Automatic integration of numerical formats examined with frequency-tagged EEG

How humans integrate and abstract numerical information across different formats is one of the most debated questions in human cognition. We addressed the neuronal signatures of the numerical integration using an EEG technique tagged at the frequency of visual stimulation. In an oddball design, participants were stimulated with standard sequences of numbers (< 5) depicted in single (digits, dots, number words) or mixed notation (dots—digits, number words—dots, digits—number words), presented at 10 Hz. Periodically, a deviant stimulus (> 5) was inserted at 1.25 Hz. We observed significant oddball amplitudes for all single notations, showing for the first time using this EEG technique, that the magnitude information is spontaneously and unintentionally abstracted, irrespectively of the numerical format. Significant amplitudes were also observed for digits—number words and number words—dots, but not for digits—dots, suggesting an automatic integration across some numerical formats. These results imply that direct and indirect neuro-cognitive links exist across the different numerical formats.

Design of Automatic Integration Algorithm for Popular Science Microanimation Works in the Context of New Media

When traditional methods integrate popular science microanimation works, the integration effect of the animation works is not good. In this paper, we propose an automatic integration algorithm of popular science microanimation works in the context of new media. The system first analyzes the characteristics of the new media context and gives the meaning of microanimation in the context of new media. It simplifies the edge folding of popular science microanimation integration and calculates the Facial Animation Parameter (FAP) value to realize the automatic integration of popular science microanimation works. We conducted a number of experiments using various size datasets to test the proposed system. We achieved an average integration accuracy of 96.3% with datasets of 500 to 3000 animation works, having the highest accuracy of 99% with a dataset of 500 animation works. On the other hand, the integration time of the animation works was recorded just 1.25 seconds with a dataset of 3000 animation works which is much lower than the existing work.

Emotion-modulated recall: Congruency effects of nonverbal facial and vocal cues on semantic recall

The current study investigated whether automatic integration of crossmodal stimuli (i.e. facial emotions and emotional prosody) facilitated or impaired the intake and retention of unattended verbal content. The study borrowed from previous bimodal integration designs and included a two-alternative forced-choice (2AFC) task, where subjects were instructed to identify the emotion of a face (as either ‘angry’ or ‘happy’) while ignoring a concurrently presented sentence (spoken in an angry, happy, or neutral prosody), after which a surprise recall was administered to investigate effects on semantic content retention. While bimodal integration effects were replicated (i.e. faster and more accurate emotion identification under congruent conditions), congruency effects were not found for semantic recall. Overall, semantic recall was better for trials with emotional (vs. neutral) faces, and worse in trials with happy (vs. angry or neutral) prosody. Taken together, our findings suggest that when individuals focus their attention on evaluation of facial expressions, they implicitly integrate nonverbal emotional vocal cues (i.e. hedonic valence or emotional tone of accompanying sentences), and devote less attention to their semantic content. While the impairing effect of happy prosody on recall may indicate an emotional interference effect, more research is required to uncover potential prosody-specific effects. All supplemental online materials can be found on OSF (https://osf.io/am9p2/).

Deep integrative models for large-scale human genomics

Polygenic risk scores (PRSs) are expected to play a critical role in achieving precision medicine. PRS predictors are generally based on linear models using summary statistics, and more recently individual- level data. However, these predictors generally only capture additive relationships and are limited when it comes to what type of data they use. Here, we develop a deep learning framework (EIR) for PRS prediction which includes a model, genome-local-net (GLN), we specifically designed for large scale genomics data. The framework supports multi-task (MT) learning, automatic integration of clinical and biochemical data and model explainability. GLN outperforms LASSO for a wide range of diseases, particularly autoimmune disease which have been researched for interaction effects. We showcase the flexibility of the framework by training one MT model to predict 338 diseases simultaneously. Furthermore, we find that incorporating measurement data for PRSs improves performance for virtually all (93%) diseases considered (ROC-AUC improvement up to 0.36) and that including genotype data provides better model calibration compared to measurements alone. We use the framework to analyse what our models learn and find that they learn both relevant disease variants and clinical measurements. EIR is open source and available at https://github.com/arnor-sigurdsson/EIR.

Fluoroless ablation of atrial fibrillation using intracardiac ultrasound integrated with 3D electroanatomical mapping system

Funding Acknowledgements Type of funding sources: None. Introduction Intracardiac echocardiography (ICE) is gaining increasingly wider adoption in interventional electrophysiology (EP) and represents an all-round tool for ablation of atrial fibrillation (AF). The key upgrade to the usefulness of ICE is its integration into three-dimensional (3D) electroanatomic mapping (EAM) system (ICE/EAM automatic integration system). Purpose The aim of this single-centre retrospective study was to evaluate feasibility, safety and acute efficacy of ICE/EAM automatic integration system guided fluoroless ablation of AF. Methods Patients with symptomatic paroxysmal or persistent AF referred for first pulmonary vein isolation (PVI) radiofrequency catheter ablation (RFCA) from September 2017 to August 2020 were included in the study. Those who underwent additional ablations for concomitant arrhythmias were excluded from statistical analysis. All procedures were performed without the use of fluoroscopy. A detailed 3D virtual anatomy of the left atrium (LA) and structures relevant to AF ablation was constructed from ultrasound contours obtained with ICE probe inside the LA. Pulmonary veins (PVs) and antral regions were additionally mapped with fast anatomical mapping. PVI was performed with contact force (CF) sensing catheter. Procedural endpoint was successful PVI. Results A total of 56 patients underwent RFCA (35.7% females, median age 62.7 years, 53.6% paroxysmal AF). Acute PVI was achieved in all patients (100%). Adverse events were detected in two patients (3.6%). The median procedure duration was 110.5 min (IQR 100.0-133.8). First-pass isolation was achieved in 50/56 LPVs (89.3%) and in 44/56 RPVs (78.6%). In patients where first-pass isolation was no achieved, intravenous carina had to be ablated in 3/6 (50%) of LPVs and 9/12 (75%) of RPVs. Conclusions Flouroless PVI using ICE/EAM automatic integration system is feasible, safe and acutely effective. We achieved high rate of first-pass isolation.

TranSyT, the Transport Systems Tracker

The importance and rate of development of GSM models have been growing for the last years, increasing the demand for software solutions that automatise several steps of this process. However, since TRIAGE's release, software development for automatic integration of transport reactions into models has stalled. Here we present the Transport Systems Tracker (TranSyT), the next iteration of TRIAGE. Unlike its predecessor, TranSyT does not rely on manual curation to expand its internal database, derived from highly-curated records retrieved from TCDB and complemented with information from other data sources. TranSyT compiles information regarding TC families, transport proteins, and derives reactions into its internal database, making it available for rapid annotation of complete genomes. All transport reactions have GPR associations and can be exported with identifiers from four different metabolite databases. TranSyT is currently available as a plugin for merlin v4.0 and an app for KBase.