Sequential Selection Research Articles

Exploring Sequential Feature Selection in Deep Bi-LSTM Models for Speech Emotion Recognition

Exploring Sequential Feature Selection in Deep Bi-LSTM Models for Speech Emotion Recognition

AIoT-based embedded systems optimization using feature selection for Parkinson's disease diagnosis through speech disorders

This study aims to build a pre-diagnosis tool for predicting Parkinson's disease based on a speech disorder which appears as a symptom in approximately 90 % of people with this disease. Recently, some technologies such as AIoT and IoMT aim to integrate Artificial Intelligence and the Internet of Things or Internet of Medical Things to provide an intelligent remote diagnosis for enhancing medical services. Thus, the classification speed and reliability of the systems in these fields are highly recommended. In this work, we compared five ML algorithms (LR, RF, XGB, SVM, KNN) based on their performance, classification speed and reliability. We employed the sequential forward feature selection in order to select the optimal relevant feature for reducing the dimensionality of the used acoustic dataset to enhance both the performance and computation cost for the proposed system. Furthermore, the stratified cross-validation approach has been used to obtain a fair estimation for the proposed system across each point in the dataset. In this paper, we used a vocal dataset of Parkinson's disease consisting of 195 samples and 22 features. We found that 10 features provide the optimal performance. So, we proposed the K-Nearest Neighbours algorithm as a classifier for our system. It reached 98.46 %, 99.33 % and 98.67 % of the accuracy, sensitivity and precision respectively. Moreover, this work provides a detailed explanation of the employed techniques and the obtained results. The novelty of this work, compared to the existing literature, is to enhance both computation cost and performance for building a real-world system to diagnose Parkinson's disease through speech disorder.

Personalized Feature Importance Ranking for Affect Recognition From Behavioral and Physiological Data

Designing affect-based personalized technology involves dealing with large datasets. Machine learning (ML) algorithms are employed to predict affect and similar human factors from in-game metrics, behavioral patterns, or physiological responses. The classification performance is usually presented as a global point estimate without providing user-specific interpretations. This approach is incompatible with effective personalization in games because it disregards the variability of body responses between players. This paper proposes a methodology to classify subjective human factors from large multimodal data. A public VR dataset (CEAP-360VR) was used to extensively compare three ML classifiers and five feature importance techniques. The produced models could reduce the original feature space by 82% (from 113 to 20 features) without compromising predictive performance (F1 score). A random forest (RF) using forward sequential feature selection (fSFS) yielded the best prediction of binary valence (F1=0.761) and arousal (F1=0.748). Finally, feature importance rankings are discussed with emphasis on global and user-specific patterns that may improve affect recognition. The proposed methodology is envisioned to help game designers and researchers create customized user-centric games and VR experiences inferring possible explanations from multimodal datasets.

Early prognostication of overall survival for pediatric diffuse midline gliomas using MRI radiomics and machine learning: A two-center study.

Diffuse midline gliomas (DMG) are aggressive pediatric brain tumors that are diagnosed and monitored through MRI. We developed an automatic pipeline to segment subregions of DMG and select radiomic features that predict patient overall survival (OS). We acquired diagnostic and post-radiation therapy (RT) multisequence MRI (T1, T1ce, T2, and T2 FLAIR) and manual segmentations from 2 centers: 53 from 1 center formed the internal cohort and 16 from the other center formed the external cohort. We pretrained a deep learning model on a public adult brain tumor data set (BraTS 2021), and finetuned it to automatically segment tumor core (TC) and whole tumor (WT) volumes. PyRadiomics and sequential feature selection were used for feature extraction and selection based on the segmented volumes. Two machine learning models were trained on our internal cohort to predict patient 12-month survival from diagnosis. One model used only data obtained at diagnosis prior to any therapy (baseline study) and the other used data at both diagnosis and post-RT (post-RT study). Overall survival prediction accuracy was 77% and 81% for the baseline study, and 85% and 78% for the post-RT study, for internal and external cohorts, respectively. Homogeneous WT intensity in baseline T2 FLAIR and larger post-RT TC/WT volume ratio indicate shorter OS. Machine learning analysis of MRI radiomics has potential to accurately and noninvasively predict which pediatric patients with DMG will survive less than 12 months from the time of diagnosis to provide patient stratification and guide therapy.

