Prediction Bias Research Articles

Mitigating bias in AI mortality predictions for minority populations: a transfer learning approach

BackgroundThe COVID-19 pandemic has highlighted the crucial role of artificial intelligence (AI) in predicting mortality and guiding healthcare decisions. However, AI models may perpetuate or exacerbate existing health disparities due to demographic biases, particularly affecting racial and ethnic minorities. The objective of this study is to investigate the demographic biases in AI models predicting COVID-19 mortality and to assess the effectiveness of transfer learning in improving model fairness across diverse demographic groups.MethodsThis retrospective cohort study used a population-based dataset of COVID-19 cases from the Centers for Disease Control and Prevention (CDC), spanning the years 2020–2024. The study analyzed AI model performance across different racial and ethnic groups and employed transfer learning techniques to improve model fairness by adapting pre-trained models to the specific demographic and clinical characteristics of the population.ResultsDecision Tree (DT) and Random Forest (RF) models consistently showed improvements in accuracy, precision, and ROC-AUC scores for Non-Hispanic Black, Hispanic/Latino, and Asian populations. The most significant precision improvement was observed in the DT model for Hispanic/Latino individuals, which increased from 0.3805 to 0.5265. The precision for Asians or Pacific Islanders in the DT model increased from 0.4727 to 0.6071, and for Non-Hispanic Blacks, it rose from 0.5492 to 0.6657. Gradient Boosting Machines (GBM) produced mixed results, showing accuracy and precision improvements for Non-Hispanic Black and Asian groups, but declines for the Hispanic/Latino and American Indian groups, with the most significant decline in precision, which dropped from 0.4612 to 0.2406 in the American Indian group. Logistic Regression (LR) demonstrated minimal changes across all metrics and groups. For the Non-Hispanic American Indian group, most models showed limited benefits, with several performance metrics either remaining stable or declining.ConclusionsThis study demonstrates the potential of AI in predicting COVID-19 mortality while also underscoring the critical need to address demographic biases. The application of transfer learning significantly improved the predictive performance of models across various racial and ethnic groups, suggesting these techniques are effective in mitigating biases and promoting fairness in AI models.

A multi-stage multivariable offshore wind speed assessment system incorporating statistical analysis, decomposition, feature selection, bias correction, and attention-based hybrid prediction

Efficient and accurate analysis of wind resources is essential for offshore wind power assessment and energy system dispatch. Different from single-variable prediction, this work proposes a multi-stage multivariable offshore wind speed assessment system. Firstly, the main variable is decomposed into three terms using seasonal-trend decomposition with loess (STL). Six highly correlated variables are then selected from 13 wind speed factors using a combination of three methods. Prediction model is built using Bidirectional Long Short-Term Memory networks (BiLSTM) based on Multi-Head Self-Attention (MSA) mechanisms, wherein Convolutional Neural Networks (CNNs) and Gated Recurrent Units (GRUs) are added. Sub-term prediction biases are corrected using Artificial Neural Networks (ANNs) optimized by the Dung Beetle Optimizer (DBO). Wind speed for four specific months representing the four seasons of an offshore wind farm is computed. Comparative experiments show that the proposed method achieves excellent performance with high prediction accuracy.

A Study on Canopy Volume Measurement Model for Fruit Tree Application Based on LiDAR Point Cloud

The accurate measurement of orchard canopy volume serves as a crucial foundation for wind regulation and dosage adjustments in precision orchard management. However, existing methods for measuring canopy volume fail to satisfy the high precision and real-time requirements necessary for accurate variable-rate applications in fruit orchards. To address these challenges, this study develops a canopy volume measurement model for orchard spraying using LiDAR point cloud data. In the domain of point cloud feature extraction, an improved Alpha Shape algorithm is proposed for extracting point cloud contours. This method improves the validity judgment for contour line segments, effectively reducing contour length errors on each 3D point cloud projection plane. Additionally, improvements to the mesh integral volume method incorporate the effects of canopy gaps in height difference calculations, significantly enhancing the accuracy of canopy volume estimation. For feature selection, a random forest-based recursive feature elimination method with cross-validation was employed to filter 10 features. Ultimately, five key features were retained for model training: the number of point clouds, the 2D point cloud contour along the X- and Z-projection directions, the 2D width in the Y-projection direction, and the 2D length in the Z-projection direction. During model construction, the study optimized the hyperparameters of partial least squares regression (PLSR), backpropagation (BP) neural networks, and gradient boosting decision trees (GBDT) to build canopy volume measurement models tailored to the dataset. Experimental results indicate that the PLSR model outperformed other approaches, achieving optimal performance with three principal components. The resulting canopy volume measurement model achieved an R2 of 0.9742, an RMSE of 0.1879, and an MAE of 0.1161. These results demonstrate that the PLSR model exhibits strong generalization ability, minimal prediction bias, and low average prediction error, offering a valuable reference for precision control of canopy spraying in orchards.

