Individual Models Research Articles

Precommercial Thinning but Not Commercial Thinning Increases the Merchantable and Large Sawlog Volume Production of Lodgepole Pine on Fair Sites

Abstract The effect of thinning on stand structure and volume production is well studied. However, the effect of combinations of precommercial (PCT) and commercial thinning (CT) on the production of different log grades and differential growth responses is less understood, although it is crucial for forest managers to choose an optimal management regime. Starting with actual operational field data from a lodgepole pine (Pinus contorta var.latifolia) PCT trial on fair quality sites (site index: 16–18 m at 50 years), we used an individual tree growth model to forecast development over the entire rotation. We found thinning changed the structure (density, quadratic mean diameter) of the stands but not the total volume production. However, in terms of merchantable volume and large sawlog volume, stands with only PCT provided the largest volume, more than either CT or a combination of PCT and CT, or the unthinned control. Moreover, our study findings also showed that, by doing only PCT, stands reached maximum merchantable or large sawlog mean annual increment earlier than other thinned or unthinned stands. This indicates that PCT might shorten the rotation length and contribute to an enhanced supply of timber. Study Implications: The effect of commercial thinning (CT) on stand structure and volume production is well studied. However, the effect of combinations of precommercial (PCT) and commercial thinning (CT) on the production of different log grades and differential growth responses is less studied, although it is crucial for forest managers to choose a management option. Starting with actual operational field data from a lodgepole pine (Pinus contorta var.latifolia) PCT trial in a fair-quality site (site index [SI]: 16–18), we used an individual tree growth model to develop the stands over the entire rotation. We found thinning changed the structure (density, quadratic mean diameter) of the stands but not the total volume production. However, in terms of merchantable volume and large sawlog volume, stands with only PCT provided the largest volume, more than either CT or a combination of PCT and CT, or the unthinned control. Moreover, our study findings also showed that, by doing only PCT, stands reached maximum merchantable or large sawlog MAI earlier than other thinned or unthinned stands. This indicates that PCT might shorten the rotation length and contribute to the continuous supply of timber production.

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.

Uncovering Disease Mechanisms through AI and Visual Computing: A Multi-Scale Approach to Biological Data Interpretation

This research focuses on uncovering complex disease mechanisms by employing artificial intelligence (AI) and visual computing to interpret multi-scale biological data. By analyzing data from molecular to organ-level systems, this study aims to bridge the gap between cellular insights and clinical applications, offering new perspectives on diseases such as cancer, neurodegenerative disorders, and cardiovascular diseases. Through a comprehensive approach to data integration, the project compiles omics and imaging data to enable holistic analysis across biological scales. The use of hierarchical neural networks, graph neural networks (GNNs), and convolutional neural networks (CNNs) allows for multi-dimensional modeling and the extraction of disease-related patterns. The findings are validated with clinical data to support actionable insights, laying the groundwork for personalized disease models and precision therapeutic strategies. Expected outcomes include breakthroughs in disease understanding, advancements in precision medicine, and the establishment of a robust AI framework that will contribute to future biomedical applications. This multidisciplinary approach promises to transform diagnostics, prognosis, and treatment, facilitating a deeper understanding of disease mechanisms and their translation into clinical practice. The integration of multi-scale data enables a nuanced view of disease, capturing interactions that may be obscured in single-level analyses. By employing advanced visual computing techniques, the project highlights structural abnormalities within tissue samples, linking cellular changes to systemic disease pathways. This approach not only aids in identifying novel biomarkers but also supports the development of individualized treatment models that predict patient-specific responses. With validation from clinical datasets, the AI framework is poised to enhance clinical decision-making, offering insights tailored to the unique biological profiles of patients. The project’s multi-scale methodology serves as a foundation for future biomedical research, setting a new standard for AI-driven disease analysis.

