Features For Prediction Research Articles

Multiple patterns of EEG parameters and their role in the prediction of patients with prolonged disorders of consciousness

IntroductionPrognostication in patients with prolonged disorders of consciousness (pDoC) remains a challenging task. Electroencephalography (EEG) is a neurophysiological method that provides objective information for evaluating overall brain function. In this study, we aim to investigate the multiple features of pDoC using EEG and evaluate the prognostic values of these indicators.MethodsWe analyzed the EEG features: (i) spectral power; (ii) microstates; and (iii) mismatch negativity (MMN) and P3a of healthy controls, patients in minimally conscious state (MCS), and unresponsive wakefulness syndrome (UWS). Patients were followed up for 6 months. A combination of machine learning and SHapley Additive exPlanations (SHAP) were used to develop predictive model and interpret the results.ResultsThe results indicated significant abnormalities in low-frequency spectral power, microstate parameters, and amplitudes of MMN and P3a in MCS and UWS. A predictive model constructed using support vector machine achieved an area under the curve (AUC) of 0.95, with the top 10 SHAP values being associated with transition probability (TP) from state C to F, time coverage of state E, TP from state D to F and D to F, mean duration of state A, TP from state F to C, amplitude of MMN, time coverage of state F, TP from state C to D, and mean duration of state E. Predictive models constructed for each component using support vector machine revealed that microstates had the highest AUC (0.95), followed by MMN and P3a (0.65), and finally spectral power (0.05).DiscussionThis study provides preliminary evidence for the application of microstate-based multiple EEG features for prognosis prediction in pDoC.Clinical trial registrationchictr.org.cn, identifier ChiCTR2200064099.

Contrastive-learning of language embedding and biological features for cross modality encoding and effector prediction

Identifying and characterizing virulence proteins secreted by Gram-negative bacteria are fundamental for deciphering microbial pathogenicity as well as aiding the development of therapeutic strategies. Effector predictors utilizing pre-trained protein language models (PLMs) have shown sound performance by leveraging extensive evolutionary and sequential protein features. However, the accuracy and sensitivity of effector prediction remain challenging. Here, we introduce a model named Contrastive-learning of Language Embedding and Biological Features (CLEF) leveraging contrastive learning to integrate PLM representations with supplementary biological features. Biologically information is captured in learned contextualized embeddings to yield meaningful representations. With cross-modality biological features, CLEF outperforms state-of-the-art (SOTA) models in predicting type III, type IV, and type VI secreted effectors (T3SEs/T4SEs/T6SEs) in enteric pathogens. All experimentally verified effectors in Enterohemorrhagic Escherichia coli and 41 of 43 experimentally verified T3SEs of Salmonella Typhimurium are recognized. Moreover, 12 predicted T3SEs and 11 predicted T6SEs are validated by extensive experiments in Edwardsiella piscicida. Furthermore, integrating omics data via CLEF framework enhances protein representations to illustrate effector-effector interactions and determine in vivo colonization-essential genes. Collectively, CLEF provides a blueprint to bridge the gap between in silico PLM’s capacity and experimental biological information to fulfill complicated tasks.

Age estimation by radiomics analysis of mandibular condylar cone beam computed tomography images.

The aim of this study was to estimate the legal age using the parameters obtained from radiomic analysis of the mandibular condyle in cone beam computed tomography (CBCT) images. The study group consisted of 300 mandibular condyles, which were categorized into six groups based on the age of the patients: 8-11years, 12-14years, 15-17years, 18-20years, 21-23years, and over 24years. Each patient's condyle was segmented individually using the 3D Slicer program. Radiomic features were extracted from the segmented images using the SlicerRadiomics plugin. Subsequently, three distinct models were developed with reference to three specific subgroups of the 12-14 age group, 15-17 age group, 18-20 age group and the efficacy of radiomic features in predicting the age of the patient was evaluated. The ROC analysis of the three radiomics scores (RS) yielded AUC values of 0.927, 0.860, and 0.769 for RS12-14, RS15-17, and RS18-20, respectively. The RS12-14 model exhibited the highest sensitivity and specificity values among the models, with 88% and 84.4%, respectively. Among the radiomic features extracted from the mandibular condyle in CBCT images, the most significant features, identified based on developed models and their respective coefficients, can be applied to estimate patients' ages. Future studies hold substantial potential for advancing this method, particularly in automating both the segmentation process and the derivation of formulae for age estimation. The use of radiomic features for age prediction presents a promising alternative method for developing fully automated systems in clinical practice.

