Predictive Performance Research Articles

Model selection to achieve reproducible associations between resting state EEG features and autism

A concern in the field of autism electroencephalography (EEG) biomarker discovery is their lack of reproducibility. In the present study, we considered the problem of learning reproducible associations between multiple features of resting state (RS) neural activity and autism, using EEG data collected during a RS paradigm from 36 to 96 month-old children diagnosed with autism (N = 224) and neurotypical children (N = 69). Specifically, EEG spectral power and functional connectivity features were used as inputs to a regularized generalized linear model trained to predict diagnostic group (autism versus neurotypical). To evaluate our model, we proposed a procedure that quantified both the predictive generalization and reproducibility of learned associations produced by the model. When prioritizing both model predictive performance and reproducibility of associations, a highly reproducible profile of associations emerged. This profile revealed a distinct pattern of increased gamma power and connectivity in occipital and posterior midline regions associated with an autism diagnosis. Conversely, model selection based on predictive performance alone resulted in non-robust associations. Finally, we built a custom machine learning model that further empirically improved robustness of learned associations. Our results highlight the need for model selection criteria that maximize the scientific utility provided by reproducibility instead of predictive performance.

Visual field prediction based on temporal-spatial feature learning

Glaucoma stands as the leading irreversible cause of blindness worldwide. Regular visual field examinations play a crucial role in both diagnosing and treating glaucoma. Predicting future visual field changes can assist clinicians in making timely interventions to manage the progression of this disease. To integrate temporal and spatial features from past visual field examination results and enhance visual field prediction, a convolutional long short-term memory (ConvLSTM) network was employed to construct a predictive model. The predictive performance of the ConvLSTM model was validated and compared with other methods using a dataset of perimetry tests from the Humphrey field analyzer at the University of Washington (UWHVF). Compared to traditional methods, the ConvLSTM model demonstrated higher prediction accuracy. Additionally, the relationship between visual field series length and prediction performance was investigated. In predicting the visual field using the previous three visual field results of past 1.5~6.0 years, it was found that the ConvLSTM model performed better, achieving a mean absolute error of 2.255 dB, a root mean squared error of 3.457 dB, and a coefficient of determination of 0.960. The experimental results show that the proposed method effectively utilizes existing visual field examination results to achieve more accurate visual field prediction for the next 0.5~2.0 years. This approach holds promise in assisting clinicians in diagnosing and treating visual field progression in glaucoma patients.

Towards a Better Uncertainty Quantification in Automated Valuation Models

AbstractThis study introduces a novel framework for quantifying prediction uncertainty in automated valuation models (AVMs), crucial tools in modern real estate finance. While non-linear AVMs excel in predictive performance, their limited methods for assessing prediction uncertainty reduces reliability and practical utility. We address this gap by proposing an approach for quantifying the uncertainty associated with predicted house prices and by introducing a model-specific AVM uncertainty estimate (AVMU) for AVM comparisons. Using a dataset of 51,747 historical apartment transactions in Oslo, Norway, we train three AVMs (XGBoost, random forest, support vector machine) to predict sales prices. Thereafter, we develop three base uncertainty estimators (direct loss estimation, bootstrap ensemble, quantile regression) and three meta estimators (average regressor, voting regressor, stacked generalization) for uncertainty quantification. Conformal calibration aligns the outputted uncertainty estimates from the six estimators with standard deviations of corresponding prediction errors. Having strong positive correlations with observed absolute prediction errors, the calibrated uncertainty estimators are shown to effectively capture prediction uncertainty. While the direct loss estimation excels among base estimators, the voting regressor and stacked generalization meta estimators consistently outperform it. Furthermore, by using the AVMU estimate from the stacked generalization meta estimator we can successfully identify the best-performing AVM for three separate apartment portfolios without knowing true sales prices. This alignment of the mean estimated prediction uncertainty with observed deviations underlines the utility of pre-factual AVMU estimates for model comparisons. In conclusion, our framework helps bridge prediction accuracy and uncertainty for AVMs, enhancing their reliability and supporting informed decision making for stakeholders.