Investigating pedestrians’ red light running intentions at urban intersections in different traffic Environments: A scenario-based analysis guided by theoretical frameworks

Pedestrian risky behaviors are one of the contributing factors to crashes involving pedestrians. Therefore, it is crucial to comprehend the mechanisms by which pedestrians interact with many influential components in the traffic environment. This study aimed to evaluate pedestrians’ red light running intentions and related factors under different traffic flow scenarios, including straight traffic flow, right-turning traffic flow, and left-turning traffic flow. A theoretical approach based on the theory of planned behavior (TPB) and the prototype willingness model (PWM) was employed. Data were collected from an online survey of 2250 participants in Tehran, Iran. Structural equation modeling (SEM) was used to identify the significant factors that explain intentions. All models successfully explained the behavioral intention for red light running violation; however, the findings revealed that the integrated model was the best-performing model to represent violation and, thus, was selected for interpreting the results and drawing relevant conclusions. Different traffic flow scenarios had varied effects on violation intentions for individual characteristics and model constructs. Previous crash experiences and driving-related background variables emerged to impact pedestrian violation intention across three scenarios. The findings also suggested that the rational constructs (attitude, perceived behavioral control, and facilitating conditions) had a more robust impact on violation intention compared to reactive constructs (prototype similarity, prototype favorability), with facilitating conditions being the strongest predictor of the model, followed by attitudes toward violation as a significant predictor of intention for red light violation. According to the results, the mechanism of risk-taking varies depending on the direction of the traffic flow. Higher risk was associated with the violation at the intersections with straight traffic flow compared to the intersections with turning traffic flow. Based on the findings of this study, several implications, including interventions focusing on individuals’ transportation safety attitudes, countermeasures to increase the risk perception of pedestrians toward turning vehicles, and countermeasures regarding the use of mobile phones while walking for the context of this study were proposed.

Enhancing early detection of cognitive decline in the elderly: a comparative study utilizing large language models in clinical notes

BackgroundLarge language models (LLMs) have shown promising performance in various healthcare domains, but their effectiveness in identifying specific clinical conditions in real medical records is less explored. This study evaluates LLMs for detecting signs of cognitive decline in real electronic health record (EHR) clinical notes, comparing their error profiles with traditional models. The insights gained will inform strategies for performance enhancement. MethodsThis study, conducted at Mass General Brigham in Boston, MA, analysed clinical notes from the four years prior to a 2019 diagnosis of mild cognitive impairment in patients aged 50 and older. We developed prompts for two LLMs, Llama 2 and GPT-4, on Health Insurance Portability and Accountability Act (HIPAA)-compliant cloud-computing platforms using multiple approaches (e.g., hard prompting, retrieval augmented generation, and error analysis-based instructions) to select the optimal LLM-based method. Baseline models included a hierarchical attention-based neural network and XGBoost. Subsequently, we constructed an ensemble of the three models using a majority vote approach. Confusion-matrix-based scores were used for model evaluation. FindingsWe used a randomly annotated sample of 4,949 note sections from 1,969 patients (women: 1,046 [53.1%]; age: mean, 76.0 [SD, 13.3] years), filtered with keywords related to cognitive functions, for model development. For testing, a random annotated sample of 1,996 note sections from 1,161 patients (women: 619 [53.3%]; age: mean, 76.5 [SD, 10.2] years) without keyword filtering was utilised. GPT-4 demonstrated superior accuracy and efficiency compared to Llama 2, but did not outperform traditional models. The ensemble model outperformed the individual models in terms of all evaluation metrics with statistical significance (p<0.01), achieving a precision of 90.2% [95% CI: 81.9%-96.8%], a recall of 94.2% [95% CI: 87.9%-98.7%], and an F1-score of 92.1% [95% CI: 86.8%-96.4%]. Notably, the ensemble model showed a significant improvement in precision, increasing from a range of 70%-79% to above 90%, compared to the best-performing single model. Error analysis revealed that 63 samples were incorrectly predicted by at least one model; however, only 2 cases (3.2%) were mutual errors across all models, indicating diverse error profiles among them. InterpretationLLMs and traditional machine learning models trained using local EHR data exhibited diverse error profiles. The ensemble of these models was found to be complementary, enhancing diagnostic performance. Future research should investigate integrating LLMs with smaller, localised models and incorporating medical data and domain knowledge to enhance performance on specific tasks.

