Extra Trees Classifier Research Articles

A class-rebalancing self-training semisupervised learning for imbalanced data lithology identification

Lithologic identification plays a crucial role in petroleum geologic exploration, and machine learning (ML) has become increasingly prevalent in intelligent lithology identification in recent years. However, identifying lithologies presents challenges due to a lack of lithologic labels and an imbalanced distribution of lithologies. To address this issue and obtain satisfactory lithologic identification results, this study investigates a class-rebalancing self-training (CReST) lithology identification framework. This framework uses logging data and limited lithologic labels as input and achieves promising lithology classification through the CReST approach. Four ML algorithms with high overall performance are selected from 25 common algorithms to establish CReST models, such as bagging classifier, extra trees classifier, random forest classifier, and support vector classifier. The classification results of the models are compared and analyzed under three conditions. The experimental findings indicate that (1) under label scarcity, the effect of category recognition varies greatly with different sample numbers; (2) under self-training (ST), overall performance is improved, but the difference in performance caused by category imbalance also increases; and (3) under CReST framework, the model effectively resolves the identification problems caused by a lack of labels and an imbalanced category distribution. Specifically, the precision of identifying categories with fewer samples is improved by more than 20%.

An Optimized Arabic Multilabel Text Classification Approach Using Genetic Algorithm and Ensemble Learning

Multilabel classification of Arabic text is an important task for understanding and analyzing social media content. It can enable the categorization and monitoring of social media posts, the detection of important events, the identification of trending topics, and the gaining of insights into public opinion and sentiment. However, multilabel classification of Arabic contents can present a certain challenge due to the high dimensionality of the representation and the unique characteristics of the Arabic language. In this paper, an effective approach is proposed for Arabic multilabel classification using a metaheuristic Genetic Algorithm (GA) and ensemble learning. The approach explores the effect of Arabic text representation on classification performance using both Bag of Words (BOW) and Term Frequency-Inverse Document Frequency (TF-IDF) methods. Moreover, it compares the performance of ensemble learning methods such as the Extra Trees Classifier (ETC) and Random Forest Classifier (RFC) against a Logistic Regression Classifier (LRC) as a single and ensemble classifier. We evaluate the approach on a new public dataset, namely, the MAWQIF dataset. The MAWQIF is the first multilabel Arabic dataset for target-specific stance detection. The experimental results demonstrate that the proposed approach outperforms the related work on the same dataset, achieving 80.88% for sentiment classification and 68.76% for multilabel tasks in terms of the F1-score metric. In addition, the data augmentation with feature selection improves the F1-score result of the ETC from 65.62% to 68.80%. The study shows the ability of the GA-based feature selection with ensemble learning to improve the classification of multilabel Arabic text.

Body landmarks and genetic algorithm-based approach for non-contact detection of head forward posture among Chinese adolescents: revitalizing machine learning in medicine

Addressing the current complexities, costs, and adherence issues in the detection of forward head posture (FHP), our study conducted an exhaustive epidemiologic investigation, incorporating a comprehensive posture screening process for each participant in China. This research introduces an avant-garde, machine learning-based non-contact method for the accurate discernment of FHP. Our approach elevates detection accuracy by leveraging body landmarks identified from human images, followed by the application of a genetic algorithm for precise feature identification and posture estimation. Observational data corroborates the superior efficacy of the Extra Tree Classifier technique in FHP detection, attaining an accuracy of 82.4%, a specificity of 85.5%, and a positive predictive value of 90.2%. Our model affords a rapid, effective solution for FHP identification, spotlighting the transformative potential of the convergence of feature point recognition and genetic algorithms in non-contact posture detection. The expansive potential and paramount importance of these applications in this niche field are therefore underscored.

Empowering Lower Limb Disorder Identification through PoseNet and Artificial Intelligence.

