Model Predictions Research Articles

Rapid identification of fragrant rice using starch flavor compound via NIR spectroscopy coupled with GC–MS and Badh2 genotyping

The identification of fragrant rice varieties using near-infrared reflectance spectroscopy (NIRS) models has attracted extensive attention from regulatory authorities worldwide. In this study, 138 fragrant and 54 nonfragrant rice varieties were planted in the same region and distinguished using sensory evaluation, gas chromatography–mass spectrometry analysis, and betaine aldehyde dehydrogenase 2 (Badh2) genotyping. Then, the 2-acetyl−1-pyrroline (2-AP) content was assessed based on partial least-squares discriminant (PLS-DA) models generated after 2nd individually or combined with SNV/MSC/smoothing preprocessing successfully classified fragrant rice both in the calibration and predictive sets. Moreover, design of experiments (DoE)-based preprocessing selection was employed as an effective strategy to optimize the calibration models compared with the one variable at a time (OVAT) method. Further, Badh2 genotype sample screening assisted with identifying authentic fragrant rice and guaranteed the NIRS model's prediction accuracy in identifying fragrant rice. In conclusion, the high throughput PLS-DA multivariate method coupled with NIRS data was applied to identify fragrant rice varieties in routine monitoring and was effective, accurate, and rapid.

Prediction of moisture content ratio of emulsified asphalt chip seal based on machine learning and electrical parameters

The moisture content ratio (MCR) of the emulsified asphalt chip seal can determine its curing degree. However, the MCR of emulsified asphalt chip seal is difficult to measure on actual projects, and there is a lack of a method to assess its MCR. The objective of this study is to establish a prediction methodology for the MCR of emulsified asphalt chip seal based on machine learning and electrical parameters. Features such as electrical parameters and MCR of emulsified asphalt chip seal at different times were measured experimentally. The importance of the features was evaluated using a Random Forest (RF) model. A Back Propagation Neural Network (BPNN) prediction model was established using the important features. The weights and biases of the BPNN were optimized and initialized using the Improved Particle Swarm Optimization (IMPSO) algorithm. As a result, the RF-IMPSO-BPNN emulsified asphalt chip seal MCR prediction model was developed. This model was compared with five other models. The results show that compared to the RF-PSO-BPNN model, the improved RF-IMPSO-BPNN model can improve the ability of the neural network to find the global optimal solution. Compared to the other four machine learning models, the RF-IMPSO-BPNN model can achieve higher prediction accuracy while reducing the human and material resources of the various devices to collect part of the data. In addition, the emulsified asphalt chip seal cures at low MCR. The model predictions are more accurate at low MCR. Therefore, this study developed the RF-IMPSO-BPNN emulsified asphalt chip seal MCR prediction model, which can use fewer features to achieve higher accuracy and provide a rapid and non-destructive idea for judging its curing.

Rapid estimation of left ventricular contractility with a physics-informed neural network inverse modeling approach

Physics-based computer models based on numerical solutions of the governing equations generally cannot make rapid predictions, which in turn limits their applications in the clinic. To address this issue, we developed a physics-informed neural network (PINN) model that encodes the physics of a closed-loop blood circulation system embedding a left ventricle (LV). The PINN model is trained to satisfy a system of ordinary differential equations (ODEs) associated with a lumped parameter description of the circulatory system. The model predictions have a maximum error of less than 5% when compared to those obtained by solving the ODEs numerically. An inverse modeling approach using the PINN model is also developed to rapidly estimate model parameters (in ∼ 3 min) from single-beat LV pressure and volume waveforms. Using synthetic LV pressure and volume waveforms generated by the PINN model with different model parameter values, we show that the inverse modeling approach can recover the corresponding ground truth values for LV contractility indexed by the end-systolic elastance Ees with a 1% error, which suggests that this parameter is unique. The PINN inverse modeling approach is then applied to estimate Ees using waveforms acquired from 11 swine models, including waveforms acquired before and after administration of dobutamine (an inotropic agent) in 3 animals. The estimated Ees is about 58% to 284% higher for the data associated with dobutamine compared to those without, which implies that this approach can be used to estimate LV contractility using single-beat measurements. The PINN inverse modeling can potentially be used in the clinic to simultaneously estimate LV contractility and other physiological parameters from single-beat measurements.