Stochastic Sequential Sensory Selection for Gesture Recognition in KineticoMyoGraphy Guided Bionic Hands

Stochastic Sequential Sensory Selection for Gesture Recognition in KineticoMyoGraphy Guided Bionic Hands

Long-Term Fairness in Sequential Multi-Agent Selection With Positive Reinforcement

Long-Term Fairness in Sequential Multi-Agent Selection With Positive Reinforcement

Prediction Model of Ocular Metastases in Gastric Adenocarcinoma: Machine Learning-Based Development and Interpretation Study.

Background: Although gastric adenocarcinoma (GA) related ocular metastasis (OM) is rare, its occurrence indicates a more severe disease. We aimed to utilize machine learning (ML) to analyze the risk factors of GA-related OM and predict its risks. Methods: This is a retrospective cohort study. The clinical data of 3532 GA patients were collected and randomly classified into training and validation sets in a ratio of 7:3. Those with or without OM were classified into OM and non-OM (NOM) groups. Univariate and multivariate logistic regression analyses and least absolute shrinkage and selection operator were conducted. We integrated the variables identified through feature importance ranking and further refined the selection process using forward sequential feature selection based on random forest (RF) algorithm before incorporating them into the ML model. We applied six ML algorithms to construct the predictive GA model. The area under the receiver operating characteristic (ROC) curve indicated the model's predictive ability. Also, we established a network risk calculator based on the best performance model. We used Shapley additive interpretation (SHAP) to identify risk factors and to confirm the interpretability of the black box model. We have de-identified all patient details. Results: The ML model, consisting of 13 variables, achieved an optimal predictive performance using the gradient boosting machine (GBM) model, with an impressive area under the curve (AUC) of 0.997 in the test set. Utilizing the SHAP method, we identified crucial factors for OM in GA patients, including LDL, CA724, CEA, AFP, CA125, Hb, CA153, and Ca2+. Additionally, we validated the model's reliability through an analysis of two patient cases and developed a functional online web prediction calculator based on the GBM model. Conclusion: We used the ML method to establish a risk prediction model for GA-related OM and showed that GBM performed best among the six ML models. The model may identify patients with GA-related OM to provide early and timely treatment.

Heart Disease Diagnostics Using Meta‐Learning‐Based Hybrid Feature Selection

Heart disease, encompassing a range of conditions affecting the heart, remains a leading cause of morbidity and mortality worldwide. The urgent need for precise diagnostic techniques is crucial for improving patient outcomes, as early and accurate diagnosis can significantly influence the effectiveness of treatment and management strategies. This study introduces an innovative approach to diagnosing heart disease by combining classifiers in a meta‐learning‐based approach and utilizing advanced feature selection methods in a hybrid model. Using the extensive heart disease dataset provided by the IEEE DataPort, our method aims to enhance the accuracy of diagnosis by gradually refining the selection of relevant features at two separate stages. Initially, in level 0, a collection of filter‐based algorithms, such as mutual information, relief, ANOVA, systematic uncertainty, and minimum redundancy‐maximum relevance, is utilized to identify a subset of relevant features. Subsequently, the Sequential Forward Selection (SFS) wrapper method is employed in level 1 to further fine‐tune the feature set. The model utilizes a meta‐learning‐based ensemble strategy for classification, combining basic classifiers such as K‐nearest neighbor (KNN), Naïve Bayes (NB), decision tree (DT), and extreme gradient boost (XGBoost) in level 0. In addition, a logistic regression (LR) meta‐classifier is incorporated at level 1. The proposed hybrid technique has been tested on the heart disease dataset, and the results show that the meta‐learning‐based hybrid feature selection approach performs exceptionally well in terms of performance metrics. The model produces impressive results with only eight precisely chosen features, including an accuracy of 96.2185%, precision of 96.0317%, recall of 96.8%, F1‐score of 96.4143%, and an area under the curve (AUC) of 0.9619. Furthermore, our approach significantly outperforms state‐of‐the‐art techniques, indicating its potential to revolutionize heart disease diagnosis and improve patient care.

Estimation of Biochemical Pigment Content in Poplar Leaves Using Proximal Multispectral Imaging and Regression Modeling Combined with Feature Selection.