Evaluation of inflammatory-thrombosis panel as a diagnostic tool for vascular Behçet's disease.

Vascular Behçet's disease (VBD) is prevalent in 40% of BD, but lacks laboratory biomarker for timely diagnosis. We aimed to establish a diagnostic panel for discerning VBD and non-VBD patients and identify hemostatic-thrombotic markers most related to VBD pathogenesis using machine learning algorithm. A total of 338 BD patients comprising 123 VBD and 215 non-VBD were enrolled. Twenty-six clinical and laboratory features selected from LassoCV were included in multiple classifier to choose the optimal model for VBD differentiation. The Shapley Additive exPlanations (SHAP) was employed to interpret the contribution of model features for VBD prediction. Logistic regression analysis and nomogram were conducted to screen risk factors of VBD. Inflammatory (neutrophils%, NK cells, IL-6), hematological (hemoglobin, hemoglobin distribution width (HDW)) and thrombosis (activated partial thromboplastin clotting time (APTT), D-dimer) parameters were elevated in VBD. Then we chose top contributors from XGBoost model and performed ten-fold cross validation, the diagnostic accuracy of which exceeded 0.90. Utilizing SHAP method, we identified higher incidence of arterial thrombosis or aneurysm and deep vein thrombosis, upregulated NK cell count, HDW, APTT and D-dimer, downregulated reticulocyte%, B cell count, red blood cell distribution width, cellular hemoglobin (CH) and TNF-α would ultimately generate the phenotype of VBD. Severity, hemoglobin, mean corpuscular hemoglobin, CH, HDW, APTT and D-dimer were found as potential risk factors for vascular outcomes among BD. Our study developed a well-performed model leveraging clinical and laboratory parameters for differentiating VBD. Inflammatory and thrombotic risk factors are potential contributors to VBD. Key Points • Inflammatory (neutrophils%, NK cells, IL-6), hematological (HGB, HDW) and thrombosis (APTT, D-dimer) parameters were elevated in VBD. • We firstly developed an inflammatory-thrombosis model as a diagnostic tool for VBD. • HGB, MCH, CH, HDW, APTT and D-dimer are potential risk factors for VBD.

Developing an interpretable machine learning model for diagnosing gout using clinical and ultrasound features.

To develop a machine learning (ML) model using clinical data and ultrasound features for gout prediction, and apply SHapley Additive exPlanations (SHAP) for model interpretation. This study analyzed 609 patients' first metatarsophalangeal (MTP1) joint ultrasound data from two institutions. Institution 1 data (n=571) were split into training cohort (TC) and internal testing cohort (ITC) (8:2 ratio), while Institution 2 data (n=92) served as external testing cohort (ETC). Key predictors were selected using Random Forest (RF), Least Absolute Shrinkage and Selection Operator (LASSO), and Extreme Gradient Boosting (XGBoost) algorithms. Six ML models were evaluated using standard performance metrics, with SHAP analysis for model interpretation. Five key predictors were identified: serum uric acid (SUA), deep learning (DL) model predictions, tophus, bone erosion, and double contour sign (DCs). The logistic regression (LR) model demonstrated optimal performance, achieving Area Under the Curve (AUC) values of 0.870 (95% CI: 0.820-0.920) in ITC and 0.854 (95% CI: 0.804-0.904) in ETC. The model showed good calibration with Brier scores of 0.138 and 0.159 in ITC and ETC, respectively. This study developed an interpretable ML model for gout prediction and utilized SHAP to elucidate feature contributions, establishing a foundation for future applications in clinical decision support for gout diagnosis.