Hybrid real-time wave forecasting model combining Gaussian process regression and neural networks

To safely and effectively conduct offshore operations, a real-time wave forecast model that can accurately and rapidly forecast waves hours or days in advance is highly valuable. Recent studies have investigated models with neural networks (NNs) and have demonstrated their advantages in terms of forecasting speed and accuracy. However, NNs are limited by their requirement for a large training dataset, which can hinder the construction of the model if sufficient training data are unavailable. Therefore, this study constructed a hybrid real-time wave forecast model by combining an NN model and Gaussian process regression (GPR) model. The objective was to obtain reasonably accurate forecasts that were more accurate than those from individual NN models even when sufficient training data for the NN model was unavailable. To derive an optimal prediction, the values predicted by the NN and GPR models were combined by utilizing the standard deviation of the latter and considering the accuracy of each model. Consequently, predictions that were more accurate compared to those of the individual models for an entire year were obtained. This improvement was particularly significant for predicting relatively large wave heights, as indicated by the daily maximum wave heights. Additionally, various accuracy evaluation metrics demonstrated that despite using a small training dataset, the optimal prediction of the hybrid model was still superior to those of the individual models for predicting the daily maximum wave heights.

Artificial intelligence-enabled electrocardiogram for mortality and cardiovascular risk estimation: a model development and validation study

Artificial intelligence (AI)-enabled electrocardiography (ECG) can be used to predict risk of future disease and mortality but has not yet been adopted into clinical practice. Existing model predictions do not have actionability at an individual patient level, explainability, or biological plausibi. We sought to address these limitations of previous AI-ECG approaches by developing the AI-ECG risk estimator (AIRE) platform. The AIRE platform was developed in a secondary care dataset (Beth Israel Deaconess Medical Center [BIDMC]) of 1 163 401 ECGs from 189 539 patients with deep learning and a discrete-time survival model to create a patient-specific survival curve with a single ECG. Therefore, AIRE predicts not only risk of mortality, but also time-to-mortality. AIRE was validated in five diverse, transnational cohorts from the USA, Brazil, and the UK (UK Biobank [UKB]), including volunteers, primary care patients, and secondary care patients. AIRE accurately predicts risk of all-cause mortality (BIDMC C-index 0·775, 95% CI 0·773-0·776; C-indices on external validation datasets 0·638-0·773), future ventricular arrhythmia (BIDMC C-index 0·760, 95% CI 0·756-0·763; UKB C-index 0·719, 95% CI 0·635-0·803), future atherosclerotic cardiovascular disease (0·696, 0·694-0·698; 0·643, 0·624-0·662), and future heart failure (0·787, 0·785-0·789; 0·768, 0·733-0·802). Through phenome-wide and genome-wide association studies, we identified candidate biological pathways for the prediction of increased risk, including changes in cardiac structure and function, and genes associated with cardiac structure, biological ageing, and metabolic syndrome. AIRE is an actionable, explainable, and biologically plausible AI-ECG risk estimation platform that has the potential for use worldwide across a wide range of clinical contexts for short-term and long-term risk estimation. British Heart Foundation, National Institute for Health and Care Research, and Medical Research Council.

Signature analysis of high-throughput transcriptomics screening data for mechanistic inference and chemical grouping.

High-throughput transcriptomics (HTTr) uses gene expression profiling to characterize the biological activity of chemicals in in vitro cell-based test systems. As an extension of a previous study testing 44 chemicals, HTTr was used to screen an additional 1,751 unique chemicals from the EPA's ToxCast collection in MCF7 cells using 8 concentrations and an exposure duration of 6 h. We hypothesized that concentration-response modeling of signature scores could be used to identify putative molecular targets and cluster chemicals with similar bioactivity. Clustering and enrichment analyses were conducted based on signature catalog annotations and ToxPrint chemotypes to facilitate molecular target prediction and grouping of chemicals with similar bioactivity profiles. Enrichment analysis based on signature catalog annotation identified known mechanisms of action (MeOAs) associated with well-studied chemicals and generated putative MeOAs for other active chemicals. Chemicals with predicted MeOAs included those targeting estrogen receptor (ER), glucocorticoid receptor (GR), retinoic acid receptor (RAR), the NRF2/KEAP/ARE pathway, AP-1 activation, and others. Using reference chemicals for ER modulation, the study demonstrated that HTTr in MCF7 cells was able to stratify chemicals in terms of agonist potency, distinguish ER agonists from antagonists, and cluster chemicals with similar activities as predicted by the ToxCast ER Pathway model. Uniform manifold approximation and projection (UMAP) embedding of signature-level results identified novel ER modulators with no ToxCast ER Pathway model predictions. Finally, UMAP combined with ToxPrint chemotype enrichment was used to explore the biological activity of structurally related chemicals. The study demonstrates that HTTr can be used to inform chemical risk assessment by determining in vitro points of departure, predicting chemicals' MeOA and grouping chemicals with similar bioactivity profiles.