Proteomic and serological markers for diagnosing cardia gastric cancer and precursor lesions in a Chinese population

Cardia gastric cancer (CGC) is prevalent in East Asia, and noninvasive, cost-effective screening methods are needed. This study investigated the diagnostic value of serum pepsinogen (PG), gastrin-17 (G-17), Helicobacter pylori (H. pylori) antibodies, and proteomic profiling for CGC and precancerous lesions. We conducted a case-control study involving biopsy-confirmed patients with CGC (n = 60), low-grade intraepithelial neoplasia (CLGD, n = 60), high-grade intraepithelial neoplasia (CHGD, n = 64), and healthy controls (n = 120) matched for age and sex from high-incidence areas in China. Serological markers including PGI, PGII, G-17, and H. pylori were measured using ELISA and Western blot, while plasma protein markers were assessed using Olink® technology. The VSOLassoBag algorithm and nine machine learning (ML) algorithms were employed to identify crucial features and construct predictive models. Various evaluation metrics, including the area under the receiver-operating-characteristic curve (AUC), were utilized to compare predictive performance. Elevated PGII levels, decreased PGR, and H. pylori infection were significantly associated with an increased risk of CGC and precancerous lesions (P for trend < 0.05). The eXtreme Gradient Boosting (XGBoost) model performed best in discriminative ability among the 9 ML models. Following feature reduction based on predictive performance, a final explainable XGBoost model was developed, incorporating five protein biomarkers (CDHR2, ICAM4, PTPRM, CDC27, and FLT1). This model exhibited excellent performance in distinguishing individuals with CGC and precancerous lesions from healthy controls (AUC = 0.931 for CGC, 0.867 for CHGD, and 0.763 for CLGD), surpassing the traditional serological marker-based model. This study underscores the diagnostic potential of serological markers and proteomic profiling in the detection of CGC. Further validation and exploration of combined biomarker approaches are warranted to enhance early diagnosis and improve outcomes in high-risk populations.

Construction of a prediction model for induction of labor based on a small sample of clinical indicator data

Because of the diversity and complexity of clinical indicators, it is difficult to establish a comprehensive and reliable prediction model for induction of labor (IOL) outcomes with existing methods. This study aims to analyze the clinical indicators related to IOL and to develop and evaluate a prediction model based on a small-sample of data. The study population consisted of a total of 90 pregnant women who underwent IOL between February 2023 and January 2024 at the Shanghai First Maternity and Infant Healthcare Hospital, and a total of 52 clinical indicators were recorded. Maximal information coefficient (MIC) was used to select features for clinical indicators to reduce the risk of overfitting caused by high-dimensional features. Then, based on the features selected by MIC, the support vector machine (SVM) model based on small samples was compared and analyzed with the fully connected neural network (FCNN) model based on large samples in deep learning, and the receiver operating characteristic (ROC) curve was given. By calculating the MIC score, the final feature dimension was reduced from 55 to 15, and the area under curve (AUC) of the SVM model was improved from 0.872 before feature selection to 0.923. Model comparison results showed that SVM had better prediction performance than FCNN. This study demonstrates that SVM successfully predicted IOL outcomes, and the MIC feature selection effectively improves the model's generalization ability, making the prediction results more stable. This study provides a reliable method for predicting the outcome of induced labor with potential clinical applications.

Machine Learning Approach for Rock Mass Classification with Imbalanced Database of TBM Tunnelling in Himalayan Geology

AbstractThe geological condition of the Himalayan region is very complex and challenging. So far, empirical and analytical approaches for rock mass characterization have been a common practice in the Himalayas. Due to the limitations of input parameters and governing equations in design practices, rock mass characterization in tunnel boring machine (TBM) excavated tunnels is crucial. This research introduces robust machine learning (ML) approaches to predict rock mass quality conditions in complex geological environments, leveraging a large database of TBM parameters and rock mass rating (RMR) values. To do so, a total of 6879 stable phase TBM cycle data were collected from 12 km long tunnel in Nepal. The pre-processed parameters were randomly split into a training set (80%) and a testing set (20%). Seven individual classifiers consisting of logistic regression (LR), support vector machine (SVM), decision tree (DT), random forest (RF), k-nearest neighbor (KNN), extreme gradient boosting (XGBoost), and bagging, and stacking ensemble classifier were exploited with optimal hyperparameters. The comprehensive assessment carried out has shown that the ensemble classifier gave highest overall accuracy as compared to other individual classifiers. More importantly, the synthetic minority over-sampling technique (SMOTE) performs better to handle the imbalanced database, while the RF and stacking classifier demonstrated the best prediction performance with accuracy of 92%. Moreover, for the minority rock mass class, the RF shows better performance compared to stacking classifier. The authors emphasize that the effective application of ML-based data-driven approach shows substantial potential for rock mass characterization in TBM tunnelling.