The Application of Digital Twin for Indoor Air Quality Management: A Case Study in Hong Kong

With the advocacy of the development of smart cities, buildings are also evolving to become smarter and more user-friendly. Employing cyber-physical technologies, such as the Internet of Things (IoT), Digital Twin (DT) and Building Information Modeling (BIM), can effectively monitor and control the built environment. DT is an emerging tool that connects the building to its digital replica so that the building’s situation can be monitored and presented to users effectively. This paper proposes a DT framework for building management to discover the applications of DT for indoor environment monitoring and control. A case study of a DT-driven indoor air quality (IAQ) system is completed to demonstrate the implementation of DT in the built environment. The system demonstrated the key features of DT, including monitoring, simulation, control and prediction, and showed their roles and functions in IAQ management. Finally, the difficulties and limitations, i.e. sensor selection and installation, of implementing the DT system in the built environment are presented after the case study.

P-879. Predicting Antimicrobial Resistance and Empiric Treatment Failure in Hospital-Onset Sepsis Using Machine Learning on Electronic Health Records

Abstract Background Antimicrobial resistance (AMR) is a major cause of treatment failure in hospital-onset sepsis. Approaches to guide empiric antimicrobial therapy are urgently needed. We utilized machine learning to predict empiric treatment failure and AMR patterns within a high-risk hospital-onset sepsis cohort using electronic health records (EHRs). Methods We examined hospitalizations with documented bacterial infection from 2010 to 2023 at a tertiary-care academic hospital system. Hospital-onset sepsis was defined using CDC Adult Sepsis Event criteria. Empiric therapy was considered adequate if all pathogens isolated were susceptible to antimicrobials administered. AMR patterns included methicillin (MRSA), extended-spectrum beta-lactamase (ESBL), vancomycin (VRE), ceftriaxone, and carbapenem (CRE). Features for prediction were patient characteristics, hospitalization details, and clinical severity metrics measured within 24 hours prior to sepsis onset. Linear and non-linear machine learning models were evaluated using bootstrap validation and out-of-bag area under the receiver operating characteristic curve (AUC). Results Among 49,581 hospitalizations, 2019 (4%) hospital-onset sepsis encounters were identified. Mean (SD) age was 65 (16.4) years and 900 (45%) were female. At 24 hours after sepsis onset, 377 (19%) received inadequate empiric treatment and 911 (45%) infections expressed AMR. Those who received inadequate empiric treatment were 8.2 (95%CI: 6.2-10.9) times more likely to have an AMR infection. Random forest model was best-performing, with AUC (95%CI) for any AMR pattern, VRE, and CRE at 0.64 (0.63-0.64), 0.71 (0.69-0.72), and 0.70 (0.67-0.73), respectively. The AUC (95%CI) for predicting inadequate empiric antimicrobial treatment was 0.65 (0.64-0.66) with a sensitivity of 64% (58-69%) and specificity of 52% (46-68%) at optimal probability thresholds. Conclusion In a hospital-onset sepsis cohort with prevalent AMR, machine learning approaches utilizing limited EHR data accessible at initial sepsis recognition had moderate, but clinically insufficient, discriminatory ability in predicting AMR and empiric antimicrobial treatment failure. Future research will evaluate the utilization of large language models using unstructured EHRs. Disclosures All Authors: No reported disclosures

Accuracy of smartwatches in predicting distance running performance

ObjectiveThis study examined the accuracy of smartwatches in predicting running performance.MethodsA total of 154 amateur runners (123 males and 31 females) were recruited. After wearing the HUAWEI WATCH GT Runner for a minimum of six weeks, the runners' actual completion times for 5 km, 10 km, and half marathon distances were measured, resulting in 288 test instances. The predicted completion times for the same distances displayed on the watch on the test day were recorded simultaneously.ResultsThe actual and predicted performances for the 5, 10, and 21.1 km distances were highly correlated, with r ≥ 0.95 (p < 0.001) and r2 ≥ 0.9 for all three distances, an error rate between the measured and predicted values of less than 3%, and intraclass correlation coefficient ≥0.9. The bias ± 95% limits of agreement were −20.4 ± 44.2 s for 5 km, 4.1 ± 299.1 s for 10 km, and 143.8 ± 400.4 s for the half marathon.ConclusionsThis study confirmed that the smartwatch exhibits high precision in predicting 5 km, 10 km, and half marathon performances, with an accuracy exceeding 97%. The performance prediction features of smartwatches can effectively guide amateur runners in setting reasonable competition goals and preparing for races.