Hyperspectral Inversion of Soil Cu Content in Agricultural Land Based on Continuous Wavelet Transform and Stacking Ensemble Learning

Heavy metal pollution in agricultural land poses significant threats to both the ecological environment and human health. Therefore, the rapid and accurate prediction of heavy metal content in agricultural soil is crucial for environmental protection and soil remediation. Acknowledging the limitations of traditional single linear or nonlinear machine learning models in terms of prediction accuracy, this study developed an ensemble learning model that integrates multiple linear or nonlinear learning models with a random forest (RF) model to improve both the prediction accuracy and reliability. In this study, we selected a typical copper (Cu) polluted area in the Pearl River Delta of Guangdong Province as the research site and collected Cu content data and indoor soil reflectance spectral data from 269 surface soil samples. First, the soil spectral data were preprocessed using Savitzky–Golay (SG) smoothing, multiplicative scattering correction (MSC), and continuous wavelet transform (CWT) to reduce noise interference. Next, principal components analysis (PCA) was employed to reduce the dimensionality of the preprocessed spectral data, eliminating redundant features and lowering the computational complexity. Finally, based on the dimensionality-reduced data and Cu content, we established a stacked ensemble learning model, where the base models included SVR, PLSR, BPNN, and XGBoost, with RF serving as the meta-model to estimate the soil heavy metal content. To evaluate the performance of the stacking model, we compared its prediction accuracy with that of individual models. The results indicate that, compared to the traditional machine learning models, the prediction accuracy of the stacking model was superior (R2 = 0.77; RMSE = 7.65 mg/kg; RPD = 2.29). This suggests that the integrated algorithm demonstrates a greater robustness and generalization capability. This study presents a method to improve soil heavy metal content estimation using hyperspectral technology, ensuring a robust model that supports policymakers in making informed decisions about land use, agriculture, and environmental protection.

Enhancing the differential diagnosis of small pulmonary nodules: a comprehensive model integrating plasma methylation, protein biomarkers, and LDCT imaging features

BackgroundAccurate differentiation between malignant and benign pulmonary nodules, especially those measuring 5–10 mm in diameter, continues to pose a significant diagnostic challenge. This study introduces a novel, precise approach by integrating circulating cell-free DNA (cfDNA) methylation patterns, protein profiling, and computed tomography (CT) imaging features to enhance the classification of pulmonary nodules.MethodsBlood samples were collected from 419 participants diagnosed with pulmonary nodules ranging from 5 to 30 mm in size, before any disease-altering procedures such as treatment or surgical intervention. High-throughput bisulfite sequencing was used to conduct DNA methylation profiling, while protein profiling was performed utilizing the Olink proximity extension assay. The dataset was divided into a training set and an independent test set. The training set included 162 matched cases of benign and malignant nodules, balanced for sex and age. In contrast, the test set consisted of 46 benign and 49 malignant nodules. By effectively integrating both molecular (DNA methylation and protein profiling) and CT imaging parameters, a sophisticated deep learning-based classifier was developed to accurately distinguish between benign and malignant pulmonary nodules.ResultsOur results demonstrate that the integrated model is both accurate and robust in distinguishing between benign and malignant pulmonary nodules. It achieved an AUC score 0.925 (sensitivity = 83.7%, specificity = 82.6%) in classifying test set. The performance of the integrated model was significantly higher than that of individual methylation (AUC = 0.799, P = 0.004), protein (AUC = 0.846, P = 0.009), and imaging models (AUC = 0.866, P = 0.01). Importantly, the integrated model achieved a higher AUC of 0.951 (sensitivity = 83.9%, specificity = 89.7%) in 5–10 mm small nodules. These results collectively confirm the accuracy and robustness of our model in detecting malignant nodules from benign ones.ConclusionsOur study presents a promising noninvasive approach to distinguish the malignancy of pulmonary nodules using multiple molecular and imaging features, which has the potential to assist in clinical decision-making.Trial registration: This study was registered on ClinicalTrials.gov on 01/01/2020 (NCT05432128). https://classic.clinicaltrials.gov/ct2/show/NCT05432128.

