Sea ice, as an important component of the Earth\'s ecosystem, has a profound impact on global climate and human activities due to its thickness. Therefore, the inversion of sea ice thickness has important research significance. Due to environmental and equipment-related limitations, the number of samples available for remote sensing inversion is currently insufficient. At high spatial resolutions, remote sensing data contain limited information and noise interference, which seriously affect the accuracy of sea ice thickness inversion. In response to the above issues, we conducted experiments using ice draft data from the Beaufort Sea and designed an improved GBDT method that integrates feature-enhancement and active-learning strategies (IFEAL-GBDT). In this method, the incident angle and time series are used to perform spatiotemporal correction of the data, reducing both temporal and spatial impacts. Meanwhile, based on the original polarization information, effective multi-attribute features are generated to expand the information content and improve the separability of sea ice with different thicknesses. Taking into account the growth cycle and age of sea ice, attributes were added for month and seawater temperature. In addition, we studied an active learning strategy based on the maximum standard deviation to select more informative and representative samples and improve the model\'s generalization ability. The improved GBDT model was used for training and prediction, offering advantages in dealing with nonlinear, high-dimensional data, and data noise problems, further expanding the effectiveness of feature-enhancement and active-learning strategies. Compared with other methods, the method proposed in this paper achieves the best inversion accuracy, with an average absolute error of 8 cm and a root mean square error of 13.7 cm for IFEAL-GBDT and a correlation coefficient of 0.912. This research proves the effectiveness of our method, which is suitable for the high-precision inversion of sea ice thickness determined using Sentinel-1 data.

In order to improve the accuracy of transformer fault diagnosis and improve the influence of unbalanced samples on the low accuracy of model identification caused by insufficient model training, this paper proposes a transformer fault diagnosis method based on SMOTE and NGO-GBDT. Firstly, the Synthetic Minority Over-sampling Technique (SMOTE) was used to expand the minority samples. Secondly, the non-coding ratio method was used to construct multi-dimensional feature parameters, and the Light Gradient Boosting Machine (LightGBM) feature optimization strategy was introduced to screen the optimal feature subset. Finally, Northern Goshawk Optimization (NGO) algorithm was used to optimize the parameters of Gradient Boosting Decision Tree (GBDT), and then the transformer fault diagnosis was realized. The results show that the proposed method can reduce the misjudgment of minority samples. Compared with other integrated models, the proposed method has high fault identification accuracy, low misjudgment rate and stable performance.

School bullying among primary and secondary school students has received increasing attention, and identifying relevant factors is a crucial way to reduce the risk of bullying victimization. Machine learning methods can help researchers predict and identify individual risk behaviors. Through a machine learning approach (i.e., the gradient boosting decision tree model, GBDT), the present longitudinal study aims to systematically examine individual, family, and school environment factors that can predict the risk of bullying victimization among primary and secondary school students a year later. A total of 2767 participants (2065 secondary school students, 702 primary school students, 55.20% female students, mean age at T1 was 12.22) completed measures of 24 predictors at the first wave, including individual factors (e.g., self-control, gender, grade), family factors (family cohesion, parental control, parenting style), peer factor (peer relationship), and school factors (teacher-student relationship, learning capacity). A year later (i.e., T2), they completed the Olweus Bullying Questionnaire. The GBDT model predicted whether primary and secondary school students would be exposed to school bullying after one year by training a series of base learners and outputting the importance ranking of predictors. The GBDT model performed well. The GBDT model yielded the top 6 predictors: teacher-student relationship, peer relationship, family cohesion, negative affect, anxiety, and denying parenting style. The protective factors (i.e., teacher-student relationship, peer relationship, and family cohesion) and risk factors (i.e., negative affect, anxiety, and denying parenting style) associated with the risk of bullying victimization a year later among primary and secondary school students are identified by using a machine learning approach. The GBDT model can be used as a tool to predict the future risk of bullying victimization for children and adolescents and to help improve the effectiveness of school bullying interventions.

Over the past few decades, scholars have employed a wide range of methodologies to determine the factors influencing firms\' voluntary carbon disclosure. Most of these studies have been conducted in advanced markets. This article aims to examine the trend of voluntary carbon disclosure in the Korean financial market by utilizing machine learning models such as Random Forest and Gradient Boosted Decision Tree. Based on a set of hand-collected carbon disclosure data, we initially demonstrated significantly better performance of machine learning models compared to the traditional logistic model. Regarding the factors influencing disclosure, we consistently find the importance of environmental scores, emphasizing the role of the emerging mega-trend of ESG management practices in disclosure decisions. However, in contrast to recent studies, we do not find that the unique Korean governance structure, chaebol, has any significantly different implications in terms of prediction performance and variable importance in carbon disclosure decisions.