Machine-Learning Predictions of Cochlear Implant Functional Outcomes: A Systematic Review.

Cochlear implant (CI) user functional outcomes are challenging to predict because of the variability in individual anatomy, neural health, CI device characteristics, and linguistic and listening experience. Machine learning (ML) techniques are uniquely poised for this predictive challenge because they can analyze nonlinear interactions using large amounts of multidimensional data. The objective of this article is to systematically review the literature regarding ML models that predict functional CI outcomes, defined as sound perception and production. We analyze the potential strengths and weaknesses of various ML models, identify important features for favorable outcomes, and suggest potential future directions of ML applications for CI-related clinical and research purposes. We conducted a systematic literature search with Web of Science, Scopus, MEDLINE, EMBASE, CENTRAL, and CINAHL from the date of inception through September 2024. We included studies with ML models predicting a CI functional outcome, defined as those pertaining to sound perception and production, and excluded simulation studies and those involving patients without CIs. Using Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we extracted participant population, CI characteristics, ML model, and performance data. Sixteen studies examining 5058 pediatric and adult CI users (range: 4 to 2489) were included from an initial 1442 publications. Studies predicted heterogeneous outcome measures pertaining to sound production (5 studies), sound perception (12 studies), and language (2 studies). ML models use a variety of prediction features, including demographic, audiological, imaging, and subjective measures. Some studies highlighted predictors beyond traditional CI audiometric outcomes, such as anatomical and imaging characteristics (e.g., vestibulocochlear nerve area, brain regions unaffected by auditory deprivation), health system factors (e.g., wait time to referral), and patient-reported measures (e.g., dizziness and tinnitus questionnaires). Used ML models were tree-based, kernel-based, instance-based, probabilistic, or neural networks, with validation and test methods most commonly being k-fold cross-validation and train-test split. Various statistical measures were used to evaluate model performance, however, for studies reporting accuracy, the best-performing models for each study ranged from 71.0% to 98.83%. ML models demonstrate high predictive performance and illuminate factors that contribute to CI user functional outcomes. While many models showed favorable evaluation statistics, the majority were not adequately reported with regard to dataset characteristics, model creation, and validation. Furthermore, the extent of overfitting in these models is unclear and will likely result in poor generalization to new data. This suggests the need for more robust validation procedures and standardization in reporting, with the ultimate hope that the iterative improvement of these models will allow for their adoption as a future clinical tool.

Estimation of Pressure Pain in the Lower Limbs Using Electrodermal Activity, Tissue Oxygen Saturation, and Heart Rate Variability

Quantification of pain or discomfort induced by pressure is essential for understanding human responses to physical stimuli and improving user interfaces. Pain research has been conducted to investigate physiological signals associated with discomfort and pain perception. This study analyzed changes in electrodermal activity (EDA), tissue oxygen saturation (StO2), heart rate variability (HRV), and Visual Analog Scale (VAS) under pressures of 10, 20, and 30 kPa applied for 3 min to the thigh, knee, and calf in a seated position. Twenty participants were tested, and relationships between biosignals, pressure intensity, and pain levels were evaluated using Friedman tests and post-hoc analyses. Multiple linear regression models were used to predict VAS and pressure, and five machine learning models (SVM, Logistic Regression, Random Forest, MLP, KNN) were applied to classify pain levels (no pain: VAS 0, low: VAS 1–3, moderate: VAS 4–6, high: VAS 7–10) and pressure intensity. The results showed that higher pressure intensity and pain levels affected sympathetic nervous system responses and tissue oxygen saturation. Most EDA features and StO2 significantly changed according to pressure intensity and pain levels, while NN interval and HF among HRV features showed significant differences based on pressure intensity or pain level. Regression analysis combining biosignal features achieved a maximum R2 of 0.668 in predicting VAS and pressure intensity. The four-level classification model reached an accuracy of 88.2% for pain levels and 81.3% for pressure intensity. These results demonstrated the potential of EDA, StO2, HRV signals, and combinations of biosignal features for pain quantification and prediction.