The global, regional, and national patterns of change in the burden of congenital birth defects, 1990–2021: an analysis of the global burden of disease study 2021 and forecast to 2040

Congenital birth defects (CBDs) present enormous challenges to global healthcare systems. These conditions severely impact patients' health and underscore issues related to socioeconomic development and healthcare accessibility and efficiency. Previous studies have been geographically limited and lacked comprehensive global analysis. This study provides global, regional, and national disability-adjusted life years (DALYs) data for four major congenital birth defects-congenital heart defects (CHD), neural tube defects (NTDs), digestive congenital anomalies (DCAs), and Down syndrome (DS) from 1990 to 2021, emphasizing health inequalities. The goal is to offer scientific evidence for optimizing resource allocation, focusing on high-burden populations, and reducing disease burden. This study systematically evaluated the global, regional, and national burden of CBDs and their changes from 1990 to 2021 using the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2021. To conduct a more focused analysis, four specific CBDs were selected: CHD, NTDs, DCAs, and DS. DALYs were used as the metric, combined with the sociodemographic index (SDI). Analyses included the slope index of inequality and concentration index to measure health inequalities, frontier analysis to estimate achievable outcomes based on development levels, decomposition analysis to identify drivers of disease burden changes, Joinpoint regression analysis to assess temporal trends, and the Bayesian age-period-cohort (BAPC) model to predict future disease burden trends. Compared to 1990, the global burden of the CBDs in 2021 showed a downward trend. Males had a higher burden than females, with the highest burden observed in low-SDI regions. When examining CHD, NTDs, DCAs, and DS specifically, trends in burden changes varied across different CBDs at the global, regional, and national levels. Frontier analysis revealed potential for burden improvement in various countries and territories. Decomposition analysis highlighted differences in disease burden drivers across SDI regions, showing the greatest improvement observed in low-SDI regions. Joinpoint regression analysis indicated a downward trend in DALYs burden across SDI regions, and BAPC model predictions suggested that the burden of CBDs will continue to decline in the future. CBDs pose a major challenge to global public health. Despite an overall decline in disease burden, health inequalities remain prominent, particularly in countries and territories with lower levels of development. Future public health interventions should focus on countries and territories with low levels of development by optimizing healthcare resource allocation, improving basic health infrastructure, enhancing health education, and reducing disease burden inequalities. Global collaboration and data sharing are essential to promote a lifecycle management model for CBDs research and treatment, advancing global health development. This study was supported by the National Natural Science Foundation of China (No. 82270310) and the Jiangsu Provincial Key Research and Development Program (No. BE2023662).

Practical guide to SHAP analysis: Explaining supervised machine learning model predictions in drug development.

Despite increasing interest in using Artificial Intelligence (AI) and Machine Learning (ML) models for drug development, effectively interpreting their predictions remains a challenge, which limits their impact on clinical decisions. We address this issue by providing a practical guide to SHapley Additive exPlanations (SHAP), a popular feature-based interpretability method, which can be seamlessly integrated into supervised ML models to gain a deeper understanding of their predictions, thereby enhancing their transparency and trustworthiness. This tutorial focuses on the application of SHAP analysis to standard ML black-box models for regression and classification problems. We provide an overview of various visualization plots and their interpretation, available software for implementing SHAP, and highlight best practices, as well as special considerations, when dealing with binary endpoints and time-series models. To enhance the reader's understanding for the method, we also apply it to inherently explainable regression models. Finally, we discuss the limitations and ongoing advancements aimed at tackling the current drawbacks of the method.