Applying integrated enhanced cognitive behaviour therapy (I-CBTE) to severe and longstanding eating disorders (SEED) Paper 2: An in-depth case study for clinicians

BackgroundThis case study examines the application of Integrated Enhanced Cognitive Behavioural Therapy (I-CBTE) for a patient with severe, longstanding anorexia nervosa and multiple comorbidities, including organic hallucinosis, complex post-traumatic stress disorder (CPTSD), and severe self-harm. Such complex presentations often result in patients falling between services, which can lead to high chronicity and increased mortality risk. Commentaries from two additional patients who have recovered from severe and longstanding anorexia nervosa are included.Case studyThe patient developed severe anorexia nervosa and hallucinosis after a traumatic brain injury in 2000. Despite numerous hospitalisations and various psychotropic medications in the UK and France, standard treatments were ineffective for 17 years. However, Integrated Enhanced Cognitive Behaviour Therapy (I-CBTE) using a whole-team approach and intensive, personalised psychological treatment alongside nutritional rehabilitation proved effective.MethodsIn this paper, we describe the application of the I-CBTE model for individuals with severe, longstanding, and complex anorexia nervosa, using lived experience perspectives from three patients to inform clinicians. We also outline the methodology for adapting the model to different presentations of the disorder.OutcomesThe patient achieved and maintained full remission from her eating disorder over the last 6 years, highlighting the benefit of the I-CBTE approach in patients with complex, longstanding eating disorder histories. Successful treatment also saved in excess of £360 k just by preventing further hospitalisations and not accounting for the improvement in her quality of life. This suggests that this method can improve outcomes and reduce healthcare costs.ConclusionThis case study, with commentaries from two patients with histories of severe and longstanding anorexia nervosa, provides a detailed description of the practical application of I-CBTE for patients with severe and longstanding eating disorders with complex comorbidities, and extensive treatment histories. This offers hope for patients and a framework for clinicians to enhance existing treatment frameworks, potentially transforming the trajectory of those traditionally deemed treatment resistant.RecommendationsWe advocate the broader integration of CBT for EDs into specialist services across the care pathway to help improve outcomes for patients with complex eating disorders. Systematic training and supervision for multidisciplinary teams in this specialised therapeutic approach is recommended. Future research should investigate the long-term effectiveness of I-CBTE through longitudinal studies. Patient feedback on experiences of integrated models of care such as I-CBTE is also needed. In addition, systematic health economics studies should be conducted.

Longitudinal relationships between free-living activities, fatigue, and symptom severity in myasthenia gravis using cohort and individualized models.

Fluctuating symptoms and fatigue are common issues in myasthenia gravis (MG), but it is unclear if these symptoms are related to physical activity or sleep patterns. This study sought to determine the day-to-day relationship between patient-reported symptoms and physical activity and sleep over 12 weeks. Sixteen participants with generalized MG wore a wrist-mounted accelerometer continuously for the study duration and reported their symptoms and fatigue each evening. Cumulative link mixed models were used to analyze whether clinical and demographic characteristics, physical activity, and sleep were related to symptom severity and fatigue over the study period. Three types of models were constructed: a cohort model, a model in which data was scaled to each participant, and individual models. The cohort model indicated that higher disease severity, female sex, more comorbidities, less physical activity, more inactive time, and lower quantity of sleep were significantly associated with increased symptom severity and fatigue (p < .05). However, in the within-participant scaled model, there were almost no significant associations with physical activity or sleep. In the individual models, some participants showed similar results to the cohort model, but others showed no associations or the opposite response in some variables. While physical activity and sleep were associated with self-reported symptoms and fatigue within this population, this was not necessarily applicable to individuals. This demonstrates the importance of an individualized analysis for determining how physical activity and sleep may impact outcomes in MG, with implications for clinical and self-management.

Forecasting of Solar Irradiation Based on the Deep Learning Model Using Complete Ensemble Empirical Mode Decomposition Technique

Since solar energy plants have been developing so quickly in recent years, accurate solar power forecasting plays a significant role in contemporary intelligent grid systems and setup a bilateral contract negotiation between suppliers and customers. This paper introduces a novel three-step method for predicting two-channel solar irradiance that combines cutting-edge techniques like Bidirectional Long Short-Term Memory, Wasserstein Generative adversarial Networks and Complete Ensemble Empirical Mode Decomposition with adaptive noise. This methodology that is being suggested involves a framework of decomposition integration which allows for the effective isolation of the solar output signal into sub-sequences that are simple and can be differentiated based on their unique frequency disparities. This is achieved after the subsequences have been separated, and then they are subjected to individual prediction models. The high subsequences are predicted using the WGAN model, while the low subsequences are forecasted using the Bi-LSTM model. This approach ensures the specific characteristics of each subsequence are captured in the prediction process. The experimental result shows that the proposed model surpasses the performance of the suboptimal model in terms of evaluation metrics: RMSE, MAPE and R2. Compared to the suboptimal model, the RMSE, MAPE of all seasons