A novel approach is presented in this study for the classification of lower limb disorders, with a specific emphasis on the knee, hip, and ankle. The research employs gait analysis and the extraction of PoseNet features from video data in order to effectively identify and categorize these disorders. The PoseNet algorithm facilitates the extraction of key body joint movements and positions from videos in a non-invasive and user-friendly manner, thereby offering a comprehensive representation of lower limb movements. The features that are extracted are subsequently standardized and employed as inputs for a range of machine learning algorithms, such as Random Forest, Extra Tree Classifier, Multilayer Perceptron, Artificial Neural Networks, and Convolutional Neural Networks. The models undergo training and testing processes using a dataset consisting of 174 real patients and normal individuals collected at the Tehsil Headquarter Hospital Sadiq Abad. The evaluation of their performance is conducted through the utilization of K-fold cross-validation. The findings exhibit a notable level of accuracy and precision in the classification of various lower limb disorders. Notably, the Artificial Neural Networks model achieves the highest accuracy rate of 98.84%. The proposed methodology exhibits potential in enhancing the diagnosis and treatment planning of lower limb disorders. It presents a non-invasive and efficient method of analyzing gait patterns and identifying particular conditions.

AN ENSEMBLE APPROACH FOR COFFEE CROP YIELD PREDICTION BASED ON AGRONOMIC FACTORS

Coffee is the most burned-through handled drink beside water, which is said to be the most exchanged cultivating product followed by oil in the entire globe. The two most significant sorts of coffee assortment filled in India are Arabica and Robusta out of 103 assortments of class coffee bean variety, which are economically exchanged around the planet. In this regard, we are taking major plantation crop in India i.e., Coffee for our research to explore and develop a predictive model for the development of coffee planters to take precise decisions in time during adverse situations in advance. Hence we propose a framework for coffee yield prediction which using machine learning ensemble approach to estimate the influence of agronomic factors to get a good coffee yield. Here, for our research work, the historic dataset is considered which is obtained from Central Coffee Research Institute (CCRI), Karnataka for the year (2008-2019). For the coffee yield prediction, we are considering agronomic factors like Age, Soil Nutrients: Organic carbon (OC), Phosphorus (P), Potassium (K), Alkaline (pH), Zone and Respective yield obtained in chikkamagaluru region, Karnataka state, India. Different classifiers are used namely, Extra Tree Classifier, Random Forest Classifier, Decision Tree and Boosting Algorithms for prediction and performance of each is compared and analyzed. Our results shown that Extra Tree Classifier and Random forest (RF) classifier with a precision of 91% with good results based on performance metrics considered respectively is an effective and versatile machine-learning method compared to other algorithms used.

A novel approach for explicit song lyrics detection using machine and deep ensemble learning models

The content of music is not always suitable for all ages. Industries that manage music content are looking for ways to help adults determine what is appropriate for children. Lyrics of songs have become increasingly inappropriate for kids and can negatively impact their mental development. However, it is difficult to filter explicit musical content because it is mostly done manually, which is time-consuming and prone to errors. Existing approaches lack the desired accuracy and are complex. This study suggests using a combination of machine learning and deep learning models to automatically screen song lyrics in this regard. The proposed model, called ELSTM-VC, combines extra tree classifier and long short-term memory and its performance is compared to other models. The ELSTM-VC can detect explicit content in English lyrics and can be useful for the music industry. The study used a dataset of 100 songs from Spotify for training, and the results show that the proposed approach effectively detects explicit lyrics. It can censor offensive content for children with a 96% accuracy. The performance of the proposed approach is better than existing approaches including machine learning models and encoding-decoding models.

Machine learning strengthened prediction of tracheal, bronchus, and lung cancer deaths due to air pollution.

This work pointed out the use of machine learning tools to predict the effect of CO, O3, CH4, and CO2 on TBL (tracheal, bronchus, and lung cancer) deaths from 1990 to 2019. In this study, data from 203 countries/locations were used. We used evaluation metrics like accuracy, area under curve (AUC), recall, precision, and Matthews correlation coefficient (MCC) to determine the prediction efficiency of the models. The models that yielded accuracy between 89 and 90 were selected in this study. The essential features in the prediction process were extracted, and it was found that CO influenced the prediction process. Extra trees classifier, random forest classifier, gradient boosting classifier, and light gradient boosting machine were selected from 14 other classifiers based on the accuracy metric. The best-performing models, according to our benchmark standards, are the extra trees classifier (90.83%), random forest classifier (89.17%), gradient boosting classifier (89.17%), and light gradient boosting machine (89.17). We conclude that machine learning models can be used in predicting mortality, i.e., the number of deaths, and could assist us in predicting the role of air pollutants on TBL deaths globally.