Detection of genetic variability in dairy cattle infectivity for bovine tuberculosis.

This study investigated the genetics of bovine tuberculosis (bTB) infectivity in Holstein-Friesian dairy cows using British national data. The analyses included cows with recorded sires from herds affected by bTB outbreaks between 2000 and 2022. Animals were considered bTB-positive if they reacted positively to the skin test and/or had positive post-mortem findings. We introduced the "index case approach," based on the assumption that once the initial positively tested animals (index cases) are detected in a herd, subsequent infections (secondary cases) in the early stages of the breakdown are likely to be attributed to these animals. Genetic analysis of the number of secondary cases (NSC) associated with a given index case was used to establish evidence of genetic variability in bTB infectivity of cattle, and derive Estimated Breeding Values (EBVs) for infectivity for the sires of the index cases. Data were analyzed by employing Markov Chain Monte Carlo techniques to fit Generalized Linear Mixed Models with either Poisson, Zero-Inflated Poisson (ZIP), Hurdle Poisson, or Geometric distributions. All 4 models demonstrated presence of genetic variance in cattle infectivity, with the strongest evidence provided by the ZIP and Hurdle Poisson models. The Hurdle Poisson model offered the most accurate and least biased predictions. Sire infectivity EBVs from the Poisson, ZIP, and Geometric models showed strong concordance, with pairwise correlations of 0.90 or higher. In contrast, correlations between EBVs from the Hurdle Poisson model and the other models ranged from 0.36 to 0.39. The association of the sire infectivity EBVs with the average observed NSC per sire and the proportion of infectious index case daughters per sire was generally moderate with correlations between 44 and 47% and 65-69%, respectively. Agreement among models for identifying the genetically most infectious sires was also reasonable, with 151 out of 285 sires appearing in the top 10% across models, and 122 (42.8%) also aligning with the top 10% based on observed average NSC. Results provide novel evidence for exploitable genetic variance in bTB infectivity allowing the derivation of meaningful EBVs. Based on the estimated posterior mean genetic variances obtained, reduction in infectivity by one genetic standard deviation would result in a 32% - 44% decrease in the expected NSC per index case. Further research is warranted to refine the phenotypic definition of infectivity and assess correlation with other dairy traits.

Fair and explainable Myocardial Infarction (MI) prediction: Novel strategies for feature selection and class imbalance correction

The rising incidences of myocardial infarction (MI), often affecting individuals without traditional risk factors, highlight the urgent need for improved early detection using personal health data. However, health surveys and electronic health records (EHRs) frequently suffer from class imbalances, leading to prediction biases and differences between specificity and sensitivity, which hinder reliable model development despite the valuable insights contained in these datasets. To address this, we have introduced a novel approach to enhance MI risk prediction using self-reported attributes from the Behavioral Risk Factor Surveillance System (BRFSS) and the National Health Interview Survey (NHIS) dataset. Our approach incorporates three innovative techniques: the Dual-Path Artificial Neural Network (DP-ANN) to mitigate biased decision making across imbalanced datasets, the Triple Criteria Selection (TCS) for unbiased feature selection, and Minority Weighted Sampling (MWS) to tackle challenges of uncontrolled minority class sampling. These methods collectively enhance MI prediction and feature relevance. The DP-ANN model has achieved balanced performance, with an average specificity of 80%, sensitivity of 82%, and AUC–ROC of 89.5%, improving imbalance variance by approximately 14.96% compared to prior studies. By outperforming other models across four heavily imbalanced datasets, our approach demonstrates robustness and generalizability. Additionally, SHapley Additive exPlanations (SHAP) analysis has revealed key predictors and risk factors for MI, such as coronary heart disease and bronchitis in females, and stroke among individuals aged 35–54. In conclusion, our study provides a robust model for healthcare professionals to assess MI risk through targeted factors, promoting early detection and potentially improving patient outcomes.