BACKGROUND: Lysine glutarylation (Kglu) is one of the most important Post-translational modifications (PTMs), which plays significant roles in various cellular functions, including metabolism, mitochondrial processes, and translation. Therefore, accurate identification of the Kglu site is important for elucidating protein molecular function. Due to the time-consuming and expensive limitations of traditional biological experiments, computational-based Kglu site prediction research is gaining more and more attention.
RESULTS: In this paper, we proposed GBDT_KgluSite, a novel Kglu site prediction model based on GBDT and appropriate feature combinations, which achieved satisfactory performance. Specifically, seven features including sequence-based features, physicochemical property-based features, structural-based features, and evolutionary-derived features were used to characterize proteins. NearMiss-3 and Elastic Net were applied to address data imbalance and feature redundancy issues, respectively. The experimental results show that GBDT_KgluSite has good robustness and generalization ability, with accuracy and AUC values of 93.73%, and 98.14% on five-fold cross-validation as well as 90.11%, and 96.75% on the independent test dataset, respectively.
CONCLUSIONS: GBDT_KgluSite is an effective computational method for identifying Kglu sites in protein sequences. It has good stability and generalization ability and could be useful for the identification of new Kglu sites in the future. The relevant code and dataset are available at https://github.com/flyinsky6/GBDT_KgluSite .

Machine learning (ML) techniques improve disease prediction by identifying the most relevant features in multidimensional data. We compared the accuracy of ML algorithms for predicting incident diabetic kidney disease (DKD).
We utilized longitudinal data from 1365 Chinese, Malay, and Indian participants aged 40-80 y with diabetes but free of DKD who participated in the baseline and 6-year follow-up visit of the Singapore Epidemiology of Eye Diseases Study (2004-2017). Incident DKD (11.9%) was defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 with at least 25% decrease in eGFR at follow-up from baseline. A total of 339 features, including participant characteristics, retinal imaging, and genetic and blood metabolites, were used as predictors. Performances of several ML models were compared to each other and to logistic regression (LR) model based on established features of DKD (age, sex, ethnicity, duration of diabetes, systolic blood pressure, HbA1c, and body mass index) using area under the receiver operating characteristic curve (AUC).
ML model Elastic Net (EN) had the best AUC (95% CI) of 0.851 (0.847-0.856), which was 7.0% relatively higher than by LR 0.795 (0.790-0.801). Sensitivity and specificity of EN were 88.2 and 65.9% vs. 73.0 and 72.8% by LR. The top 15 predictors included age, ethnicity, antidiabetic medication, hypertension, diabetic retinopathy, systolic blood pressure, HbA1c, eGFR, and metabolites related to lipids, lipoproteins, fatty acids, and ketone bodies.
Our results showed that ML, together with feature selection, improves prediction accuracy of DKD risk in an asymptomatic stable population and identifies novel risk factors, including metabolites.
This study was supported by the National Medical Research Council, NMRC/OFLCG/001/2017 and NMRC/HCSAINV/MOH-001019-00. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

The length of hospital stay (LOHS) caused by COVID-19 has imposed a financial burden, and cost on the healthcare service system and a high psychological burden on patients and health workers. The purpose of this study is to adopt the Bayesian model averaging (BMA) based on linear regression models and to determine the predictors of the LOHS of COVID-19.
In this historical cohort study, from 5100 COVID-19 patients who had registered in the hospital database, 4996 patients were eligible to enter the study. The data included demographic, clinical, biomarkers, and LOHS. Factors affecting the LOHS were fitted in six models, including the stepwise method, AIC, BIC in classical linear regression models, two BMA using Occam\'s Window and Markov Chain Monte Carlo (MCMC) methods, and GBDT algorithm, a new method of machine learning.
The average length of hospitalization was 6.7 ± 5.7 days. In fitting classical linear models, both stepwise and AIC methods (R 2 = 0.168 and adjusted R 2 = 0.165) performed better than BIC (R 2 = 0.160 and adjusted = 0.158). In fitting the BMA, Occam\'s Window model has performed better than MCMC with R 2 = 0.174. The GBDT method with the value of R 2 = 0.64, has performed worse than the BMA in the testing dataset but not in the training dataset. Based on the six fitted models, hospitalized in ICU, respiratory distress, age, diabetes, CRP, PO2, WBC, AST, BUN, and NLR were associated significantly with predicting LOHS of COVID-19.
The BMA with Occam\'s Window method has a better fit and better performance in predicting affecting factors on the LOHS in the testing dataset than other models.