Improving the accuracy of epileptogenic zone localization in stereo EEG with machine learning algorithms

The precise identification of the epileptogenic zone (EZ) is paramount in the presurgical evaluation of epilepsy patients to ensure successful surgical outcomes. The analysis of Stereo EEG, an instrumental tool for EZ localization, poses considerable challenges even for experienced epileptologists. Consequently, the development of machine learning (ML)-based computational tools for enhanced EZ localization is imperative. In this investigation, we developed ML models utilizing Stereo EEG from 15 patients, who remained seizure-free (Engel 1 a-d) following EZ resection, over an average follow-up period of 44.4 months. Utilizing Delphos and MNI detectors, spikes and High Frequency Oscillations (HFOs) were identified from Stereo EEG in Resected Zone (RZ) and non-Resected Zone (non-RZ). Linear and non-linear features were estimated from each modality using MATLAB. A total of 27,744 spikes, 7,790 ripples, and 7,632 fast ripples, along with their combinations, were employed to train the ML models. The Gradient Boosting classifier demonstrated the highest prediction accuracy of 98.5% for EZ localization in Mesial Temporal Lobe Epilepsy (MTLE) when trained with features derived from the spike-ripple combination. In the case of Neocortical Epilepsy (NE), the Extra Trees classifier achieved an accuracy of 87.6% when utilizing features from fast ripples. The Random Forest, Extra Trees, and Gradient Boosting algorithms were the most effective for predicting the RZ. Linear features outperformed non-linear features in predicting epileptogenic zones within the epileptic brain. Our study establishes the capability of ML methodologies in localizing epileptogenic zones with high accuracy. Future studies that focus on increasing the training sample size and incorporating more advanced machine learning (ML) algorithms have the potential to significantly improve the accuracy of these models in pinpointing epileptogenic networks. Additionally, implementing this ML approach across multiple research centers would contribute to the broader validation and generalizability of this technique.

Managing Healthcare Infodemic by deep learning in providing healthcare services

Digital Health care data acquisition and processing is performed by Artificial Intelligence and Internet of Things technologies and digitization of data and information affects the patients’ behavior. News about COVID-19, a global pandemic, is circulating on social media worldwide providing a collection of big data. Awareness about the pandemic is spreading drastically in the form of messages, social media posts, tweets, and videos. It is, therefore, significant to assess the early flow of information on social media during the pandemic to prevent alarmism. This study aims to perform sentiment analysis of social media big data about COVID-19 by deep learning on a dataset provided by IEEE Data Port. The goal is to assist healthcare professionals in developing social media policies that can be used to change public opinion. The Dataset used consists of 11,858 COVID-19-related tweets collected on May 30, 2020. Data are labeled as positive or negative in the first step using TextBlob and VADER. In step II, various machine learning models are compared using three feature extraction techniques in combination with VADER and TextBlob. The results show that Extra Tree Classifier using TF-IDF features outperforms with an accuracy of 0.9474.

Revolutionizing Diabetes Disease Prediction Through Novel Machine Learning Techniques

Diabetes is a chronic disease that affects millions of people worldwide. Accurate and timely diagnosis of diabetes is crucial for its effective treatment and management. While machine learning has shown promise in predicting the disease, missing data, outliers, class imbalance and limitations of classifiers can hinder accuracy. To address these challenges, we propose a novel machine learning approach that combines adaptive iterative imputation (AII) for missing value imputation, dynamic ensemble isolation forest (DE-IF) for outlier detection and removal, Iterated KMeans SMOTEENN (IKMSENN) for class imbalance, and an adaptive extra tree classifier (AETC) for classification. Our approach is evaluated using the Pima Indian Diabetes Dataset (PIDD), a widely used benchmark dataset in diabetes disease prediction. Experimental results show that our approach outperforms several state-of-the-art machine learning models in terms of accuracy, precision, recall, [Formula: see text]-measure, and the area under the receiver operating characteristic (ROC) curve (AUC-ROC). Our approach achieved an accuracy of 98.58%, with a precision of 0.986, recall of 0.987, [Formula: see text]-measure of 0.985, and ROC of 0.965 on the PIDD dataset. Our research presents a significant contribution to the field of diabetes disease prediction by introducing novel machine learning approaches that address common challenges such as missing data, outliers and class imbalance, as well as limitations of classifiers. Our approach has the potential to greatly improve the accuracy and effectiveness of diabetes disease prediction and has important implications for the diagnosis and management of the disease.