Preliminary Evaluation of a Numerical System of Prediction for Surface Solar Irradiance and Cloudiness in a Site with a Subtropical Humid Climate

This study explores a prediction system for global horizontal irradiance and cloudiness in a humid subtropical terrestrial region. This system consists of regional simulations performed with the Weather Research and Forecasting model using the initial and boundary conditions from the Global Forecast System. The predictions show significant biases for the variable of interest, with notable variations within the daily and annual cycles. This study also finds significant biases in cloud incidence and clarity index predictions, with relevant diurnal and seasonal variations. During austral summer, multiplying the relative humidity of initial and boundary conditions by a fixed factor improves the forecasts of global horizontal irradiance and cloud incidence for the central hours of the day and the afternoon. During austral winter, an empirical correction of the clarity index obtained from the simulation’s outputs also shows the potential to improve the forecasts’ biases. This work proposes a hypothesis about the causes of the forecast biases.

Clock bias prediction of navigation satellite based on BWO-CNN-BiGRU-attention model

Clock bias prediction of navigation satellite based on BWO-CNN-BiGRU-attention model

Equations to Predict Carbon Monoxide Emissions from Amazon Rainforest Fires

Earth systems models (ESMs), which can simulate the complex feedbacks between climate and fires, struggle to predict fires well for tropical rainforests. This study provides equations that predict historic carbon monoxide emissions from Amazon rainforest fires for 2003–2018, which could be implemented within ESMs’ current structures. We also include equations to convert the predicted emissions to burned area. Regressions of varying mathematical forms are fitted to one or both of two fire CO emission inventories. Equation accuracy is scored on r2, bias of the mean prediction, and ratio of explained variances. We find that one equation is best for studying smoke consequences that scale approximately linearly with emissions, or for a fully coupled ESM with online meteorology. Compared to the deforestation fire equation in the Community Land Model ver. 4.5, this equation’s linear-scale accuracies are higher for both emissions and burned area. A second equation, more accurate when evaluated on a log scale, may better support studies of certain health or cloud process consequences of fires. The most accurate recommended equation requires that meteorology be known before emissions are calculated. For all three equations, both deforestation rates and meteorological variables are key groups of predictors. Predictions nevertheless fail to reproduce most of the variation in emissions. The highest linear r2s for monthly and annual predictions are 0.30 and 0.41, respectively. The impossibility of simultaneously matching both emission inventories limits achievable fit. One key cause of the remaining unexplained variability appears to be noise inherent to pan-tropical data, especially meteorology.

A Dual Filter Based on Radial Basis Function Neural Networks and Kalman Filters with Application to Numerical Wave Prediction Models.

The aim of this study is to introduce and evaluate a dual filter that combines Radial Basis Function neural networks and Kalman filters to enhance the accuracy of numerical wave prediction models. Unlike the existing methods, which focus solely on systematic errors, the proposed framework concurrently targets both systematic and non-systematic parts of forecast errors, significantly reducing the bias and variability in significant wave height predictions. The produced filter is self-adaptive, identifying optimal Radial Basis Function network configurations through an automated process involving various network parameters tuning. The produced computational system is assessed using a time-window procedure applied across divergent time periods and regions in the Aegean Sea and the Pacific Ocean. The results reveal a consistent performance, outperforming classic Kalman filters with an average reduction of 53% in bias and 28% in RMSE, underlining the dual filter's potential as a robust post-processing tool for environmental simulations.

Dynamically downscaled seasonal ocean forecasts for North American east coast ecosystems