Preoperative prediction of early recurrence in patients with BRAF mutant colorectal cancer using a intergrated nomogram

Objective To explore the predictive value of radiomics nomogram combining with CT radiomics features and clinical features for postoperative early recurrence in patients with BRAF-mutant colorectal cancer. Methods A total of 220 patients with surgically and pathologically confirmed BRAF-mutant colorectal cancer from 2 institutions were retrospectively included. All patients from institution 1 were randomized at a 7:3 ratio into a training cohort (n = 108) and an internal validation cohort (n = 45), and patients from institution 2 were used as an external validation cohort (n = 67). The association between the radiomics features and early recurrence was assessed in the training cohort and verified in the validation cohort. Furthermore, the performance of the radiomics nomogram was evaluated by combining the rad-score and clinical risk factors. The predictive performance was evaluated by receiver operating characteristic curve analysis, calibration curve analysis, and decision curve analysis. Results The dierenees in the Lymphocyte/monocyte ratio (LMR) and Peripheral nerve infiltration (PNI) were statistically significant between the early recurrence in BRAF-mutant colorectal cancer groups and the early non-recurrence in BRAF-mutant colorectal cancer groups (P < 0.05); The two groups showed no significant differenee in clinical parameters including age, sex, and biochemistry serum markers (P > 0.05). Comparing with the pure radiomics or clinical data, combined models can be seen that the addition of LMR and PNI further improveed the predictive efficiency of the model. The rad-score based on LR, generated by 4 selected radiomics features, demonstrated a favorable ability to predict early recurrence in both the training (AUC 0.81), internal validation (AUC 0.73), external validation (AUC 0.63) cohorts. Subsequently, integrating two independent predictors into a nomogram exhibited more favorable discriminatory performance, with the AUC improved to 0.88 and 0.81 in both cohorts. Conclusions The proposed CT-based radiomics signature is associated with early recurrence among the patients with BRAF-mutant colorectal cancer. The present study also proposes a combined model can potentially be applied in the individual preoperative prediction of early recurrence in patients with BRAF-mutant colorectal cancer. Advances in knowledge CT-based radiomics showed satisfactory diagnostic significance for early recurrence in patients with BRAF-mutant colorectal cancer. Key baseline clinical characteristics were associated with early recurrence in patients with BRAF-mutant colorectal cancer. The combined model may be applied in the individual preoperative prediction of early recurrence in patients with BRAF-mutant colorectal cancer.

Prediction of optical chaos using a multi-stage extreme learning machine with data uncertainty

In this paper, we study the problem of predicting optical chaos for semiconductor lasers, where data uncertainty can severely degrade the performance of chaos prediction. We hereby propose a multi-stage extreme learning machine (MSELM) based approach for the continuous prediction of optical chaos, which handles data uncertainty effectively. Rather than relying on pilot signals for conventional reservoir learning, the proposed approach enables the use of predicted optical intensity as virtual training samples for the MSELM model learning, which leads to enhanced prediction performance and low overhead. To address the data uncertainty in virtual training, total least square (TLS) is employed for the update of the proposed MSELM’s parameters with simple updating rule and low complexity. Simulation results demonstrate that the proposed MSELM can execute the continuous optical chaos predictions effectively. The chaotic time series can be continuously predicted for a time period in excess of 4 ns with a normalized mean squared error (NMSE) lower than 0.012. It also demands much fewer training samples than state-of-the-art learning-based methods. In addition, the simulation results show that with the help of TLS, the length of prediction is improved significantly as the uncertainty is handled properly. Finally, we verify the prediction ability of the multi-stage ELM under various laser parameters, and make the median boxplot of the predicted results, which shows that the proposed MSELM continues to produce accurate and continuous predictions on time-varying optical chaos.

Early prediction of sepsis in emergency department patients using various methods and scoring systems.