Prompt Recovery for Large Language Models

Understanding and recovering prompts in large language models (LLMs) is vital for addressing concerns related to privacy, copyright, and beyond. However, there is a lack of extensive research in this area. To fill this gap, we implemented model stacking techniques, such as utilizing mean prompts and embedding models, tailored for specific datasets. While these individual models were designed for particular datasets, our combined stacking model demonstrated improved accuracy in prompt recovery across diverse datasets. Although there was a slight decline in performance on the initial dataset, our comprehensive evaluation across multiple LLMs and prompt benchmarks indicates that our stacking model exceeds the performance of existing methods. Notably, this approach uses a single LLM without depending on external resources, making it an efficient and accessible solution for prompt recovery.

Identifying the spatial pattern and driving factors of nitrate in groundwater using a novel framework of interpretable stacking ensemble learning

Groundwater nitrate contamination poses a potential threat to human health and environmental safety globally. This study proposes an interpretable stacking ensemble learning (SEL) framework for enhancing and interpreting groundwater nitrate spatial predictions by integrating the two-level heterogeneous SEL model and SHapley Additive exPlanations (SHAP). In the SEL model, five commonly used machine learning models were utilized as base models (gradient boosting decision tree, extreme gradient boosting, random forest, extremely randomized trees, and k-nearest neighbor), whose outputs were taken as input data for the meta-model. When applied to the agricultural intensive area, the Eden Valley in the UK, the SEL model outperformed the individual models in predictive performance and generalization ability. It reveals a mean groundwater nitrate level of 2.22 mg/L-N, with 2.46% of sandstone aquifers exceeding the drinking standard of 11.3 mg/L-N. Alarmingly, 8.74% of areas with high groundwater nitrate remain outside the designated nitrate vulnerable zones. Moreover, SHAP identified that transmissivity, baseflow index, hydraulic conductivity, the percentage of arable land, and the C:N ratio in the soil were the top five key driving factors of groundwater nitrate. With nitrate threatening groundwater globally, this study presents a high-accuracy, interpretable, and flexible modeling framework that enhances our understanding of the mechanisms behind groundwater nitrate contamination. It implies that the interpretable SEL framework has great promise for providing valuable evidence for environmental management, water resource protection, and sustainable development, particularly in the data-scarce area.

Deep Learning Ensemble for Flood Probability Analysis

Predicting flood events is complex due to uncertainties from limited gauge data, high data and computational demands of traditional physical models, and challenges in spatial and temporal scaling. This research innovatively uses only three remotely sensed and computed factors: rainfall, runoff and temperature. We also employ three deep learning models—Feedforward Neural Network (FNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM)—along with a deep neural network ensemble (DNNE) using synthetic data to predict future flood probabilities, utilizing the Savitzky–Golay filter for smoothing. Using a hydrometeorological dataset from 1993–2022 for the Nile River basin, six flood predictors were derived. The FNN and LSTM models exhibited high accuracy and stable loss, indicating minimal overfitting, while the CNN showed slight overfitting. Performance metrics revealed that FNN achieved 99.63% accuracy and 0.999886 ROC AUC, CNN had 95.42% accuracy and 0.893218 ROC AUC, and LSTM excelled with 99.82% accuracy and 0.999967 ROC AUC. The DNNE outperformed individual models in reliability and consistency. Runoff and rainfall were the most influential predictors, while temperature had minimal impact.