Abstract. Using a 1/12° regional model of the Northwest Atlantic Ocean (MOM6-NWA12), we downscale an ensemble of retrospective seasonal forecasts from a 1° global forecast model. To evaluate whether downscaling improved the forecast skill for surface temperature and salinity and bottom temperature, the global and downscaled forecasts are compared with each other and with a reference forecast of persistence using anomaly correlation. Both sets of forecasts are also evaluated on the basis of mean bias and ensemble spread. We find that downscaling significantly improved the forecast skill for monthly sea surface temperature anomalies in the Northeast US Large Marine Ecosystem, a region that global models have historically struggled to skillfully predict. The downscaled sea surface temperature (SST) predictions for this region were also more skillful than the persistence baseline across most initialization months and lead times. Although some of the SST prediction skill in this region stems from the recent rapid warming trend, prediction skill above persistence is generally maintained after removing the contribution of the trend, and patterns of skill suggestive of predictable processes are also preserved. While downscaling mainly improved the SST anomaly prediction skill in the Northeast US region, it improved bottom temperature and sea surface salinity anomaly skill across many of the marine ecosystems along the North American east coast. Although improvements in anomaly prediction via downscaling were ubiquitous, the effects of downscaling on prediction bias were mixed. Downscaling generally reduced the mean surface salinity biases found in the global model, particularly in regions with sharp salinity gradients (the Northern Gulf of Mexico and the Northeast US). In some cases, however, downscaling amplified the surface and bottom temperature biases found in the global predictions. We discuss several processes that are better resolved in the regional model and contribute to the improved skill, including the autumn reemergence of temperature anomalies and advection of water masses by coastal currents. Overall, the results show that a downscaled high-resolution model can produce improved seasonal forecast skill by representing fine-scale processes that drive predictability.

Seeing Beyond Noise: Improving Cryptocurrency Forecasting with Linear Bias Correction

Cryptocurrency is recognized as a leading digital currency by its peer-to-peer transfer capabilities and secure features. Accurately forecasting cryptocurrency price trends holds substantial significance for investors and traders, as they inform critical decisions regarding the acquisition, divestment, or retention of cryptocurrencies, guided by expectations of value, risk assessment, and potential returns. This study also aims to identify a resourceful technique to efficiently forecast prices of cryptocurrencies such as Bitcoin (BTC), Binance (BNB), Ripple (XRP), and Tether (USDT) using optimal data-driven models (LSTM, GRU, and BiLSTM models) using bias correction. The proposed methodology includes collecting cryptocurrency data and precious metal data from Coindesk and BullionVault, respectively, and then finding the optimal model input combination for each cryptocurrency by lag adjustment and correlating feature selection. Hyperparameter tuning was performed by trial-and-error technique, and an early stopping function was applied to minimize time and space complexity. Bias correction (BC) is applied to model-forecasted price trends to reduce errors in evaluation and to enhance accuracy by adjusting model outputs to reduce prediction bias, providing a refined alternative to traditional unadjusted deep learning methods. GRU-BC outperformed other models in forecasting Bitcoin (with MAE 25.291, RMSE 31.266, MAPE 2.999) and USDT (with MAE 0.0006, RMSE 0.0012, MAPE 0.0622) price trends, while BiLSTM-BC was superior in predicting XRP (with MAE 0.0129, RMSE 0.0171, MAPE 2.9013) and BNB (with MAE 2.2759, RMSE 2.8357, MAPE 1.9785) market price flow.

A Large-Scale Inter-Comparison and Evaluation of Spatial Feature Engineering Strategies for Forest Aboveground Biomass Estimation Using Landsat Satellite Imagery

Aboveground biomass (AGB) estimates derived from Landsat’s spectral bands are limited by spectral saturation when AGB densities exceed 150–300 Mg ha−1. Statistical features that characterize image texture have been proposed as a means to alleviate spectral saturation. However, apart from Gray Level Co-occurrence Matrix (GLCM) statistics, many spatial feature engineering techniques (e.g., morphological operations or edge detectors) have not been evaluated in the context of forest AGB estimation. Moreover, many prior investigations have been constrained by limited geographic domains and sample sizes. We utilize 176 lidar-derived AGB maps covering ∼9.3 million ha of forests in the Pacific Northwest of the United States to construct an expansive AGB modeling dataset that spans numerous biophysical gradients and contains AGB densities exceeding 1000 Mg ha−1. We conduct a large-scale inter-comparison of multiple spatial feature engineering techniques, including GLCMs, edge detectors, morphological operations, spatial buffers, neighborhood vectorization, and neighborhood similarity features. Our numerical experiments indicate that statistical features derived from GLCMs and spatial buffers yield the greatest improvement in AGB model performance out of the spatial feature engineering strategies considered. Including spatial features in Random Forest AGB models reduces the root mean squared error (RMSE) by 9.97 Mg ha−1. We contextualize this improvement model performance by comparing to AGB models developed with multi-temporal features derived from the LandTrendr and Continuous Change Detection and Classification algorithms. The inclusion of temporal features reduces the model RMSE by 18.41 Mg ha−1. When spatial and temporal features are both included in the model’s feature set, the RMSE decreases by 21.71 Mg ha−1. We conclude that spatial feature engineering strategies can yield nominal gains in model performance. However, this improvement came at the cost of increased model prediction bias.