Employees’ perceptions of the fairness of AI-based performance prediction features

Employees’ perceptions of the fairness of AI-based performance prediction features

Remaining Useful Life Prediction of Rolling Bearings Based on CBAM-CNN-LSTM.

Predicting the Remaining Useful Life (RUL) is vital for ensuring the reliability and safety of equipment and components. This study introduces a novel method for predicting RUL that utilizes the Convolutional Block Attention Module (CBAM) to address the problem that Convolutional Neural Networks (CNNs) do not effectively leverage data channel features and spatial features in residual life prediction. Firstly, Fast Fourier Transform (FFT) is applied to convert the data into the frequency domain. The resulting frequency domain data is then used as input to the convolutional neural network for feature extraction; Then, the weights of channel features and spatial features are assigned to the extracted features by CBAM, and the weighted features are then input into the Long Short-Term Memory (LSTM) network to learn temporal features. Finally, the effectiveness of the proposed model is verified using the PHM2012 bearing dataset. Compared to several existing RUL prediction methods, the mean squared error, mean absolute error, and root mean squared error of the proposed method in this paper are reduced by 53%, 16.87%, and 31.68%, respectively, which verifies the superiority of the method. Meanwhile, the experimental results demonstrate that the proposed method achieves good RUL prediction accuracy across various failure modes.

Integrated Crop Protection Management

This paper presents the development of an innovative mobile application designed to assist farmers by providing critical information for optimizing agricultural productivity. The app integrates three core functionalities: soil health analysis, monsoon prediction, and a marketplace for buying and selling crops. Soil health analysis uses advanced algorithms to assess soil quality, offering recommendations for improved crop growth. The monsoon prediction feature leverages meteorological data to forecast rainfall patterns, enabling farmers to plan for planting and irrigation more effectively. Additionally, the app’s buy-sell module facilitates direct trade between farmers, enhancing market accessibility and ensuring fair pricing. Through empirical testing and user feedback, the app demonstrated significant potential in improving farm management practices, increasing crop yields, and enhancing financial outcomes for farmers. This research underscores the importance of integrating technology into agriculture, offering a sustainable and data-driven solution to modern farming challenges.

Improving drug-target affinity prediction by adaptive self-supervised learning.

Computational drug-target affinity prediction is important for drug screening and discovery. Currently, self-supervised learning methods face two major challenges in drug-target affinity prediction. The first difficulty lies in the phenomenon of sample mismatch: self-supervised learning processes drug and target samples independently, while actual prediction requires the integration of drug-target pairs. Another challenge is the mismatch between the broadness of self-supervised learning objectives and the precision of biological mechanisms of drug-target affinity (i.e., the induced-fit principle). The former focuses on global feature extraction, while the latter emphasizes the importance of local precise matching. To address these issues, an adaptive self-supervised learning-based drug-target affinity prediction (ASSLDTA) was designed. ASSLDTA integrates a novel adaptive self-supervised learning (ASSL) module with a high-level feature learning network to extract the feature. The ASSL leverages a large amount of unlabeled training data to effectively capture low-level features of drugs and targets. Its goal is to maximize the retention of original feature information, thereby bridging the objective gap between self-supervised learning and drug-target affinity prediction and alleviating the sample mismatch problem. The high-level feature learning network, on the other hand, focuses on extracting effective high-level features for affinity prediction through a small amount of labeled data. Through this two-stage feature extraction design, each stage undertakes specific tasks, fully leveraging the advantages of each model while efficiently integrating information from different data sources, providing a more accurate and comprehensive solution for drug-target affinity prediction. In our experiments, ASSLDTA is much better than other deep methods, and the result of ASSLDTA is significantly increased by learning adaptive self-supervised learning-based features, which validates the effectiveness of our ASSLDTA.