Prediction of postpartum hemorrhage (PPH) using machine learning algorithms in a Kenyan population.

Postpartum hemorrhage (PPH) is a significant cause of maternal mortality worldwide, particularly in low- and middle-income countries. It is essential to develop effective prediction models to identify women at risk of PPH and implement appropriate interventions to reduce maternal morbidity and mortality. This study aims to predict the occurrence of postpartum hemorrhage using machine learning models based on antenatal, intrapartum, and postnatal visit data obtained from the Kenya Antenatal and Postnatal Care Research Collective cohort. Four machine learning models - logistic regression, naïve Bayes, decision tree, and random forest - were constructed using 67% training data (1,056/1,576). The training data was further split into 67% for model building and 33% cross validation. Once the models are built, the remaining 33% (520/1,576) independent test data was used for external validation to confirm the models' performance. Models were fine-tuned using feature selection through extra tree classifier technique. Model performance was assessed using accuracy, sensitivity, and area under the curve (AUC) of the receiver operating characteristics (ROC) curve. The naïve Bayes model performed best with 0.95 accuracy, 0.97 specificity, and 0.76 AUC. Seven factors (anemia, limited prenatal care, hemoglobin concentrations, signs of pallor at intrapartum, intrapartum systolic blood pressure, intrapartum diastolic blood pressure, and intrapartum respiratory rate) were associated with PPH prediction in Kenyan population. This study demonstrates the potential of machine learning models in predicting PPH in the Kenyan population. Future studies with larger datasets and more PPH cases should be conducted to improve prediction performance of machine learning model. Such prediction algorithms would immensely help to construct a personalized obstetric path for each pregnant patient, improve resource allocation, and reduce maternal mortality and morbidity.

Flood Prediction using MLP, CATBOOST and Extra-Tree Classifier

Flooding can be one of the many devastating natural catastrophes, resulting in the annihilation of life and damaging property. Additionally, it can harm farmland and kill growing crops and trees. Nowadays, rivers and lakes are being destroyed, and the natural water reservoirs are converted into development sites and buildings. Due to this, even just a bit of rain can cause a flood. To minimize the number of fatalities, property losses, and other flood-related issues, an early flood forecast is necessary. Therefore, machine learning methods can be used for the prediction of floods.To forecast the frequency of floods brought on by rainfall, a forecasting system is built using rainfall data. The dataset is trained using various techniques like the MLP classifier, the CatBoost classifier, and the Extra-Tree classifier to predict the occurrence of floods. Finally, the three models' performances are compared and the best model for flood prediction is presented. The MLP, Extra-Tree, and CatBoost models achieved accuracy of 94.5%, 97.9%, and 98.34%, respectively, and it is observed that CatBoost performed well with high accuracy to predict the occurrence of floods.

Deep Learning for Discrimination of Hypertrophic Cardiomyopathy and Hypertensive Heart Disease on MRI Native T1 Maps.

Native T1 and radiomics were used for hypertrophic cardiomyopathy (HCM) and hypertensive heart disease (HHD) differentiation previously. The current problem is that global native T1 remains modest discrimination performance and radiomics requires feature extraction beforehand. Deep learning (DL) is a promising technique in differential diagnosis. However, its feasibility for discriminating HCM and HHD has not been investigated. To examine the feasibility of DL in differentiating HCM and HHD based on T1 images and compare its diagnostic performance with other methods. Retrospective. 128 HCM patients (men, 75; age, 50 years ± 16) and 59 HHD patients (men, 40; age, 45 years ± 17). 3.0T; Balanced steady-state free precession, phase-sensitive inversion recovery (PSIR) and multislice native T1 mapping. Compare HCM and HHD patients baseline data. Myocardial T1 values were extracted from native T1 images. Radiomics was implemented through feature extraction and Extra Trees Classifier. The DL network is ResNet32. Different input including myocardial ring (DL-myo), myocardial ring bounding box (DL-box) and the surrounding tissue without myocardial ring (DL-nomyo) were tested. We evaluate diagnostic performance through AUC of ROC curve. Accuracy, sensitivity, specificity, ROC, and AUC were calculated. Independent t test, Mann-Whitney U-test and Chi-square test were adopted for HCM and HHD comparison. P < 0.05 was considered statistically significant. DL-myo, DL-box, and DL-nomyo models showed an AUC (95% confidential interval) of 0.830 (0.702-0.959), 0.766 (0.617-0.915), 0.795 (0.654-0.936) in the testing set. AUC of native T1 and radiomics were 0.545 (0.352-0.738) and 0.800 (0.655-0.944) in the testing set. The DL method based on T1 mapping seems capable of discriminating HCM and HHD. Considering diagnostic performance, the DL network outperformed the native T1 method. Compared with radiomics, DL won an advantage for its high specificity and automated working mode. 4 TECHNICAL EFFICACY STAGE: 2.