Anthropogenic Extremely Low Volatility Organics (ELVOCs) Govern the Growth of Molecular Clusters Over the Southern Great Plains During the Springtime

AbstractNew particle formation (NPF) often drives cloud condensation nuclei concentrations and the processes governing nucleation of molecular clusters vary substantially in different regions. The growth of these clusters from ∼2 to >10 nm diameters is often driven by the availability of extremely low volatility organic vapors (ELVOCs). Although the pathways to ELVOC formation from the oxidation of biogenic terpenes are better understood, the mechanistic pathways for ELVOC formation from oxidation of anthropogenic organics are less well understood. We integrate measurements and detailed regional model simulations to understand the processes governing NPF and secondary organic aerosol formation at the Southern Great Plain (SGP) observatory in Oklahoma and compare these with a site within the Bankhead National Forest (BNF) in Alabama, southeast USA. During our two simulated NPF event days, nucleation rates are predicted to be at least an order of magnitude higher at SGP compared to BNF largely due to lower sulfuric acid (H2SO4) concentrations at BNF. Among the different nucleation mechanisms in WRF‐Chem, we find that the dimethylamine (DMA) + H2SO4 nucleation mechanism dominates at SGP. We find that anthropogenic ELVOCs are critical for explaining the growth of particles observed at SGP. Treating organic particles as semisolid, with strong diffusion limitations for organic vapor uptake in the particle phase, brings model predictions into closer agreement with observations. We also simulate two non‐NPF event days observed at the SGP site and show that low‐level clouds reduce photochemical activity with corresponding reductions in H2SO4 and anthropogenic ELVOC concentrations, thereby explaining the lack of NPF.

Numerical Study on Droplet Splashing Behavior of Slag‐Splashing for Converter Protection Using Volume of Fluid‐Discrete Phase Model Two‐Way Conversion Model

Limited quantitative research exists on the complex phenomenon of slag splashing in converters. A comprehensive modeling framework is developed to fill this gap, integrating a 3D two‐phase flow model with a volume of fluid‐discrete phase model two‐way conversion model. This framework explores droplet behavior during splashing for converter protection, encompassing the mutual conversion between slag and discrete droplets induced by top‐blowing oxygen. It also tracks the trajectories of all droplets, enabling a detailed quantitative analysis of the splashing process. The Eulerian wall film model is used to simulate liquid film formation on the converter wall. Model predictions are validated against 1:10 scale experimental data from a 50 t converter. This assessment covers splashing rates, droplet locations, and size distribution under various operating conditions. Parametric studies identified an optimal top‐blowing flow rate of 8.80 Nm3 h−1 and an oxygen lance height of 233.2 mm, balancing splashing effectiveness with production cost and safety.

Evolution and prognostic impact of left ventricular myocardial work indices after transcatheter aortic valve replacement in patients with severe aortic stenosis