Early recognition of sepsis, a common life-threatening condition in intensive care units (ICUs), is beneficial for improving patient outcomes. However, most sepsis prediction models were trained and assessed in the ICU, which might not apply to emergency department (ED) settings. To establish an early predictive model based on basic but essential information collected upon ED presentation for the follow-up diagnosis of sepsis observed in the ICU. This study developed and validated a reliable model of sepsis prediction among ED patients by comparing 10 different methods based on retrospective electronic health record data from the MIMIC-IV database. In-ICU sepsis was identified as the primary outcome. The potential predictors encompassed baseline demographics, vital signs, pain scale, chief complaints and Emergency Severity Index (ESI). 80% and 20% of the total of 425 737 ED visit records were randomly selected for the train set and the test set for model development and validation, respectively. Among the methods evaluated, XGBoost demonstrated an optimal predictive performance with an area under the curve (AUC) of 0.90 (95% CI: 0.90-0.91). Logistic regression exhibited a comparable predictive ability to XGBoost, with an AUC of 0.89 (95% CI: 0.89-0.90), along with a sensitivity and specificity of 85% (95% CI: 0.83-0.86) and 78% (95% CI: 0.77-0.80), respectively. Neither of the five commonly used severity scoring systems demonstrated satisfactory performance for sepsis prediction. The predictive ability of using ESI as the sole predictor (AUC: 0.79, 95% CI: 0.78-0.80) was also inferior to the model integrating ESI and other basic information. The use of ESI combined with basic clinical information upon ED presentation accurately predicted sepsis among ED patients, strengthening its application in ED. The proposed model may assist nurses in risk stratification management and prioritize interventions for potential sepsis patients, even in low-resource settings.

Short-Term Prediction Method for Gas Concentration in Poultry Houses Under Different Feeding Patterns

Ammonia (NH3) and carbon dioxide (CO2) are the main gases that affect indoor air quality and the health of the chicken flock. Currently, the environmental control strategy for poultry houses mainly relies on real-time temperature, resulting in lag and singleness. Indoor air quality can be improved by predicting the change in CO2 concentration and proposing an optimal control strategy. Combining the advantages of seasonal-trend decomposition using loess (STL), Granger causality (GC), long short-term memory (LSTM), and extreme gradient boosting (XGBoost), an ensemble method called the STL-GC-LSTM-XGBoost model is proposed. This model can set fast response prediction results at a lower cost and has strong generalization ability. The comparative analysis shows that the proposed STL-GC-LSTM-XGBoost model achieved high prediction accuracy, performance, and confidence in predicting CO2 levels under different environmental regulation modes and data volumes. However, its prediction accuracy for NH3 was slightly lower than that of the STL-GC-LSTM model. This may be due to the limited variability and regularity of the NH3 dataset, which likely increased model complexity and decreased predictive ability with the introduction of XGBoost. Nevertheless, in general, the proposed integrated model still provides a feasible approach for gas concentration prediction and health-related risk control in poultry houses.

Fuzzy-based ensemble methodology for accurate long-term prediction and interpretation of extreme significant wave height events

Providing an accurate prediction of Significant Wave Height (SWH), and specially of extreme SWH events, is crucial for coastal engineering activities and holds major implications in several sectors as offshore renewable energy. With the aim of overcoming the challenge of skewness and imbalance associated with the prediction of these extreme SWH events, a fuzzy-based cascade ensemble of regression models is proposed. This methodology allows to remarkably improve the predictive performance on the extreme SWH values, by using different models specialised in different ranges on the target domain. The method’s explainability is enhanced by analysing the contribution of each model, aiding in identifying those predictor variables more characteristic for the detection of extreme SWH events. The methodology has been validated tackling a long-term SWH prediction problem, considering two case studies over the southwest coast of the United States of America. Both reanalysis data, providing information on various meteorological factors, and SWH measurements, obtained from the nearby stations and the station under examination, have been considered. The goodness of the proposed approach has been validated by comparing its performance against several machine learning and deep learning regression techniques, leading to the conclusion that fuzzy ensemble models perform much better in the prediction of extreme events, at the cost of a slight deterioration in the rest of the samples. The study contributes to advancing the SWH prediction field, specially, to understanding the behaviour behind extreme SWH events, critical for various sectors reliant on oceanic conditions.