Comparative Analysis of Fall Risk Assessment Features in Community-Elderly and Stroke Survivors: Insights from Sensor-Based Data.

Fall-risk assessment studies generally focus on identifying characteristics that affect postural balance in a specific group of subjects. However, falls affect a multitude of individuals. Among the groups with the most recurrent fallers are the community-dwelling elderly and stroke survivors. Thus, this study focuses on identifying a set of features that can explain fall risk for these two groups of subjects. Sixty-five community dwelling elderly (forty-nine female, sixteen male) and thirty-five stroke-survivors (twenty-two male, thirteen male) participated in our study. With the use of an inertial sensor, some features are extracted from the acceleration data of a Timed Up and Go (TUG) test performed by both groups of individuals. A short-form berg balance scale (SFBBS) score and the TUG test score were used for labeling the data. With the use of a 100-fold cross-validation approach, Relief-F and Extra Trees Classifier algorithms were used to extract sets of the top 5, 10, 15, 20, 25, and 30 features. Random Forest classifiers were trained for each set of features. The best models were selected, and the repeated features for each group of subjects were analyzed and discussed. The results show that only the stand duration was an important feature for the prediction of fall risk across all clinical tests and both groups of individuals.

Evaluation of the Thermogravimetric Profile of Hybrid Cellulose Acetate Membranes using Machine Learning Approaches

Thermogravimetric analysis (TGA) is a characterization technique routinely used in materials science. In this particular case, TGA determines the variation of weight with temperature. The thermogravimetric analysis of cellulose acetate (CA) hybrid membranes can provide similar results, despite their different chemical composition. The present study uses machine learning algorithms to correlate data from thermogravimetric analyses with variations in chemical composition. Experimental points relating to temperature and weight from these analyses were treated in different ways and used to estimate the composition of the membranes. The Extra-Trees Classifier, Random Forest, Decision Tree, and K-Nearest Neighbors (KNN) algorithms were applied to this data and then evaluated using a confusion and accuracy matrix. The decision tree-based algorithms demonstrated a superior capacity for estimating the composition, albeit with negligible disparities in the thermogravimetric profile. The Extra-Trees Classifier algorithm, in particular, stood out for its ability to estimate composition in all tests, achieving 90% accuracy.

A stacking ensemble classifier-based machine learning model for classifying pollution sources on photovoltaic panels

Solar energy is a very efficient alternative for generating clean electric energy. However, pollution on the surface of solar panels reduces solar radiation, increases surface transmittance, and raises the surface temperature. All these factors cause photovoltaic (PV) panels to be less efficient. To address this problem, a stacking ensemble classifier-based machine learning model is proposed. In this study, different sources of pollution on each solar panel are used, and their power generation is recorded. The proposed model includes gradient boost, extra tree, and random forest classifiers, with the extra tree classifier serving as a meta-learner. The model takes into account various weather features during the training process, including irradiance and temperature, aiming to increase its accuracy and robustness in identifying pollution sources on the PV panel. Moreover, the proposed model is evaluated using various methods in order to examine performance metrics such as accuracy, F1 score, and precision. Results show that the model can achieve an accuracy score of 97.37%. The model’s performance is also compared to state-of-the-art machine learning models, demonstrating its superiority in accurately classifying pollution sources on PV panels. By utilizing different sources of pollution and weather features during training, the model can accurately classify different pollution sources, resulting in increased power generation efficiency and the longevity of PV panels. The main results of this study can be used to manage and maintain PV panels since the model can identify PV modules that need to be cleaned to keep producing the most power. Furthermore, the efficiency, reliability, and sustainability of PV panels can be further enhanced by the proposed model.