Abstract Introduction Left ventricular myocardial work (LVMW) is an echocardiographic-based technique to evaluate LV function using pressure-strain loops and therefore correcting for the left ventricular (LV) afterload. LVMW indices have been shown to be of prognostic importance in patients with aortic stenosis (AS) before transcatheter aortic valve replacement (TAVR), but no data is available on the evolution and the prognostic value of these parameters after TAVR. Purpose To measure LVMW indices before and after TAVR in patients with severe AS and explore the association with outcome. Methods As previously validated, LV global longitudinal strain (GLS) and LV systolic pressure (as the sum of the brachial arterial blood pressure and mean aortic valve gradient before TAVR and as brachial arterial blood pressure after TAVR) were incorporated to build pressure-strain loops and to calculate the following LVMW indices: global work index (LV GWI), global constructive work (LV GCW), global wasted work (LV GWW) and global work efficiency (LV GWE). The endpoint was all-cause mortality. Results A total of 255 patients were included with a median age of 82 years (IQR 77-85); 51% were men and 67% presented with NYHA class III/IV. After TAVR, LVEF and LV GLS remained in average unchanged (from 56 to 55%, p=0.470 and from 13.6 to 13.2%, p=0.068). While LV GWW remained unchanged after TAVR (from 247 to 258mmHg%, p=0.080), LV GWI, LV GCW and LV GWE significantly decreased (from 1882 to 1291mmHg%, p <0.001, from 2248 to 1671mmHg%, p <0.001 and from 89 to 85%, p <0.001 respectively). During a median follow-up of 59 months (IQR 40-72), 129 patients died. Univariable Cox regression analysis identified an association between all-cause mortality and male sex, diabetes mellitus, chronic obstructive pulmonary disease (COPD), renal function, atrial fibrillation, Charlson comorbidity index, pacemaker implantation, and post-TAVR LV GLS, LV GWI and LV GCW. These clinical parameters were used as a basal multivariable Cox regression model of which on top either post-TAVR LV GLS, LV GWI or LV GCW were added. All these parameters of post-TAVR LV systolic function showed an independent association with all-cause mortality, but LV GWI demonstrated the highest increase in model predictivity (Figure 1). Given the prognostic relevance of post-TAVR LV GWI, the population was divided into LV GWI tertiles. The group with the lowest LV GWI tertile showed a significantly higher proportion of male patients, higher prevalence of atrial fibrillation and also worse pre-TAVR LVEF, LV GLS, LV GWI, LV GCW and LV GWE. Kaplan Meier survival analysis confirmed the worse survival in the group with the lowest LV GWI compared to the highest tertile (log-rank ꭓ2: 8,224, p 0.016, Figure 2). Conclusion LVMW indices change after TAVR and post-TAVR LV GLS, LV GWI, LV GCW are independently associated with all-cause mortality. Among these parameters, LV GWI yielded the highest prognostic value.

Population displacement after earthquakes: benchmarking predictions based on housing damage

In the aftermath of an earthquake, the number of residents whose housing was destroyed is often used to approximate the number of people displaced (i.e., rendered homeless) after the event. While this metric can provide rapid situational awareness regarding potential long-term housing needs, more recent research highlights the importance of additional factors beyond housing damage within the scope of household displacement and return (e.g., utility disruption, tenure, place attachment). This study benchmarks population displacement estimates using this simplified conventional approach (i.e., only considering housing destruction) through three scenario models for recent earthquakes in Haiti, Japan, and Nepal. These model predictions are compared with officially reported values and alternate mobile location data-based estimates from the literature. The results highlight the promise of scenario models to realistically estimate population displacement and potential long-term housing needs after earthquakes, but also highlight a large range of uncertainty in the predicted values. Furthermore, purely basing displacement estimates on housing damage offers no view on how the displaced population counts vary with time as compared to more comprehensive models that include other factors influencing population return or alternative approaches, such as using mobile location data.

DELINEATE-Regurgitation: deep learning for automated assessment of aortic, mitral, and tricuspid regurgitation from echocardiography

Abstract Background Valvular heart diseases, including aortic regurgitation (AR), mitral regurgitation (MR), and tricuspid regurgitation (TR) are significant contributors to cardiovascular morbidity and mortality worldwide. Artificial intelligence (AI) may improve echocardiographic assessment, but no models to date have successfully evaluated valvular regurgitation. Purpose The goal of this study was to develop an AI system to improve the accuracy of echocardiographic AR, MR, and TR interpretation. Methods Complete transthoracic echocardiogram (TTE) studies performed at a single center were split into train (n=65,301), validation (14,018), and test sets (13,746). The Deep Learning for Echo Analysis, Tracking, and Evaluation of Valvular Regurgitation (DELINEATE-Regurgitation) system intakes entire TTE studies, identifies color Doppler clips showing each valve, and yields a study level classification of AR, MR, and TR on a 6-grade scale (none/trace, mild, mild-moderate, moderate, moderate-severe, severe) using the cardiologist interpretation as a reference standard. This system used a hybrid neural network, leveraging the spatiotemporal feature learning capabilities of convolutional neural networks with the sequential processing strengths of transformer networks. The end-to-end hybrid approach allows the model to learn nuanced representations of the data including features within and across color Doppler videos. Model accuracy was assessed using quadratically weighted Cohen’s kappa (k), area under the receiver operator characteristic curve (AUROC) for the detection of both "moderate or greater" and "severe" regurgitation, and AI-cardiologist agreement the 6-grade scale. Additionally, an analysis was conducted using a 4-grade regurgitation scale (none/trace, mild, moderate, and severe) by rolling up intermediate classes (mild-moderate and moderate-severe) into the next higher class for both cardiologist determinations and model predictions. Results The DELINEATE-Regurgitation system demonstrated a high accuracy in the classification of AR (k=0.857), MR (k=0.867), and TR (k=0.847) (Figure 1). DELINEATE predicted the same severity as cardiologists (AR 91.8%, MR 79.9%, TR 77.7%) or within ±1 grade accuracy (AR 99.5%, MR 98.6%, TR 98.6%). The models had excellent binary discrimination in the detection of moderate or greater (AR AUROC 99.4, MR 98.6, TR 98.1) and severe (AR 99.7, MR 99.3, TR 99.6) regurgitation (Figure 2). In the 4-grade scale, the model had exact agreement with cardiologists in 92.3% of AR, 82.4% of MR, and 80.6% of TR cases. Panel adjudication of AI-cardiologist disagreement and correlation with quantitative metrics at higher levels of regurgitation will be presented. Conclusions The DELINEATE-Regurgitation AI system demonstrates a high level of accuracy in the assessment of AR, MR, and TR using color Doppler TTE videos. Future work will focus on automating quantification and developing strategies for optimal clinical deployment.Figure 1 - Confusion MatricesFigure 2 - ROC Curves