Prediction of induction motor faults using machine learning.

Unplanned downtime in industrial sectors presents significant challenges, impacting both production efficiency and profitability. To tackle this issue, companies are actively working towards optimizing their operations and reducing disruptions that hinder their ability to meet customer demands and financial goals. Predictive maintenance, utilizing advanced technologies like data analytics, machine learning, and IoT devices, offers real-time equipment data monitoring and analysis. This research study centers on the development of a versatile machine-learning model for predicting faults in induction motors within industrial environments. Implementing such a model can enable proactive maintenance, ultimately leading to decreased downtime in industrial operations. The study involved the acquisition of a dataset comprising healthy and faulty conditions of four 3-phase induction motors, along with relevant features for fault prediction. Multiple machine learning algorithms were trained using this dataset, exhibiting promising performance. The Random Forest (RF) model achieved the highest accuracy at 0.91, closely followed by the Artificial Neural Network (ANN) and k-nearest Neighbors (k-NN) models, both achieving an accuracy of 0.9. Meanwhile, the Decision Tree (DT) model showed the lowest accuracy at 0.89. Further model evaluation was carried out using a confusion matrix, which provided a detailed breakdown of the models' performance for each class, revealing the number of correctly and incorrectly classified induction motor conditions. The results from the confusion matrix indicate that the models effectively classified the various states and conditions of the induction motors. To enhance model performance in future work, potential avenues include refining the ANN and RF models, exploring transfer learning or ensemble methods, and incorporating diverse datasets to improve generalization.

Additional prognostic value of coronary angiographic features for in-hospital outcomes prediction after acute coronary syndrome: Insight from ADDICT-ICCU study

Additional prognostic value of coronary angiographic features for in-hospital outcomes prediction after acute coronary syndrome: Insight from ADDICT-ICCU study

Self-interactive learning: Fusion and evolution of multi-scale histomorphology features for molecular traits prediction in computational pathology.

Predicting disease-related molecular traits from histomorphology brings great opportunities for precision medicine. Despite the rich information present in histopathological images, extracting fine-grained molecular features from standard whole slide images (WSI) is non-trivial. The task is further complicated by the lack of annotations for subtyping and contextual histomorphological features that might span multiple scales. This work proposes a novel multiple-instance learning (MIL) framework capable of WSI-based cancer morpho-molecular subtyping by fusion of different-scale features. Our method, debuting as Inter-MIL, follows a weakly-supervised scheme. It enables the training of the patch-level encoder for WSI in a task-aware optimisation procedure, a step normally not modelled in most existing MIL-based WSI analysis frameworks. We demonstrate that optimising the patch-level encoder is crucial to achieving high-quality fine-grained and tissue-level subtyping results and offers a significant improvement over task-agnostic encoders. Our approach deploys a pseudo-label propagation strategy to update the patch encoder iteratively, allowing discriminative subtype features to be learned. This mechanism also empowers extracting fine-grained attention within image tiles (the small patches), a task largely ignored in most existing weakly supervised-based frameworks. With Inter-MIL, we carried out four challenging cancer molecular subtyping tasks in the context of ovarian, colorectal, lung, and breast cancer. Extensive evaluation results show that Inter-MIL is a robust framework for cancer morpho-molecular subtyping with superior performance compared to several recently proposed methods, in small dataset scenarios where the number of available training slides is less than 100. The iterative optimisation mechanism of Inter-MIL significantly improves the quality of the image features learned by the patch embedded and generally directs the attention map to areas that better align with experts' interpretation, leading to the identification of more reliable histopathology biomarkers. Moreover, an external validation cohort is used to verify the robustness of Inter-MIL on molecular trait prediction.