Monitoring the biochemical pigment contents in individual plants is crucial for assessing their health statuses and physiological states. Fast, low-cost measurements of plants' biochemical traits have become feasible due to advances in multispectral imaging sensors in recent years. This study evaluated the field application of proximal multispectral imaging combined with feature selection and regressive analysis to estimate the biochemical pigment contents of poplar leaves. The combination of 6 spectral bands and 26 vegetation indices (VIs) derived from the multispectral bands was taken as the group of initial variables for regression modeling. Three variable selection algorithms, including the forward selection algorithm with correlation analysis (CORR), recursive feature elimination algorithm (RFE), and sequential forward selection algorithm (SFS), were explored as candidate methods for screening combinations of input variables from the 32 spectral-derived initial variables. Partial least square regression (PLSR) and nonlinear support vector machine regression (SVR) were both applied to estimate total chlorophyll content (Chla+b) and carotenoid content (Car) at the leaf scale. The results show that the nonlinear SVR prediction model based on optimal variable combinations, selected by SFS using multiple scatter correction (MSC) preprocessing data, achieved the best estimation accuracy and stable prediction performance for the leaf pigment content. The Chla+b and Car models developed using the optimal model had R2 and RMSE predictive statistics of 0.849 and 0.825 and 5.116 and 0.869, respectively. This study demonstrates the advantages of using a nonlinear SVR model combined with SFS variable selection to obtain a more reliable estimation model for leaf biochemical pigment content.

A Qualitative Perspective on Anti-Doping: Mixed Methods Research

National and international anti-doping programs are carried out to protect athletes' health, establish an equal and fair competition environment, and defend the basic principles of sports. This study aims to determine the situation regarding the prevention of doping in sports in Türkiye and the recommendations that can be presented to the national sports administration on anti-doping in line with the findings to be obtained. The study was designed with the explanatory sequential design participant selection sub-design, a mixed method research. Three hundred-two people, determined by the purposeful sampling technique, participated in the research. After evaluating the quantitative findings obtained using the questionnaire technique, in-depth interviews were conducted for the qualitative part of the research with 12 people who met the criteria determined among all the participants. Descriptive statistics were used to analyze quantitative data, and theme analysis, descriptive analysis, content analysis, analytical generalization, and discourse analysis were performed in qualitative data analysis. As a result of this research, it is understood that training activities prepared to increase the knowledge and awareness level of sports professionals are considered a more effective method for doping prevention than criminal sanctions. In line with the findings obtained, it is understood that there is a need to establish an independent, multi-faceted anti-doping organization in which all relevant stakeholders are involved and a national action plan in which more importance is attached to guidance and training activities.

Prediction of malaria positivity using patients’ demographic and environmental features and clinical symptoms to complement parasitological confirmation before treatment

BackgroundCurrent malaria diagnosis methods that rely on microscopy and Histidine Rich Protein-2 (HRP2)-based rapid diagnostic tests (RDT) have drawbacks that necessitate the development of improved and complementary malaria diagnostic methods to overcome some or all these limitations. Consequently, the addition of automated detection and classification of malaria using laboratory methods can provide patients with more accurate and faster diagnosis. Therefore, this study used a machine-learning model to predict Plasmodium falciparum (Pf) antigen positivity (presence of malaria) based on sociodemographic behaviour, environment, and clinical features.MethodData from 200 Nigerian patients were used to develop predictive models using nested cross-validation and sequential backward feature selection (SBFS), with 80% of the dataset randomly selected for training and optimisation and the remaining 20% for testing the models. Outcomes were classified as Pf-positive or Pf-negative, corresponding to the presence or absence of malaria, respectively.ResultsAmong the three machine learning models examined, the penalised logistic regression model had the best area under the receiver operating characteristic curve for the training set (AUC = 84%; 95% confidence interval [CI]: 75–93%) and test set (AUC = 83%; 95% CI: 63–100%). Increased odds of malaria were associated with higher body weight (adjusted odds ratio (AOR) = 4.50, 95% CI: 2.27 to 8.01, p < 0.0001). Even though the association between the odds of having malaria and body temperature was not significant, patients with high body temperature had higher odds of testing positive for the Pf antigen than those who did not have high body temperature (AOR = 1.40, 95% CI: 0.99 to 1.91, p = 0.068). In addition, patients who had bushes in their surroundings (AOR = 2.60, 95% CI: 1.30 to 4.66, p = 0.006) or experienced fever (AOR = 2.10, 95% CI: 0.88 to 4.24, p = 0.099), headache (AOR = 2.07; 95% CI: 0.95 to 3.95, p = 0.068), muscle pain (AOR = 1.49; 95% CI: 0.66 to 3.39, p = 0.333), and vomiting (AOR = 2.32; 95% CI: 0.85 to 6.82, p = 0.097) were more likely to experience malaria. In contrast, decreased odds of malaria were associated with age (AOR = 0.62, 95% CI: 0.41 to 0.90, p = 0.012) and BMI (AOR = 0.47, 95% CI: 0.26 to 0.80, p = 0.006).ConclusionNewly developed routinely collected baseline sociodemographic, environmental, and clinical features to predict Pf antigen positivity may be a valuable tool for clinical decision-making.