A BIONIC-INSPIRED DRAGONFLY ALGORITHM FOR PARAMETRIC OPTIMIZATION AND DAMAGE INVESTIGATION DURING MACHINING OF MODIFIED POLYMER NANOCOMPOSITE

Polymer nanocomposite is commonly used to develop structural components of space, aircraft, biomedical, sensor, automobile, and battery sector applications. It remarkably substitutes the heavyweight metallic and nonmetallic engineering materials. The machining principles of polymer nanocomposites are intensely different and complex from traditional metals and alloys. The nonhomogeneity, abrasive, and anisotropic nature differs its machining aspect from conventional metallic materials. This investigation aims to execute the CNC drilling of modified nanocomposite using Graphene–carbon (G-C) @ epoxy matrix. The process constraints, namely, cutting speed (S), feed (F), and wt.% of graphene oxide (GO) vary up to three levels and are designed according to the response surface methodology (RSM) array. The nonlinear model is created to predict surface roughness (Ra) and delamination (Fd) on regression analysis. It has been found that the average error for Ra is 0.94% and for Fd it is 3.27%, which is acceptable in model predictions. The metaheuristics-based evolutionary Dragonfly algorithm (DA) evaluated the optimal parametric condition. The optimal setting prediction for the DA is observed as cutting speed (S)-37.68[Formula: see text]m/min, feed (F)-80[Formula: see text]mm/min, and wt.% of graphene oxide (GO)-1%. This algorithm demonstrates a higher application potential than the previous efforts in controlling Ra and Fd values. Both the drilling response values are found to be minimized when the cutting speed increases and the feed decreases. The best fitness value for the DA is 1.626 for surface roughness and 5.086 for delamination. This study agreed with the prediction model’s outcomes and the process parameters’ optimal condition. The defects generated during the sample drilling, such as fiber pull out, uncut/burr, and fiber breakage, were examined using FE-SEM analysis. The optimal findings of the DA module significantly controlled the damages during machining.

Protein language models learn evolutionary statistics of interacting sequence motifs

Protein language models (pLMs) have emerged as potent tools for predicting and designing protein structure and function, and the degree to which these models fundamentally understand the inherent biophysics of protein structure stands as an open question. Motivated by a finding that pLM-based structure predictors erroneously predict nonphysical structures for protein isoforms, we investigated the nature of sequence context needed for contact predictions in the pLM Evolutionary Scale Modeling (ESM-2). We demonstrate by use of a "categorical Jacobian" calculation that ESM-2 stores statistics of coevolving residues, analogously to simpler modeling approaches like Markov Random Fields and Multivariate Gaussian models. We further investigated how ESM-2 "stores" information needed to predict contacts by comparing sequence masking strategies, and found that providing local windows of sequence information allowed ESM-2 to best recover predicted contacts. This suggests that pLMs predict contacts by storing motifs of pairwise contacts. Our investigation highlights the limitations of current pLMs and underscores the importance of understanding the underlying mechanisms of these models.