Comprehensive propagation of errors for the prediction of woody biomass

Abstract Managing vegetation to sequester carbon in biomass requires estimates to meet standards for accuracy, with methods that are transparent, verifiable and cost‐effective. Allometric models are commonly used to predict biomass from non‐destructive field inventory data. Although a number of studies have addressed biomass error propagation, none have provided a general set of methods for linking errors all the way from initial allometric model development through to the final site‐based biomass prediction, for both above‐ and below‐ground biomass. Error sources in total biomass (above‐ + below‐ground) were quantified using a combination of analytical and Monte Carlo methods, illustrated with four contrasting case studies using either site‐ and‐species‐specific, species‐specific or generalised allometric models. Sampling error was found to be the most important contributor to site‐level biomass uncertainty, arising from the interaction between spatial variability and the field sampling design. The contribution of allometric model covariance to total error was also quantified, with errors in the determination of moisture content during allometric model development identified as a potentially important yet often overlooked error source. Application of different allometric models to the same inventory data suggested the error from generalised models was no greater than that from site‐ or species‐specific models, with increases in the generalised model prediction error balanced by decreases in other error sources associated with the increased sample size on which generalised models are based. Recommendations for reducing errors in predicted biomass include increasing field survey sample size, adopting field survey designs that ensure spatial representativeness and improving moisture content measurement protocols and increasing the moisture content sample size during allometric model development. To reduce costs while maintaining acceptable accuracy, the use of generalised allometric models is recommended, with the caveat that additional biomass sampling for model validation may be required to limit the potential for biased predictions.

Bridging the Gap Between Global Weather Prediction and Global Storm‐Resolving Simulation: Introducing the GFDL 6.5‐km SHiELD

AbstractWe introduce a 6.5‐km version of the Geophysical Fluid Dynamics Laboratory (GFDL)'s System for High‐resolution prediction on Earth‐to‐Local Domains (SHiELD). This global model is designed to bridge the gap between global medium‐range weather prediction and global storm‐resolving simulation while remaining practical for real‐time forecast. The 6.5‐km SHiELD represents a significant advancement over GFDL's flagship global forecast system, the 13‐km SHiELD. This global model features a holistically‐developed scale‐aware suite of physical parameterizations, stepping into the formidable convective “gray zone” of resolutions below 10 km. Comparative analyses with the 13‐km SHiELD, conducted over a 3‐year hindcast period, highlight noteworthy improvements across global‐scale, regional‐scale, tropical cyclone (TC), and continental convection predictions. In particular, the 6.5‐km SHiELD excels in predicting considerably finer‐scale convective systems associated with large‐scale frontal systems and extratropical cyclones. The predictions of global temperature, wind, cloud, and precipitation are significantly improved in this global model. Regionally, over the contiguous United States and the Maritime Continent, substantial reductions in prediction biases of precipitation, cloud cover, and wind fields are also found. In the mesoscale realm, the model demonstrates prominent improvements in global TC intensity and continental convective precipitation prediction: biases are relieved, and skill is higher. These findings affirm the superiority of the 6.5‐km SHiELD compared to the current 13‐km SHiELD, which will advance weather prediction by successfully addressing both synoptic weather systems and specific storm‐scale phenomena in the same global model.

Development of a machine learning model in prediction of the rapid progression of interstitial lung disease in patients with idiopathic inflammatory myopathy.