A Comprehensive Analysis of Provider Fraud Detection through Machine Learning

This research paper presents a comprehensive analysis of healthcare provider fraud detection and analysis using machine learning, drawing insights from diverse literature surveys. The study employs a systematic approach to evaluate methodologies and insights from various academic fields. Leveraging the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, the research synthesizes findings from 27 relevant studies out of 450 articles. The focus lies on characterizing healthcare fraud, emphasizing addressing the limitations and gaps identified in existing literature. The paper introduces a Sequential Forward Selection (SFS) method and SMOTE oversampling for fraud detection, utilizing K-Nearest Neighbors, Artificial Neural Network, Linear Discriminant Analysis, and Gradient Boosting Machine, Classification using a bagging classifier and a stacking meta-estimator. It is recommended to use the Stacking aggregator because it is statistically significant. This study intends to offer insightful information to researchers, legislators, and healthcare practitioners by resolving issues with and gaps in existing methodologies.

Investigating the feasibility of differentiating MS active lesions from inactive ones using texture analysis and machine learning methods in DWI images

BackgroundMagnetic resonance imaging (MRI) is commonly used in conjunction with a gadolinium-based contrast agent (GBCA) to distinguish active multiple sclerosis (MS) lesions. However, recent studies have raised concerns regarding the long-term effects of the accumulation of GBCA in the body. Thus, the purpose of this study is to investigate the possibility of using texture analysis in diffusion-weighted imaging (DWI) and machine learning algorithms to discriminate active from inactive MS lesions without the use of GBCA. MethodsTo achieve this purpose, we introduce an image processing pipeline. In the proposed pipeline, following registration and alignment of slices, MS lesions from DWI images are segmented and quantized. Next, different texture analysis methods are employed to extract texture features from the lesions. Then, a two-stage feature reduction method is applied, in which the first stage involves a statistical t-test and the second stage relies on principal component analysis (PCA), sequential forward selection (SFS), sequential backward selection (SBS), and ReliefF algorithms. Finally, we use five classifiers logistic regression (LR), support vector machine (SVM), decision tree (DT), K nearest neighbor (KNN), and linear discriminant analysis (LDA) in a 5-fold cross-validation procedure to determine active and inactive MS lesions. ResultsIn this study, we collected and prepared 255 active/inactive MS lesions from MRI scans of 34 patients diagnosed with MS, with a mean age of 35.56±10.89. Among 89 texture features extracted, 63 features showed statistically significant differences between the means of active and inactive lesions (P<0.05). The SVM classifier with the PCA feature reduction algorithm demonstrated the best performance with an average accuracy of 0.960 (±0.024), specificity and precision of 1.0, sensitivity of 0.913 (±0.053), and AUC of 0.957 (±0.027). ConclusionOur study indicates that DWI changes detected using texture analysis-based machine learning models can precisely differentiate active from inactive MS lesions. This finding provides valuable clinical information for the early diagnosis and effective monitoring of MS disease.

Consider the time dimension: theorizing and formalizing sequential media selection

Abstract Existing media-selection theories predominantly consider media selection at a static moment-in-time. However, such theorizing is out-of-step with today’s media landscape, which is dominated by sequential media consumption where future media selection is dependent on previously selected media. Ignoring the dependencies among sequential media selection leads to a failure to theorize and model the time-evolving nature of media selection. To bridge this gap, we review computational modeling methods and offer an integrative theoretical framework for studying sequential media selection. In doing so, we lay the theoretical and methodological foundation necessary for state-of-the-art research focused on understanding the underlying mechanisms of, and sequential dependencies among, media selection. Our sequential media-selection framework helps media researchers by theorizing and formalizing processes related to learning, exploration vs. exploitation, and foraging. The outcome is a manuscript that builds on existing theory and research to offer a roadmap for next-generation media-selection inquiry.