Cluster radioactivity of heavy elements within an improved Coulomb and Proximity potential model

The Coulomb and Proximity potential Model (CPPM) is improved by introducing an accurate empirical expression for the nuclear surface diffuseness parameter in the expression for the proximity potential. Within the improved CPPM, the decay half-lives of experimentally detected cluster emissions from heavy elements are calculated. The results of the calculations using the Universal Decay Law (UDL) [Qi et al., Phys. Rev. Lett. 103 (2009) 072501], the improved empirical formula of Sahu et al. (ImSahuB) [Wang et al., Chin. Phys. C 45 (2021) 044111], and the empirical formula of Horoi [J. Phys. G. Nucl. Part. Phys. 30 (2004) 945] are compared with the predictions of the improved CPPM. The lowest standard deviation is obtained for the predictions using the improved CPPM. The model’s predictive power increases significantly when the proximity potential is modified by including the diffuseness parameter. Furthermore, improved CPPM is employed to calculate the decay half-lives of undiscovered cluster emissions from heavy elements, a valuable information for upcoming experimental investigations. The calculated values are compared with the predictions of the UDL, Horoi formula and the improved empirical formula of Sahu et al. (ImSahuB). There is a high degree of agreement between the improved CPPM, UDL, and Horoi formula predictions. The ImSahuB formula predict decay half-life one or two order magnitudes lower than the predictions of the other models.

Spectroscopic analysis of the strongly lensed SN Encore: Constraints on cosmic evolution of Type Ia supernovae

Abstract Strong gravitational lensing magnifies the light from a background source, allowing us to study these sources in detail. Here, we study the spectra of a z = 1.95 lensed Type Ia supernova SN Encore for its brightest Image A, taken 39 days apart. We infer the spectral age with template matching using the supernova identification (SNID) software and find the spectra to be at 29.0 ±5.0 and 37.4 ±2.8 rest-frame days post maximum respectively, consistent with separation in the observer frame after accounting for time-dilation. Since SNe Ia measure dark energy properties by providing relative distances between low- and high-z SNe, it is important to test for evolution of spectroscopic properties. Comparing the spectra to composite low-z SN Ia spectra, we find strong evidence for similarity between the local sample and SN Encore. The line velocities of common SN Ia spectral lines, Si II 6355 and Ca II NIR triplet are consistent with the distribution for the low-z sample as well as other lensed SNe Ia, e.g. iPTF16geu (z = 0.409)and SN H0pe (z = 1.78). The consistency between the low-z sample and lensed SNe at high-z suggests no obvious cosmic evolution demonstrating their using as high-z distance indicators, though this needs to be confirmed/refuted via a larger sample. We also find that the spectra of SN Encore match the predictions for explosion models very well. With future large samples of lensed SNe Ia e.g. with the Vera C. Rubin Observatory, spectra at such late phases will be important to distinguish between different explosion scenarios.

Prediction of overweight/obesity risk in adolescence using static and dynamic multi-exposure models

Abstract Background Studying the risk factors for obesity development and timely identification of individuals at increased risk are key for obesity prevention. We aimed to assess the associations of multiple prenatal, infancy and childhood exposures with overweight/obesity at 13 years (y) and develop static and dynamic multiple-exposure prediction models. Methods In 4232 mother-child pairs from the Generation XXI birth cohort, we assessed, during pregnancy/infancy and childhood, 55 sociodemographic factors, anthropometric measures, lifestyle behaviors, and clinical exposures related to the mother and the child and categorized body mass index z-scores as normal weight and overweight/obesity at 13y. Static models for pregnancy/infancy, 4, 7 and 10y, and a dynamic model were fitted using Elastic Net and logistic regression and were compared in terms of predictive performance. Results Mothers with lower educational level and household income, no partner, overweight/obesity before and after pregnancy, excessive gestational weight gain, pregnancy complications and who were smokers during pregnancy had children with a higher risk of overweight/obesity at 13y. Children with higher BMI during infancy, overweight/obesity during childhood, an early introduction of solid foods (before 4 months), a higher consumption of soft drinks at 4y and females had a higher risk of overweight/obesity at 13y. We observed the lowest predictive performance for the pregnancy/infancy model (prediction error of 36%) and the highest predictive performance for the 10y and dynamic models (prediction errors of 17%). Conclusions We identified the most important predictors of overweight/obesity in adolescence, most of them modifiable, and able to be targeted in prevention strategies. Predicting overweight/obesity risk in adolescence might be suboptimal if only considering pregnancy/infancy data and improves as closer to the age of the outcome, with no additional benefit in using a complex dynamic model. Key messages • Prevention of overweight/obesity should start during prenatal life. • Children at higher risk of overweight/obesity may be identified years earlier using prediction models in clinical practice.