Rapidly progressive interstitial lung disease (RP-ILD) significantly impacts the prognosis of patients with idiopathic inflammatory myopathies (IIM). High-resolution computed tomography (HRCT) is a crucial noninvasive technique for evaluating interstitial lung disease (ILD). Utilizing quantitative computed tomography (QCT) enables accurate quantification of disease severity and evaluation of prognosis, thereby serving as a crucial computer-aided diagnostic method. This study aimed to establish and validate a machine learning (ML) model to predict RP-ILD in patients with idiopathic inflammatory myopathy-related interstitial lung disease (IIM-ILD) based on QCT and clinical features. A total of 514 patients (367 females, median age 54 years) with IIM-ILD in the China-Japan Friendship Hospital were retrospectively included, out of which 249 cases (165 females, median age 55 years) were identified as having RP-ILD. To extract the quantitative features on HRCT, deep learning (DL) methods were employed, along with demographic factors, pulmonary function test results, and blood gas analysis results; these factors were integrated into a final prediction model. Logistic regression was chosen as the final model due to its superior area under the curve (AUC) and explainability compared to the other seven ML models. The validation dataset yielded an AUC of 0.882 [95% confidence interval (CI): 0.797-0.967], indicating that the combined QCT and clinical features model outperformed both the QCT-only model and the clinically-only model. In calibration and clinical decision curve analysis, the final model demonstrated minimal prediction bias (concordance index: 0.887, 95% CI: 0.800-0.974, P<0.001) and provided greater net benefit across most thresholds. The nomogram encompassed the incorporation of the following variables: subtype, gender, forced expiratory volume in one second (FEV1%), diffusing capacity for carbon monoxide (DLCO%), oxygenation index (OI), and quantitative ground-glass opacities (GGOs), consolidation, pulmonary vascular, and branches on HRCT. When utilizing ML techniques, the baseline QCT has the potential to predict rapid progression in patients with IIM-ILD. The prediction performance will be further improved by incorporating clinical data alongside HRCT features. Idiopathic inflammatory myopathy (IIM); rapidly progressive interstitial lung disease (RP-ILD); high-resolution computed tomography (HRCT); machine learning (ML); quantitative computed tomography (QCT).

Deep Learning for Karolinska Sleepiness Scale Classification Based On Eye Aspect Ratio with SMOTE-Enhanced Data Balancing

This paper addresses the challenge of accurately classifying sleepiness levels based on the Karolinska Sleepiness Scale (KSS) using Eye Aspect Ratio (EAR) data, especially when class imbalance leads to biased predictions. The research proposes a deep learning framework that integrates a Multi-Layer Perceptron (MLP) with the Synthetic Minority Over-sampling Technique (SMOTE) to balance the training data. EAR features, representing eye closure patterns, are extracted from video frames, and SMOTE is applied to generate synthetic data for underrepresented sleepiness classes. By training the MLP model on this balanced dataset, the system achieves a 97.6% classification accuracy in distinguishing four distinct sleepiness levels based on the KSS, demonstrating its effectiveness in reducing prediction bias and managing class imbalance, both crucial for real-time drowsiness detection systems

Optimizing Diabetic Neuropathy Severity Classification Using Electromyography Signals Through Synthetic Oversampling Techniques

Electromyography signals are electrical signals generated by muscle activity and are very useful for analyzing the health conditions of muscles and nerves. Data imbalance is a prevalent issue in EMG signal data, especially when addressing patients with varied health conditions and restricted data availability. A major difficulty for machine learning models is class imbalance in datasets, which frequently leads to biased predictions favoring the dominant class and neglecting the minority classes. The data augmentation method employs the Synthetic Minority Over Sampling Technique (SMOTE) and Random Over Sampling (ROS) to address data imbalances and enhance the performance of classification models for underrepresented classes. This study employs an oversampling technique to enhance the efficacy of the XG Boost model. SMOTE exhibits better efficacy relative to competing methods; the application of appropriate oversampling techniques allows models to integrate patterns from both majority and often neglected minority data.

The Prioritization of Prospection.

Humans frequently engage in mental time travel, reflecting on the past and anticipating the future. Although these processes may seem similar, research documents systematic differences between retrospection and prospection. We propose a conceptual framework to organize and explain these differences based on three axiomatic temporal asymmetries: The past occurs before the future; the past is more certain than the future; and the past is less controllable than the future. People's experience with these axiomatic differences is internalized and overgeneralized to shape mental representations of the past and future. Our review shows that people generally prioritize prospection over retrospection, attending more to the future than the past and reacting more intensely to future events than to past events. We consider potential moderators of and constraints on the generality of prioritizing prospection. We explore the implications of these temporal asymmetries, emphasizing their theoretical and practical significance. While daily life centers on the present, people often reflect on the past and anticipate the future. But which direction of mental time travel-backward or forward-has more influence? We identify three key differences that shape how people engage with the past and future: time flows from past to future, the future is more uncertain, and people have more control over the future. These differences affect the frequency, intensity, and nature of thoughts and feelings, leading to predictable biases in how we mentally represent and emotionally engage with events over time. Because focusing on the future often provides greater benefits, people tend to prioritize prospection over retrospection in everyday life.