Text Data Augmentation Techniques for Fake News Detection in the Romanian Language

This paper aims to investigate the use of a Romanian data source, different classifiers, and text data augmentation techniques to implement a fake news detection system. The paper focusses on text data augmentation techniques to improve the efficiency of fake news detection tasks. This study provides two approaches for fake news detection based on content and context features found in the Factual.ro data set. For this purpose, we implemented two data augmentation techniques, Back Translation (BT) and Easy Data Augmentation (EDA), to improve the performance of the models. The results indicate that the implementation of the BT and EDA techniques successfully improved the performance of the classifiers used in our study. The results of our content-based approach show that an Extra Trees Classifier model is the most effective, whether data augmentation is used or not, as it produced the highest accuracy, precision, F1 score, and Kappa. The Random Forest Classifier with BT yielded the best results of the context-based experiment overall, with the highest accuracy, recall, F1 score, and Kappa. Furthermore, we found that BT and EDA led to an increase in the AUC scores of all models in both content-based and context-based data sets.

Predicting Mechanical Thrombectomy Outcome and Time Limit through ADC Value Analysis: A Comprehensive Clinical and Simulation Study Using Machine Learning

Predicting outcomes after mechanical thrombectomy (MT) remains challenging for patients with acute ischemic stroke (AIS). This study aimed to explore the usefulness of machine learning (ML) methods using detailed apparent diffusion coefficient (ADC) analysis to predict patient outcomes and simulate the time limit for MT in AIS. A total of 75 consecutive patients with AIS with complete reperfusion in MT were included; 20% were separated to test data. The threshold ranged from 620 × 10−6 mm2/s to 480 × 10−6 mm2/s with a 20 × 10−6 mm2/s step. The mean, standard deviation, and pixel number of the region of interest were obtained according to the threshold. Simulation data were created by mean measurement value of patients with a modified Rankin score of 3–4. The time limit was simulated from the cross point of the prediction score according to the time to perform reperfusion from imaging. The extra tree classifier accurately predicted the outcome (AUC: 0.833. Accuracy: 0.933). In simulation data, the prediction score to obtain a good outcome decreased according to increasing time to reperfusion, and the time limit was longer among younger patients. ML methods using detailed ADC analysis accurately predicted patient outcomes in AIS and simulated tolerance time for MT.

Enhancing Heart Disease Prediction through Ensemble Learning Techniques with Hyperparameter Optimization

Heart disease is a significant global health issue, contributing to high morbidity and mortality rates. Early and accurate heart disease prediction is crucial for effectively preventing and managing the condition. However, this remains a challenging task to achieve. This study proposes a machine learning model that leverages various preprocessing steps, hyperparameter optimization techniques, and ensemble learning algorithms to predict heart disease. To evaluate the performance of our model, we merged three datasets from Kaggle that have similar features, creating a comprehensive dataset for analysis. By employing the extra tree classifier, normalizing the data, utilizing grid search cross-validation (CV) for hyperparameter optimization, and splitting the dataset with an 80:20 ratio for training and testing, our proposed approach achieved an impressive accuracy of 98.15%. These findings demonstrated the potential of our model for accurately predicting the presence or absence of heart disease. Such accurate predictions could significantly aid in early prevention, detection, and treatment, ultimately reducing the mortality and morbidity associated with heart disease.

Chronic Kidney Disease Prediction System Using Machine Learning

Abstract: Chronic kidney disease (CKD) is a life-threatening condition that can be difficult to diagnose early because there are no symptoms. The purpose of the proposed study is to develop and validate a predictive model for the prediction of chronic kidney disease. Machine learning algorithms are often used in medicine to predict and classify diseases. Medical records are often skewed. Chronic Kidney Disease (CKD) or chronic renal disease has become a significant issue with a steady growth rate. A person can only survive without kidneys for an average of 18 days, which makes a huge demand for a kidney transplant and Dialysis. It is important to have effective methods for the early prediction of CKD. Machine learning methods are effective in CKD prediction. This work proposes a workflow to predict CKD status based on clinical data, incorporating data prepossessing, a missing value handling method with collaborative filtering and attribute selection. Out of the 11 machine learning methods considered, the extra tree classifier and random forest classifier are shown to result in the highest accuracy and minimal bias to the attributes. The project also considers the practical aspects of data collection and highlights the importance of incorporating domain knowledge when using machine learning for CKD status prediction.