An inherently interpretable deep learning model for local explanations using visual concepts.

Over the past decade, deep learning has become the leading approach for various computer vision tasks and decision support systems. However, the opaque nature of deep learning models raises significant concerns about their fairness, reliability, and the underlying inferences they make. Many existing methods attempt to approximate the relationship between low-level input features and outcomes. However, humans tend to understand and reason based on high-level concepts rather than low-level input features. To bridge this gap, several concept-based interpretable methods have been developed. Most of these methods compute the importance of each discovered concept for a specific class. However, they often fail to provide local explanations. Additionally, these approaches typically rely on labeled concepts or learn directly from datasets, leading to the extraction of irrelevant concepts. They also tend to overlook the potential of these concepts to interpret model predictions effectively. This research proposes a two-stream model called the Cross-Attentional Fast/Slow Thinking Network (CA-SoftNet) to address these issues. The model is inspired by dual-process theory and integrates two key components: a shallow convolutional neural network (sCNN) as System-I for rapid, implicit pattern recognition and a cross-attentional concept memory network as System-II for transparent, controllable, and logical reasoning. Our evaluation across diverse datasets demonstrates the model's competitive accuracy, achieving 85.6%, 83.7%, 93.6%, and 90.3% on CUB 200-2011, Stanford Cars, ISIC 2016, and ISIC 2017, respectively. This performance outperforms existing interpretable models and is comparable to non-interpretable counterparts. Furthermore, our novel concept extraction method facilitates identifying and selecting salient concepts. These concepts are then used to generate concept-based local explanations that align with human thinking. Additionally, the model's ability to share similar concepts across distinct classes, such as in fine-grained classification, enhances its scalability for large datasets. This feature also induces human-like cognition and reasoning within the proposed framework.

Biventricular modelling for investigating ventricular arrhythmias in silico

Abstract Background Ventricular arrhythmias have a significant incidence rate worldwide and may potentially lead to sudden cardiac death in severe cases. Despite therapeutic advancements, the intricate mechanisms dictating these arrhythmias remain elusive. Biventricular modelling and deep learning provide a way to investigate the complex electrophysiological substrates underlying cardiac arrhythmogenesis. Purpose This study aims to develop and validate a patient-specific biventricular computational modelling pipeline for simulating ventricular arrhythmias in silico. We provide a comprehensive platform for understanding the initiation, propagation, and treatment response of various ventricular arrhythmias. Methods To facilitate large-scale computational experiments and personalised model predictions within clinical timeframes, it is imperative to establish a rapid and reproducible modelling framework. First, we constructed biventricular volumetric models derived from late gadolinium-enhanced magnetic resonance imaging (LGE-MRI) data from the Cardiac Atlas Project (2018 Atria Segmentation Challenge), and then we calculate biventricular coordinates to facilitate data registration between measures, and fibre and fibrosis inclusion. We also apply the pipeline to cardiac magnetic resonance (CMR) data and electrocardiogram recordings from UK Biobank. Results We have successfully developed a pipeline for constructing biventricular volumetric models and efficiently processed biventricular coordinates (Figure). By systematically exploring interactions between ventricular regions, the model may identify critical factors contributing to arrhythmia susceptibility and guide personalised therapeutic strategies. Furthermore, the model may provide insights into the efficacy of antiarrhythmic interventions, ablation techniques, and device therapies, facilitating the development of precision medicine approaches. Conclusions We demonstrate a pipeline for constructing patient-specific biventricular computational models at scale from imaging data and apply this to LGE-MRI data and UK Biobank CMR data. These biventricular models enable investigation of the complex mechanisms of ventricular arrhythmias, providing novel insights for clinical decision-making and treatment strategies.