PD_EBM: An Integrated Boosting Approach Based on Selective Features for Unveiling Parkinson's Disease Diagnosis With Global and Local Explanations

ABSTRACTEarly detection and characterization are crucial for treating and managing Parkinson's disease (PD). The increasing prevalence of PD and its significant impact on the motor neurons of the brain impose a substantial burden on the healthcare system. Early‐stage detection is vital for improving patient outcomes and reducing healthcare costs. This study introduces an ensemble boosting machine, termed PD_EBM, for the detection of PD. PD_EBM leverages machine learning (ML) algorithms and a hybrid feature selection approach to enhance diagnostic accuracy. While ML has shown promise in medical applications for PD detection, the interpretability of these models remains a significant challenge. Explainable machine learning (XML) addresses this by providing transparency and clarity in model predictions. Techniques such as Local Interpretable Model‐agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) have become popular for interpreting these models. Our experiment used a dataset of 195 clinical records of PD patients from the University of California Irvine (UCI) Machine Learning repository. Comprehensive data preparation included encoding categorical features, imputing missing values, removing outliers, addressing data imbalance, scaling data, selecting relevant features, and so on. We propose a hybrid boosting framework that focuses on the most important features for prediction. Our boosting model employs a Decision Tree (DT) classifier with AdaBoost, followed by a linear discriminant analysis (LDA) optimizer, achieving an impressive accuracy of 99.44%, outperforming other boosting models.

PREDICTING TIME TO COGNITIVE IMPAIRMENT USING BRAIN AGE

Abstract The neurological changes that occur during the asymptomatic pre-clinical phase of cognitive impairment (CI) can be quantified using brain age (BA) measures and leveraged to identify those at risk of clinical manifestation. We employ a machine learning survival model to predict future CI among healthy older adults using cognitive, demographic, genetic, and neuroimaging data including regional brain volume and anatomically specific measures of BA. Participants were 2,718 adults from NACC (1,795 females, mean age=69.14±10.91). Converters (N=343) were diagnosed with CI during the study, and non-converters (N=2,375) were CN across all visits. We predicted time to conversion (days) from baseline MRIs (185 brain volumes and regional BA measures) and 69 clinical variables. The model was trained on 1,846 converters and 252 non-converters, validated on 32 converters and 238 non-converters, and tested on 150 converters and 200 non-converters with 5-fold cross-validation. The model’s predictive accuracy (c-index) was 0.82 for training, 0.83 for validation, and 0.78 in the test set. Important features for prediction included age, neurocognitive test scores, education, history of hypertension and stroke, BA, and volumes of the right hippocampus, right precentral gyrus, bilateral white matter, and left inferior temporal gyrus. Products of this model allow us to compare a) the risk, and b) the BA biomarkers across population subgroups with heightened CI risk such as racial minorities and people with fewer years of education. The ability to accurately predict future cognitive impairment (CI) benefits those most in need of therapeutic interventions and reduces the burden of CI on healthcare systems.

A GIS and Gen-AI-Driven framework for automated renewable energy resource assessment and infrastructure optimization

As the globe moves towards renewable energy, maximizing the utilization of renewable resources and enhancing infrastructure management are important issues that conventional approaches find difficult to handle successfully. This paper presents a novel approach that integrates Geographic Information Systems (GIS) with Generative Artificial Intelligence (Gen-AI) to improve an efficiency of renewable energy systems. Moreover, GIS has strong instruments for analyzing issues concerning geography, whereas Gen-AI has enhanced features for energy rate prediction, selection of areas that are suitable for energy production, and control of facilities. These technologies have been proposed to work in synergy to address some of the complex issues like determination of exact locations of the sites, estimating energy generation at any time and controlling real-time energy requirements. Furthermore, it is applied to predict and improve maintenance as well as correct configuration of smart grids to become more effective and eco-friendly energy distributing system. The applications of the framework include many REN sectors, such as solar, wind, hydro, and bioenergy, solving the issues of rearing losses, environmental effects, and varying energy demands. Through solving these urgent problems, integrating GIS and Gen-AI constructs the groundwork for a more sustainable, robust and future energy structure.