Modelling the Relationship Between Merchandise Trade Flows and Some Macroeconomic Variables in Ghana

Macroeconomic variables serve as economic indicators that offer valuable insights into the overall health and stability of an economy. Changes in these variables can have significant impacts on a country&amp;apos;s trade balance and overall economic performance. This study employed multivariate time series analysis to study the relationship between Merchandise Trade Flows (MTF), Monetary Policy Rate (MPR), Commercial Lending Rate (CLR), Nominal Growth Rate (NGR) and Consumer Price Index (CPI) with Money Supply (MoS) as exogenous variable. The nature of trend in each series was investigated. The results revealed that quadratic trend model best models MTF, MPR, CLR and NGR whiles an exponential trend best models CPI. Johansen’s co-integration test with unrestricted trend performed revealed the existence of long-run equilibrium relationships between the variables and three (3) co-integrating equations described this long-run relationship. In terms of short-run relationships, the VEC (2) model revealed that, CLR, NGR, MoS have positive and significant impact on MTF. CLR, NGR and MoS have positive and significant impact on MPR, NGR have positive and significant impact on CLR, CPI and MoS have significant impact on NGR whiles NGR and MoS have significant impact on CPI. Model diagnostics performed on the VEC (2) model showed that, all the model parameters are structurally stable over time and the residuals of the individual models are free from serial correlation and conditional heteroscedasticity. Forecast error variance decomposition (FEVD) analysis showed that each variable primarily explained its own variance and the influence of other variables increase over time. Hence, adopting a broad perspective on macroeconomic variables can help policymakers anticipate and mitigate ripple effects across various economic sectors.

Does perceived usability of a multi-component mobile device intervention improve health-related quality of life?

Abstract Background Good health-related quality of life (HRQL) is essential to healthy aging. Prior studies have demonstrated that mobile device technologies can promote HRQL in older adults. However, evidence of technology acceptance in older adults is often inconsistent and poorly understood. Purpose This study aims to 1) describe technology acceptance using efficiency, learnability, user experience (i.e., acceptability), and perceived control from the USABILITY survey of a multi-component mobile device intervention consisting of a wearable activity tracker (Fitbit Versa), access to Lose It app, and text messages (for experimental group only) in older adults at risk for cardiovascular disease, and 2) examine the correlation between technology acceptance and changes in HRQL before and after using each component of the digital intervention. Methods Data were collected using a purpose-built technology usability survey, and HRQL was assessed using the 12-Item Short Form Survey (SF-12), which measured perceived physical and mental health. Correlations were used to identify associations between technology acceptance and HRQL. Results The mean age of participants (n=54) was 65.4 ± 6.0 years; a majority were females (60%), married (53%), Hispanic (66%), and had a high school education or higher (76%). Table 1 shows that the experimental group had considerably higher HRQL at three months, but no differences in technology acceptance were noted. Fitbit technology adoption was found to be correlated with physical HRQL (r=304, p=0.027). The Lose It app showed no correlation with HRQL. Despite receiving higher usability ratings, SMS messages were not correlated with physical or mental HRQL. Conclusion Our findings demonstrate that older persons at risk for cardiovascular disease find mobile device intervention components acceptable. However, each component affected physical and mental health differently. These findings inform more complete technology acceptance models for older individuals' technology design, education, and deployment. Older persons prefer personalized communication, such as customized text messages, as the most favored aspect of the multi-component mobile device intervention. It warrants a larger clinical trial to determine its efficacy on HRQL.

Prediction of 10-year risk for recurrent atherosclerotic cardiovascular events in a secondary prevention population: Insights from CHART-2 and Suita studies in the Japanese population