Reproducible Python workflow for multi‐site resting‐state EEG analysis: From raw data to group classification

AbstractBackgroundAmong promising markers for neurodegenerative disorders, electroencephalogram (EEG) represents a non‐invasive and potentially portable alternative. Standardized methods for preprocessing may contribute to less inter‐observer variability, while pooling multi‐centric EEG data may provide more generalizable results. We aim to formulate a reproducible pipeline for group‐level classification (of Parkinson’s Disease ‐PD‐) based on various features of resting‐state EEG recordings (rsEEG) collected in multi‐site studies.MethodsWe used four datasets (acquired in Colombia, Finland, and the USA) consisting of 169 subjects (84 PD; 85 non‐PD healthy controls). Our workflow included rsEEG raw files standardization, preprocessing, spectral bandpowers and multiple entropy features extraction on sensor space, harmonization of multi‐site features, and group‐level multiclass classification using supervised machine learning methods. Center/scanner‐related variability effects in the extracted features were harmonized using the ComBat approach and its variants 1. Finally, sequential feature selection (SFS) and binary classification (Non‐PD vs. PD) was conducted using XGradientBoosting 2. The full workflow is detailed in Figure 1.Resultst‐distributed stochastic neighbor embedded (tSNE) plots showed that center‐related variability in spectral features was reduced after harmonization using the ComBat method and its bootstrapped variant, Figure 2. SFS of rsEEG features was performed on 224 harmonized features, showing the highest performance when alpha, pre‐alpha, and slow‐theta bands were considered for subsequent classification. The XGradientBoosting classifier showed an AUC of 0.826 in separating PD from non‐PD, Figure 3.ConclusionsIn multicenter studies, the harmonization of rsEEG features at the sensor space using ComBat can contribute to controlling batch effects. XGradientBoosting with rsEEG features and SFS performed well in classifying non‐PD vs. PD.Intended to be a practical tool for group‐level analysis of neurodegenerative disorders based on sensor space rsEEG features, our workflow is openly available at (https://github.com/alberto‐jj/raw_to_classification) and can be easily adapted to multi‐class classification and regression tasks.

Nonlocal Controls on Tropical Cyclogenesis: A Trajectory-Based Genesis Potential Index

Abstract Tropical cyclone (TC) genesis is initiated by convective precursors or “seeds” and influenced by environmental conditions along the seed-to-TC trajectories. Genesis potential indices (GPIs) provide a simple way to evaluate TC genesis likelihood from environmental conditions but have two limitations that may introduce bias. First, the globally fixed GPIs fail to represent interbasin differences in the relationship between environments and genesis. Second, existing GPIs are only functions of local environmental conditions, whereas nonlocal factors may have a significant impact. We address the first limitation by constructing basin- and time-scale-specific GPIs (local-GPIs) over the eastern North Pacific (ENP) and North Atlantic (NA) using Poisson regression. A sequential feature selection (SFS) algorithm identifies vertical wind shear and a heating condition as leading factors controlling TC genesis in the ENP and the NA, respectively. However, only a slight improvement in performance is achieved, motivating us to tackle the second limitation with a novel trajectory-based GPI (traj-GPI). We merge adjacent nonlocal environments into each grid point based on observed seed trajectory densities. The seed activity, driven mainly by upward motion, and the transition to TCs, controlled primarily by vertical wind shear or heating conditions, are captured simultaneously in the traj-GPI, yielding a better performance than the original GPIs. This study illustrates the importance of seed activity in modeling TC genesis and identifies key environmental factors that influence the process of TC genesis at different stages. Significance Statement The genesis potential index (GPI) is an effective tool for modeling the likelihood of tropical cyclone (TC) genesis for a given time and location. This study reveals that existing GPIs are primarily biased by a lack of information about nonlocal TC seed activity, since they are based only on local large-scale environmental variables. According to our study, upward motion and vertical wind shear are the most influential environmental factors in seed genesis and the transition from seed to TC, respectively. Based on the observed seed trajectories, we build trajectory-based GPIs that include the information from seed activity. Spatiotemporal performances of TC genesis are significantly improved over the original GPIs.