Mitigating Algorithmic Bias in AI-Driven Cardiovascular Imaging for Fairer Diagnostics.

Background/Objectives: The research addresses algorithmic bias in deep learning models for cardiovascular risk prediction, focusing on fairness across demographic and socioeconomic groups to mitigate health disparities. It integrates fairness-aware algorithms, susceptible carrier-infected-recovered (SCIR) models, and interpretability frameworks to combine fairness with actionable AI insights supported by robust segmentation and classification metrics. Methods: The research utilised quantitative 3D/4D heart magnetic resonance imaging and tabular datasets from the Cardiac Atlas Project's (CAP) open challenges to explore AI-driven methodologies for mitigating algorithmic bias in cardiac imaging. The SCIR model, known for its robustness, was adapted with the Capuchin algorithm, adversarial debiasing, Fairlearn, and post-processing with equalised odds. The robustness of the SCIR model was further demonstrated in the fairness evaluation metrics, which included demographic parity, equal opportunity difference (0.037), equalised odds difference (0.026), disparate impact (1.081), and Theil Index (0.249). For interpretability, YOLOv5, Mask R-CNN, and ResNet18 were implemented with LIME and SHAP. Bias mitigation improved disparate impact (0.80 to 0.95), reduced equal opportunity difference (0.20 to 0.05), and decreased false favourable rates for males (0.0059 to 0.0033) and females (0.0096 to 0.0064) through balanced probability adjustment. Results: The SCIR model outperformed the SIR model (recovery rate: 1.38 vs 0.83) with a -10% transmission bias impact. Parameters (β=0.5, δ=0.2, γ=0.15) reduced susceptible counts to 2.53×10-12 and increased recovered counts to 9.98 by t=50. YOLOv5 achieved high Intersection over Union (IoU) scores (94.8%, 93.7%, 80.6% for normal, severe, and abnormal cases). Mask R-CNN showed 82.5% peak confidence, while ResNet demonstrated a 10.4% accuracy drop under noise. Performance metrics (IoU: 0.91-0.96, Dice: 0.941-0.980, Kappa: 0.95) highlighted strong predictive accuracy and reliability. Conclusions: The findings validate the effectiveness of fairness-aware algorithms in addressing cardiovascular predictive model biases. The integration of fairness and explainable AI not only promotes equitable diagnostic precision but also significantly reduces diagnostic disparities across vulnerable populations. This reduction in disparities is a key outcome of the research, enhancing clinical trust in AI-driven systems. The promising results of this study pave the way for future work that will explore scalability in real-world clinical settings and address limitations such as computational complexity in large-scale data processing.

Boreal Summer Extratropical Intraseasonal Oscillation Prediction in Current Subseasonal‐To‐Seasonal Operational Models Over Eurasia

AbstractExtratropical intraseasonal oscillations (EISO), which is akin to the tropical ISO, significantly influence mid‐high‐latitude intraseasonal variability. Previous studies have shown that biases in the dynamical prediction of EISOs reduce the accuracy of subseasonal forecasts for extreme events, particularly in extratropical regions. However, the current dynamical prediction skills of EISO and associated dependence remain unclarified. Based on the recently proposed EISO metrics, which represent three dominant zonally propagating EISO modes traveling along the westerly jets, this study conducts a comprehensive evaluation of the EISOs' prediction skill in 13 state‐of‐the‐art subseasonal‐to‐seasonal (S2S) prediction project models. Firstly, the S2S models display skill to predict the EISOs out to a range of 9–20 days, which demonstrates higher predictability compared to other conventional circulation modes in the extratropics. In the subseasonal prediction, EISOs' amplitudes are underestimated, and their propagating speeds feature large random errors in all S2S models. Secondly, the EISOs' prediction skill depends on their initial/target amplitude, particularly for strong target amplitudes, which showcase higher skill than weak‐amplitude EISOs. Thirdly, although the EISOs' prediction skill is sensitive to the EISOs' phase, it is not locked to any fixed phase, indicating that EISOs do not have an inherent prediction barrier like the tropical ISO. This study also discusses how the humidity initialization and ensemble sizes influence the EISOs' prediction skill, suggesting that increasing the ensemble members in the saturation range is an effective way of improving the prediction for extratropical subseasonal variation and related extreme events.