Abstract Background Despite the European Society of Cardiology (ESC) guidelines advocating risk prediction models for secondary prevention in established atherosclerotic cardiovascular disease (ASCVD) patients, such models for Japanese individuals remain limited, owing to their comparatively lower ASCVD risk than Western populations. Given this gap, there is a need to develop tailored risk assessment tools for Japanese ASCVD patients to optimize secondary prevention strategies. Purpose Leveraging data from two large-scale, long-term cardiovascular cohorts in Japan, our aim was to develop and externally validate a 10-year risk model predicting recurrent cardiovascular events among patients with established ASCVD. Methods The CHART-2 and Suita studies, large-scale prospective observational multicenter cohort studies in Japan, served as the primary data sources for this investigation. From the CHART-2 study, which comprised 3,560 patients with a history of coronary artery disease or ischemic stroke, we derived our risk model. External validation was conducted using data from the Suita study, which included 445 patients with similar ASCVD backgrounds. The primary outcome of interest was a composite endpoint encompassing cardiovascular death, myocardial infarction, or stroke. Variable selection was performed based on clinically relevant factors previously employed in established risk scores (SMART, REACH, and TRS 2°P risk scores). Logistic regression was utilized for model development, and the predictive performance was assessed using the C-index. Results Over a median follow-up of 10.1 years, the derivation cohort experienced 756 cardiovascular events. Patients in the derivation cohort had an average age of 68.9 ± 10.6 years, with 91% and 24% having a history of coronary artery disease and ischemic stroke, respectively. Using logistic regression, the multivariate model revealed a higher recurrent risk associated with several factors including age, smoking status, systolic blood pressure, diabetes, renal function, number of ASCVD events, atrial fibrillation, lipid levels, left ventricular systolic dysfunction, and statin use. A simple risk score based on these predictors demonstrated a &amp;gt;5-fold gradient of 10-year recurrent risk with a C-index of 0.65 (95% CI 0.63-0.68, Figure). The prediction model was externally validated in the Suita study wit 245 cardiovascular events and a median follow-up of 10.4 years, showing good discriminative ability with a C-index of 0.64 (95% CI 0.60–0.68). Conclusion Our newly developed prediction model offers a valuable tool for identifying individuals at heightened risk of recurrent cardiovascular events over a 10-year period, thereby facilitating more targeted and effective secondary prevention strategies tailored to the unique characteristics of Japanese ASCVD patients.

Prediction of family inheritance in non-sarcomeric hypertrophic cardiomyopathy using clinical variables and 12-lead electrocardiogram artificial intelligence models

Abstract Background Hypertrophic cardiomyopathy (HCM) can be familial even in the absence of a causative sarcomeric variant, justifying periodical screening of first-degree relatives. However, indiscriminate screening may burden low-risk individuals unnecessarily. The prevalence and predictors of familial disease in non-sarcomeric HCM remain inadequately characterized. Purpose Our study aims to develop a tailored approach to family screening based on index cases’ clinical profiles. By leveraging artificial intelligence (AI) models, we aim to predict the likelihood of family disease using pertinent clinical variables and electrocardiograms (ECGs) from the proband. Methods From January 2008 to June 2022, all probands diagnosed with HCM at our Heart Institute genetics clinic were included if they had i) negative genetic testing for a (likely) pathogenic variant in a sarcomeric gene and ii) ≥1 first- or second-degree relative screened for HCM. The outcome was the presence of family disease defined as left ventricular wall thickness ≥13mm or verified autopsy report confirming HCM. Predictors of family disease were assessed at the first genetic clinic visit and considered age, sex, hypertension, and echocardiography metrics (maximal wall thickness, septal morphology, outflow tract obstruction at rest or Valsava). Using 12-lead ECGs and clinical variables, we used machine learning to train two distinct AI models. Model A utilized ECG signals only to train a ResNet model, while Model B used both ECGs and selected clinical variables (below 0.2 alpha level in logistic regressions) to train a random forest model to predict family disease. We performed evaluations at both ECG-level (one ECG used per patient) and patient-level (all available ECGs used in a patient) and we present individual and ensemble model performance. Results Among 356 families with sarcomere-negative HCM, 106 (30%) were familial. Probands were aged 53 ± 16 years, 30% female, 36% with a diagnosis of hypertension and 43% with peak outflow tract obstruction ≥30mmHg. In univariable logistic regressions, family disease was associated with age at diagnosis (OR 0.89 per 5-year, P=0.003), hypertension (OR 0.60, P=0.046) and obstruction (OR 0.52, P=0.013). The ensemble AI model, which integrated clinical and ECG data, achieved an AUC of 0.82 (95% CI: [0.795, 0.848]). It demonstrated a specificity of 0.891 (95% CI: [0.867, 0.913]) and a negative predictive value of 0.798 (95% CI: [0.772, 0.823]). When comparing ECG-level and patient-level data (1 to 10 ECGs per patient), the models consistently achieved AUC values between 0.78 and 0.81, indicating stable performance across various data subsets. Conclusion The prevalence of family inheritance in sarcomere-negative HCM can be accurately predicted using AI models that integrate clinical variables and 12-lead ECG data. This approach could enable tailored screening strategies based on the likelihood of family disease.