Terahertz Non-Destructive Testing of Porosity in Multi-Layer Thermal Barrier Coatings Based on Small-Sample Data

Accurately characterizing the internal porosity rate of thermal barrier coatings (TBCs) was essential for prolonging their service life. This work concentrated on atmospheric plasma spray (APS)-prepared TBCs and proposed the utilization of terahertz non-destructive detection technology to evaluate their internal porosity rate. The internal porosity rates were ascertained through a metallographic analysis and scanning electron microscopy (SEM), followed by the reconstruction of the TBC model using a four-parameter method. Terahertz time-domain simulation data corresponding to various porosity rates were generated employing the time-domain finite difference method. In simulating actual test signals, white noise with a signal-to-noise ratio of 10 dB was introduced, and various wavelet transforms were utilized for denoising purposes. The effectiveness of different signal processing techniques in mitigating noise was compared to extract key features associated with porosity. To address dimensionality challenges and further enhance model performance, kernel principal component analysis (kPCA) was employed for data processing. To tackle issues related to limited sample sizes, this work proposed to use the Siamese neural network (SNN) and generative adversarial network (GAN) algorithms to solve this challenge in order to improve the generalization ability and detection accuracy of the model. The efficacy of the constructed model was assessed using multiple evaluation metrics; the results indicate that the novel hybrid WT-kPCA-GAN model achieves a prediction accuracy exceeding 0.9 while demonstrating lower error rates and superior predictive performance overall. Ultimately, this work presented an innovative, convenient, non-destructive online approach that was safe and highly precise for measuring the porosity rate of TBCs, particularly in scenarios involving small sample sizes facilitating assessments regarding their service life.

Estimation of CO 2 saturation maps from synthetic seismic data using a deep-learning method with a multi-scale approach

In the context of Carbon Capture and Storage (CCS), monitoring the behaviour of CO 2 within subsurface reservoirs is pivotal for efficient and safe storage operations. This study proposes a new approach to enhance the accuracy of predicting CO 2 saturation maps directly from shot gathers, using a combination of deep learning (DL) and feature extraction methods. We employ a U-Net model, specifically tailored to solve regression tasks, and two-dimensional continuous wavelet transform (2D-CWT) to analyse shot gathers at different scales. We introduce a novel approach using multi-channel input data for the DL model, combining shot gathers and CWT images. The DL model was trained and tested using both single channel and multichannel input. The single channel datasets included shot gathers and CWT images at the first scale, while the multi-channel datasets integrated shot gathers with either two or four scales of CWT. We conducted a sensitivity analysis on the number of training epochs to compare the performance of the model with multichannel and single channel input. Additionally, a Transfer Learning approach was implemented to improve performance in noisy conditions by leveraging knowledge from a pre-trained model on noise-free images. Monte Carlo dropout was used to predict pixel-scale variability and provided a prediction variability with a standard deviation ranging from 0.019 to 0.194, enhancing the robustness of CO 2 saturation estimates. Our results suggest that combining feature extraction using 2D-CWT with the U-Net architecture improves the prediction performance of CO 2 saturation in a synthetic CCS model. Additionally, using multi-channel input instead of single channel is more efficient as it requires less training and prediction time to achieve comparable results. This confirms the effectiveness of 2D-CWT as a pre-processing technique, contributing valuable insights and advances in the field of data-driven geophysical inversion problems.

A Deep Learning Approach for Predicting Aerial Suppressant Drops in Wildland Firefighting Using Automatic Dependent Surveillance–Broadcast Data

This study utilizes Automatic Dependent Surveillance–Broadcast (ADS-B) data sourced by the OpenSky Network to curate a dataset aimed at enhancing the precision of aerial suppressant drop predictions in wildland firefighting. By amalgamating ADS-B data with Automated Telemetry Unit (ATU) drop information, this research constructs a reliable base for analyzing the spatial aspects of aerial firefighting operations. Using sequential machine learning models, specifically Long Short-Term Memory (LSTM) networks and 1D Convolutional Neural Networks (1DCNN), the study interprets complex flight dynamics to predict drop locations. The dataset, covering 2017 to 2023, is labeled and segmented to reflect accurate suppressant release events, facilitating the distinction between drop and non-drop activities in fixed-wing aircraft. The LSTM model demonstrated strong predictive performance with an F1 score of 0.922, effectively identifying suppressant drop events with high accuracy. This model’s reliable predictions can significantly improve situational awareness in real-time aerial firefighting operations, enabling more informed decision-making and better coordination of resources during wildfire events.

ACCI-HBV-ACLF: A Novel Predictive Model for Hepatitis B Virus-Related Acute-On-Chronic Liver Failure.

