ml, p = 0.03) and LA diameters were significantly larger in HCM patients. Anterior volume (LA-A: 112 ± 48 vs. 83 ± 26 ml, p

Abstract Joining titanium alloy and stainless steel is becoming an urgent need since their outstanding mechanical properties can be utilized integratedly. However, direct fusion joining of Ti6Al4V to SS316 can cause brittle Ti-Fe intermetallics which compromise join bonds’ mechanical properties. In this research, Laser 3D Printing was applied to explore a new Ti6Al4V to SS316 multi-metallic structure. A novel filler transition route was introduced (Ti6Al4V → V → Cr → Fe → SS316) to avoid the Ti-Fe intermetallics. Two experimental cases were performed for comparison to evaluate this novel route’s effect. In the first case, SS316 layer was directly deposited on Ti6Al4V substrate by laser 3D printing, but the sample cracked in the printing process. Then fracture morphology, phase identification, and micro-hardness were analyzed. In the second case, a multi-metallic structure was fabricated via laser 3D printing following the transition route. Microstructure characterization and composition distribution were analyzed via scanning electron microscope(SEM) and energy dispersive spectrometry(EDS). x-ray diffraction(XRD) tests demonstrated the intermetallics were effectively avoided following the transition route. Vickers hardness number(VHN) showed no significant hard brittle phases in the sample. Comparing with directly depositing SS316 on Ti6Al4V, the usage of the novel transition route can eliminate the intermetallics effectively. These research results are good contributions in joining titanium alloy and stainless steel.