Enhancing streamflow prediction in the Wujiang River basin: a two-stage decomposition approach with deep learning integration

ABSTRACT Global climate change and human activities have profoundly impacted the geological and hydrological processes in watersheds, increasing the challenges in streamflow prediction. In this study, we propose a streamflow prediction model based on deep learning and dual-mode decomposition specifically tailored for the Wujiang River basin, a significant tributary on the southern bank of the upper Yangtze River. This model effectively addresses the prediction error issue caused by high-frequency components through dual-mode decomposition of time-series data. The results demonstrate that employing the coupled complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN)–variational mode decomposition (VMD)–temporal convolutional network (TCN)–long short-term memory network (LSTM) model reduces prediction errors by at least 35% compared to single decomposition models. Furthermore, compared to individual TCN or LSTM models, the TCN–LSTM coupled model exhibits greater stability and higher prediction accuracy during training, with reductions in mean absolute error and root mean square error by 43.13 and 24.57%, respectively. This model holds promising prospects for application and can provide crucial insights for water resource management.

The phobic brain: Morphometric features correctly classify individuals with small animal phobia.

Specific phobia represents an anxiety disorder category characterized by intense fear generated by specific stimuli. Among specific phobias, small animal phobia (SAP) denotes a particular condition that has been poorly investigated in the neuroscientific literature. Moreover, the few previous studies on this topic have mostly employed univariate analyses, with limited and unbalanced samples, leading to inconsistent results. To overcome these limitations, and to characterize the neural underpinnings of SAP, this study aims to develop a classification model of individuals with SAP based on gray matter features, by using a machine learning method known as the binary support vector machine. Moreover, the contribution of specific structural macro-networks, such as the default mode, the salience, the executive, and the affective networks, in separating phobic subjects from controls was assessed. Thirty-two subjects with SAP and 90 matched healthy controls were tested to this aim. At a whole-brain level, we found a significant predictive model including brain structures related to emotional regulation, cognitive control, and sensory integration, such as the cerebellum, the temporal pole, the frontal cortex, temporal lobes, the amygdala and the thalamus. Instead, when considering macro-networks analysis, we found the Default, the Affective, and partially the Central Executive and the Sensorimotor networks, to significantly outperform the other networks in classifying SAP individuals. In conclusion, this study expands knowledge about the neural basis of SAP, proposing new research directions and potential diagnostic strategies.

Compliance of eating model of individuals with metabolic syndrome with Bulgarian dietary guidelines

Abstract Background The incidence of metabolic syndrome progressively increases among the urban population of Europe. About 1/3 of the Bulgarian adults have metabolic syndrome. Changing the dietary pattern can be a modern approach to control it. The aim of the study is to investigate the eating model of people with metabolic syndrome according to the Bulgarian dietary guidelines. Methods Two hundred individuals with metabolic syndrome aged 54.6±10.5 y. were randomly selected from general practitioners’ practices in Pleven region, Bulgaria. A Food Frequency Questionnaire method was applied to assess the eating pattern of the respondents. The frequency of food consumption of different foods was compared with the dietary guidelines for the Bulgarian population over 18 years of age. Data were processed with IBM SPSS v. 25. Results Grain consumption met the recommendations in almost all respondents. Only 29% consumed fresh fruits and 42% had intake of milk and milk products according to the requirements. Women ate more fresh vegetables and fruits and dairy products than men (p = 0,032). Risk meat consumption, 2-4 times a week, was found: men preferring more red meat (45%) and women - chicken and poultry (56%). Fish consumption met recommendation (once a week) in only 14.7% of the respondents. The food intake of eggs, vegetables, legumes and nuts were low. The preferred oils were sunflower and olive (once a day by 86% and 15% respectively), and butter (21% - 2-4 times a week). The intake of table salt corresponds to the recommendations in two thirds of the participants. One third consumed confectionery every day. Half did not drink alcohol, but 16% used concentrates once a day. Conclusions Foods that increase the nutritional risk for the development of metabolic syndrome and type 2 diabetes - high consumption of confectionery and red meat, low consumption of vegetables, fish and legumes were found. Unbalanced eating model requires new public initiatives to prevent metabolic syndrome. Key messages • Men with metabolic syndrome have a higher consumption of unhealthy foods than women. • Modern Bulgarian dietary guidelines are an important tool for prevention of metabolic syndrome.