Early identification of hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF) holds crucial importance in guiding clinical management and reducing mortality. However, existing scoring systems often overlook patient's underlying clinical condition, which significantly impacts prognosis. Use the age-adjusted Charlson comorbidity index (aCCI) to evaluate the patient's complications to develop a more precise model for predicting transplant-free mortality in HBV-ACLF patients. Nine hundred and six patients were included for investigation and were segregated into a training cohort and a temporal validation cohort according to the chronological order of admission in a ratio of 7:3. In the training cohort, univariate analysis, logistic regression analysis and LASSO regression analysis were used to construct a prognostic model and it was subsequently validated in a temporal validation cohort and an external validation cohort. We found total bilirubin, neutrophils, international normalised ratio and aCCI exhibited significant associations with 28-day transplant-free mortality and established a novel prognostic model, named aCCI-HBV-ACLF. The model demonstrated strong predictive performance, with area under the receiver operating characteristic curve (ROC) values of 0.859 for 28-day mortality, 0.822 for 90-day mortality. In the temporal validation cohort, aCCI-HBV-ACLF achieved area under the ROC values of 0.869 for 28-day mortality and 0.850 for 90-day mortality. In the external validation cohort, aCCI-HBV-ACLF had area under the ROC values of 0.868 for 28-day mortality and 0.888 for 90-day mortality. This study proposes a new prognostic model, which achieved excellent predictive ability for 28-/90-day transplant-free mortality rates among patients with HBV-ACLF.

Analyzing the posterior predictive capability and usability of landslide susceptibility maps: a case of Kerala, India

AbstractLandslide susceptibility maps serve as the basis for hazard and risk assessment, as well as risk-informed land use planning at various spatial scales. Researchers create these maps aiming to fulfil a variety of purposes, including infrastructure planning and restrictive land use zoning. These applications require accurate and specific information to fulfil these purposes, as decisions based on these maps have the potential to cost lives and cause infrastructure damage. The usability of the maps depends on whether they provide the required information and their accuracy to be utilized for the intended purpose. Therefore, assessing the usability and predictive accuracy of landslide susceptibility maps is of paramount importance. Typically, the accuracy of the maps is evaluated using the same landslide inventory that was used to create the map, which does not actually test the predictive ability of the maps in future situations. In this study, we briefly reviewed the purposes of the map creation using literature and stakeholder interviews and assessed the accuracy of three landslide susceptibility maps in a posterior manner. We generated a multi-temporal landslide event inventory after the creation dates of these landslide susceptibility maps. We devised a method to evaluate classified maps by making use of Unique Conditions Units (UCUs) to compare the posteriorly predicted susceptibility classes and the new landslide occurrences. Interviews with stakeholders revealed a disconnection between the aims set forth by map producers and the specific needs of the end users. Our posterior assessment shows that overall predictions of the maps provide plausible results; however, their interpretations for different use cases make them less likely to be used. When comparing the maps using UCUs, landslide densities overlap between the different susceptibility classes, indicating low predictive performance of the maps. Direct comparison of all maps shows a low agreement between susceptibility classes, which pinpoints the uncertainties in data and methods used to create different maps. This study highlights the need for purpose-oriented landslide susceptibility mapping and posterior assessment of the predictive capabilities of these maps aiming to fulfil respective purposes.

Construction of a nomogram for preoperative deep vein thrombosis in pelvic fracture patients

BackgroundIn recent years, the incidence of pelvic fractures has been on the rise, predominantly affecting the elderly population. Deep vein thrombosis may lead to poor prognosis in patients. monocyte-to-lymphocyte ratio is novel biomarkers of inflammation, and this study aims to verify their predictive effect and construct the nomogram model.MethodThis study used binary logistic regression analysis to predict the predictive effect of MLR on the occurrence of DVT in pelvic fractures patients. And use R studio to construct nomogram model.ResultThe results showed that Age (1.04 [1.01, 1.07], p = 0.006), WBC (1.44 [1.28, 1.61], p < 0.001), and MLR (2.11 [1.08, 4.13], p = 0.029) were independent predictive factors. The nomogram demonstrated good predictive performance with small errors in both the training and validation groups, and most clinical patients could benefit from them.ConclusionThe nomogram constructed based on MLR can assist clinicians in early assessment of the probability of DVT occurrence.

Machine Learning Models for Predicting Significant Liver Fibrosis in Patients with Severe Obesity and Nonalcoholic Fatty Liver Disease.