UNASSIGNED: Cognitive impairment (CI) is a common disorder in patients with epilepsy (PWEs). Objective assessment method for diagnosing CI in PWEs would be beneficial in reality. This study proposed to construct a diagnostic model for CI in PWEs using the clinical and the phase locking value (PLV) functional connectivity features of the electroencephalogram (EEG).
UNASSIGNED: PWEs who met the inclusion and exclusion criteria were divided into a cognitively normal (CON) group (n = 55) and a CI group (n = 76). The 23 clinical features and 684 PLV EEG features at the time of patient visit were screened and ranked using the Fisher score. Adaptive Boosting (AdaBoost) and Gradient Boosting Decision Tree (GBDT) were used as algorithms to construct diagnostic models of CI in PWEs either with pure clinical features, pure PLV EEG features, or combined clinical and PLV EEG features. The performance of these models was assessed using a five-fold cross-validation method.
UNASSIGNED: GBDT-built model with combined clinical and PLV EEG features performed the best with accuracy, precision, recall, F1-score, and an area under the curve (AUC) of 90.11, 93.40, 89.50, 91.39, and 0.95%. The top 5 features found to influence the model performance based on the Fisher scores were the magnetic resonance imaging (MRI) findings of the head for abnormalities, educational attainment, PLV EEG in the beta (β)-band C3-F4, seizure frequency, and PLV EEG in theta (θ)-band Fp1-Fz. A total of 12 of the top 5% of features exhibited statistically different PLV EEG features, while eight of which were PLV EEG features in the θ band.
UNASSIGNED: The model constructed from the combined clinical and PLV EEG features could effectively identify CI in PWEs and possess the potential as a useful objective evaluation method. The PLV EEG in the θ band could be a potential biomarker for the complementary diagnosis of CI comorbid with epilepsy.

The rapid growth of medical data has greatly promoted the wide exploitation of machine learning for paramedical diagnosis. Inversely proportional to their performance, most machine learning models generally suffer from the lack of explainability, especially the local explainability of the model, that is, the case-specific explainability.
In this paper, we proposed a GBDT (Gradient Boosting Decision Tree)-based explainable model for case-specific paramedical diagnostics, and mainly make the following contributions: (1) an adaptive gradient boosting decision tree (AdaGBDT) model is proposed to boost the path-mining for decision effectively; (2) to learn a case-specific feature importance embedding for a specific patient, the bi-side mutual information is applied to characterize the backtracking on the decision path; (3) through the collaborative decision-making by globally explainable AdaGBDT with case-based reasoning (CBR) in the case-specific metric space, some hard cases can be identified by the means of visualized interpretation. The performance of our model is evaluated on the Wisconsin diagnostic breast cancer dataset and the UCI heart disease dataset.
Experiments conducted on two datasets show that our AdaGBDT achieves the best performance, with the F1-value of 0.9647 and 0.8405 respectively. Moreover, a series of experimental analyses and case studies further illustrate the excellent performance of feature importance embedding.
The proposed case-specific explainable paramedical diagnosis via AdaGBDT has excellent predictive performance, with both promising case-level and consistent global explainability.

Epilepsy is a common neurological disorder with sudden and recurrent seizures. Early prediction of seizures and effective intervention can significantly reduce the harm suffered by patients. In this paper, a method based on nonlinear features of EEG signal and gradient boosting decision tree (GBDT) is proposed for early prediction of epilepsy seizures. First, the EEG signals were divided into two categories: those that had seizures onset over a period of time (represented by InT) and those that did not. Second, the noise in the EEG was removed using complementary ensemble empirical mode decomposition (CEEMD) and wavelet threshold denoising. Third, the nonlinear features of the two categories of EEG were extracted, including approximate entropy, sample entropy, permutation entropy, spectral entropy and wavelet entropy. Fourth, a GBDT classifier with random forest as the initial result was designed to distinguish the two categories of EEG. Fifth, a two-step \"k of n\" method was used to reduce the number of false alarms. The proposed method was evaluated on 13 patients\' EEG data from the CHB-MIT Scalp EEG Database. Based on ten-fold cross validation, the average accuracy was 91.76% when the InT was taken at 30 min, and 38 out of 39 seizures were successfully predicted. When the InT was taken for 40 min, the average accuracy was 92.50% and all 42 seizures selected were successfully predicted. The results indicate the effectiveness of the proposed method for predicting epilepsy seizures.