GAUSS-NEWTON METHOD APPLICATION IN THE PROBLEM OF PHASE FUNCTION RECONSTRUCTING FROM HILBERTOGRAMS

The problem of phase function reconstructing in Hilbert diagnostics of gaseous, condensed and reacting media is discussed in the work. The method for reconstructing the phase disturbances structure of a probing light field, based on the iterative Gauss-Newton al- gorithm, is proposed.This method does not require the second derivatives determination and greatly reduces the number of calculations.It consists in the sequential selection of a complex phase profile, which is specified by the sum of third degree Bezier curves, and the hilbertogram calculation in order to minimize the root-mean-square error between the experimental and reconstructed hilbertograms. The Jacobi matrix for the nonlinear integral operator of Hilbert visualization is calculated. The proposed algorithm was tested on test functions.The devel- opment of the method and its applications is associated with the application of the algorithm to the processing of experimental results.

An Assessment of the Mobility of Toxic Elements in Coal Fly Ash Using the Featured BPNN Model

This study aims to propose a novel backpropagation neural network (BPNN) featured with sequential forward selection (SFS), named the BPNN_s model, to master the leaching characteristics of toxic elements (TEs) in coal fly ash (CFA). A total of 400 datasets and 54 features are involved to predict the fractions of TEs. The determination coefficient (R2), root mean square error (RMSE) and variance accounted for (VAF) and Willmott’s index (WI) are used to validate the BPNN_s, and its predictive performance is compared with the other three models, including the unified BPNN (BPNN_u), the adaptive boosting (AdaBoost) and the random forest (RF) models. The results indicate that the BPNN_s outperforms others in predicting the fractions of TEs, and feature selection is an imperative step for developing a model. Moreover, the features selected with SFS suggest that the influence of the element properties is more significant than that of the chemical properties as well as the concentration on predicting the fractions of TEs. Atomic weight is found to be the most critical feature in the prediction through a shapely additive explanations (SHAP) analysis. This study helps to assess the TEs’ mobility rapidly and accurately and provides a foundation for obtaining insights into the relationship between the features and the fractions of TEs.

Palmer amaranth identification using hyperspectral imaging and machine learning technologies in soybean field

Weed infestation and their management are a critical production challenge in agricultural fields. Palmer amaranth has created management challenges because it has multiple emergence pattern and has evolved resistant to nine unique herbicide sites of action. Effective Palmer amaranth detection and positive identification in field conditions will help to improve Palmer amaranth control. A field-based hyperspectral imaging system was developed to record Palmer amaranth in soybean fields. The data were pre-processed applying Savitzky-Golay 2nd derivative, Multiplicative Scatter Correction, and Standard Normal Variate in a forward feed manner. Recursive feature elimination, SelectFromModel, sequential forward selection, and backward elimination were used to select significant wavebands from the available 224 bands. Later, supervised machine-learning models were generated to classify soybean and Palmer amaranth using the selected wavebands. Matthew’s correlation coefficient (MCC), F1 score, precision, and recall were considered as the most significant parameters to evaluate the models’ performance. The highest result was obtained by quadratic discriminant analysis with a prediction accuracy of 93.95%, a precision of 90.30%, a recall of 90.29%, an F1 score of 0.95, and an MCC score of 0.85. The findings of this study showed that the combination of hyperspectral imaging and machine-learning is a potential technique for real-time weed detection in the open field condition.

Machine Learning Insights: Predicting Hepatic Encephalopathy After TIPS Placement.

To develop and assess machine learning (ML) models' ability to predict post-procedural hepatic encephalopathy (HE) following transjugular intrahepatic portosystemic shunt (TIPS) placement. In this retrospective study, 327 patients who underwent TIPS for hepatic cirrhosis between 2005 and 2019 were analyzed. Thirty features (8 clinical, 10 laboratory, 12 procedural) were collected, and HE development regardless of severity was recorded one month follow-up. Univariate statistical analysis was performed with numeric and categoric data, as appropriate. Feature selection is used with a sequential feature selection model with fivefold cross-validation (CV). Three ML models were developed using support vector machine (SVM), logistic regression (LR) and CatBoost, algorithms. Performances were evaluated with nested fivefold-CV technique. Post-procedural HE was observed in 105 (32%) patients. Patients with variceal bleeding (p = 0.008) and high post-porto-systemic pressure gradient (p = 0.004) had a significantly increased likelihood of developing HE. Also, patients having only one indication of bleeding or ascites were significantly unlikely to develop HE as well as Budd-Chiari disease (p = 0.03). The feature selection algorithm selected 7 features. Accuracy ratios for the SVM, LR and CatBoost, models were 74%, 75%, and 73%, with area under the curve (AUC) values of 0.82, 0.83, and 0.83, respectively. ML models can aid identifying patients at risk of developing HE after TIPS placement, providing an additional tool for patient selection and management.