Although noninvasive tests can be used to predict liver fibrosis, their accuracy is limited for patients with severe obesity and nonalcoholic fatty liver disease (NAFLD). We developed machine learning (ML) models to predict significant liver fibrosis in patients with severe obesity through noninvasive tests. This prospective study included 194 patients with severe obesity who underwent wedge liver biopsy and metabolic bariatric surgery at Taipei Medical University Hospital between September 2016 and December 2020. Significant liver fibrosis was defined as a fibrosis score ≥ 2. Patients were randomly divided into a training group (70%) and a validation group (30%). ML models, including support vector machine, random forest, k-nearest neighbor, XGBoost, and logistic regression, were trained to predict significant liver fibrosis, using DM status, AST, ALT, ultrasonographic fibrosis scores, and liver stiffness measurements (LSM). An ensemble model including these ML models was also used for prediction. Among the ML models, the XGBoost model exhibited the highest AUROC of 0.77, with a sensitivity, specificity, and accuracy of 61.5%, 75.8%, and 69.5%, in validation set, while LSM, AST, ALT showed strongest effects on the model. The ensemble model outperformed all ML models in terms of sensitivity, specificity, and accuracy of 73.1%, 90.9%, and 83.1%. For patients with severe obesity and NAFLD, the XGBoost model and the ensemble model exhibit high predictive performance for significant liver fibrosis. These models may be used to screen for significant liver fibrosis in this patient group and monitor treatment response after metabolic bariatric surgery.

A regularized Cox hierarchical model for incorporating annotation information in predictive omic studies

BackgroundAssociated with high-dimensional omics data there are often “meta-features” such as biological pathways and functional annotations, summary statistics from similar studies that can be informative for predicting an outcome of interest. We introduce a regularized hierarchical framework for integrating meta-features, with the goal of improving prediction and feature selection performance with time-to-event outcomes.MethodsA hierarchical framework is deployed to incorporate meta-features. Regularization is applied to the omic features as well as the meta-features so that high-dimensional data can be handled at both levels. The proposed hierarchical Cox model can be efficiently fitted by a combination of iterative reweighted least squares and cyclic coordinate descent.ResultsIn a simulation study we show that when the external meta-features are informative, the regularized hierarchical model can substantially improve prediction performance over standard regularized Cox regression. We illustrate the proposed model with applications to breast cancer and melanoma survival based on gene expression profiles, which show the improvement in prediction performance by applying meta-features, as well as the discovery of important omic feature sets with sparse regularization at meta-feature level.ConclusionsThe proposed hierarchical regularized regression model enables integration of external meta-feature information directly into the modeling process for time-to-event outcomes, improves prediction performance when the external meta-feature data is informative. Importantly, when the external meta-features are uninformative, the prediction performance based on the regularized hierarchical model is on par with standard regularized Cox regression, indicating robustness of the framework. In addition to developing predictive signatures, the model can also be deployed in discovery applications where the main goal is to identify important features associated with the outcome rather than developing a predictive model.

Spatiotemporal risk of human brucellosis under intensification of livestock keeping based on machine learning techniques in Shaanxi, China.

As one of the most neglected zoonotic diseases, brucellosis has posed a serious threat to public health worldwide. This study is purposed to apply different machine learning models to improve the prediction accuracy of human brucellosis (HB) in Shaanxi, China from 2008 to 2020, under livestock husbandry intensification from a spatiotemporal perspective. We quantitatively evaluated the performance and suitability of ConvLSTM, RF, and LSTM models in epidemic forecasting, and investigated the spatial heterogeneity of how different factors drive the occurrence and transmission of HB in distinct sub-regions by using Kernel Density Analysis and Shapley Additional Explanations. Our findings demonstrated that ConvLSTM network yielded the best predictive performance with the lowest average RMSE of 13.875 and MAE values of 18.393. RF model generated an underestimated outcome while LSTM model had an overestimated one. In addition, climatic conditions, intensification of livestock keeping and socioeconomic status were identified as the dominant factors that drive the occurrence of HB in Shaanbei Plateau, Guanzhong Plain, and Shaannan Region, respectively. This work provided a comprehensive understanding of the potential risk of HB epidemics in Northwest China driven by both anthropogenic activities and natural environment, which can support further practice in disease control and prevention.