Mutual Information Method Research Articles

Left atrium 4D-flow segmentation with high-resolution contrast-enhanced magnetic resonance angiography.

Atrial fibrillation (AF) leads to intracardiac thrombus and an associated risk of stroke. Phase-contrast cardiovascular magnetic resonance (CMR) with flow-encoding in all three spatial directions (4D-flow) provides a time-resolved 3D volume image with 3D blood velocity, which brings individual hemodynamic information affecting thrombus formation. As the resolution and contrast of 4D-flow are limited, we proposed a semi-automated 4D-flow segmentation method for the left atrium (LA) using a standard-of-care contrast-enhanced magnetic resonance angiography (CE-MRA) and registration technique. LA of 54 patients with AF were segmented from 4D-flow taken in sinus rhythm using two segmentation methods. (1) Phase-contrast magnetic resonance angiography (PC-MRA) was calculated from 4D-flow, and LA was segmented slice-by-slice manually. (2) LA and other structures were segmented from CE-MRA and transformed into 4D-flow coordinates by registration with the mutual information method. Overlap of volume was tested by the Dice similarity coefficient (DSC) and the average symmetric surface distance (ASSD). Mean velocity and stasis were calculated to compare the functional property of LA from two segmentation methods. LA volumes from segmentation on CE-MRA were strongly correlated with PC-MRA volume, although mean CE-MRA volumes were about 10% larger. The proposed registration scheme resulted in visually successful registration in 76% of cases after two rounds of registration. The mean of DSC of the registered cases was 0.770 ± 0.045, and the mean of ASSD was 2.704 mm ± 0.668 mm. Mean velocity had no significant difference between the two segmentation methods, and mean stasis had a 3.3% difference. The proposed CE-MRA segmentation and registration method can generate segmentation for 4D-flow images. This method will facilitate 4D-flow analysis for AF patients by making segmentation easier and overcoming the limit of resolution.

A COMPARATIVE STUDY ON FEATURE SELECTION METHODS IN QSAR

A quantitative structure-activity relationship (QSAR) relates quantitative chemical structure attributes (molecular descriptors) to a biological activity. QSAR studies have now become attractive in drug discovery and development because their application can save substantial time and human resources. Several parameters are important in the prediction ability of a QSAR model. On the one hand, different statistical methods may be applied to check the linear or nonlinear behavior of a data set. On the other hand, feature selection techniques are applied to decrease the model complexity, to decrease the overfitting/overtraining risk, and to select the most important descriptors from the often more than calculated. The selected descriptors are then linked to a biological activity of the corresponding compound by means of a mathematical model. Different modeling techniques can be applied, some of which explicitly require a feature selection. A QSAR model can be useful in the design of new compounds with improved potency in the class under study. Only molecules with a predicted interesting activity will be synthesized. In the feature selection problem, a learning algorithm is faced with the problem of selecting a relevant subset of features upon which to focus attention, while ignoring the rest. This paper studied the comparative analysis of the Chi-square, Mutual Information, Anova F-value, Fisher Score and SHAP feature selection methods used in QSAR modeling. The Python code written to get experimental results in this article has been uploaded to Github.

Approach for Detecting Attacks on IoT Networks Based on Ensemble Feature Selection and Deep Learning Models.

The Internet of Things (IoT) has transformed our interaction with technology and introduced security challenges. The growing number of IoT attacks poses a significant threat to organizations and individuals. This paper proposes an approach for detecting attacks on IoT networks using ensemble feature selection and deep learning models. Ensemble feature selection combines filter techniques such as variance threshold, mutual information, Chi-square, ANOVA, and L1-based methods. By leveraging the strengths of each technique, the ensemble is formed by the union of selected features. However, this union operation may overlook redundancy and irrelevance, potentially leading to a larger feature set. To address this, a wrapper algorithm called Recursive Feature Elimination (RFE) is applied to refine the feature selection. The impact of the selected feature set on the performance of Deep Learning (DL) models (CNN, RNN, GRU, and LSTM) is evaluated using the IoT-Botnet 2020 dataset, considering detection accuracy, precision, recall, F1-measure, and False Positive Rate (FPR). All DL models achieved the highest detection accuracy, precision, recall, and F1 measure values, ranging from 97.05% to 97.87%, 96.99% to 97.95%, 99.80% to 99.95%, and 98.45% to 98.87%, respectively.

Estimation of prediction intervals for uncertainty assessment of artificial neural network based wastewater treatment plant effluent modeling

Uncertainty qualification of estimated biological variables of Tabriz wastewater treatment plant (WWTP) in Iran, was conducted by artificial intelligence-based modeling. The pollution variables at entrance of sedimentation section and its output were considered as inputs and outputs of the modeling. Non-linear mutual information (MI) and linear correlation coefficient (CC) were used to select inputs. Modeling was conducted via two feed-forward neural networks (FFNN) models for point prediction (PP) and lower and upper bound estimation (LUBE) to provide prediction intervals (PIs) of the computed variables as representative of the modeling uncertainty. In the proposed LUBE, an FFNN was trained with two outputs as upper and lower limits of the predictions. The results showed that selecting the inputs through MI led to narrower PI bound and less uncertainty. Computed determination coefficient (DC)train and DCTest for modeling effluent biological oxygen demand (BODeff) were respectively as 0.9 and 0.7, whereas the values for modeling effluent chemical oxygen demand (CODeff) were 0.8 and 0.7. Comparison of non-linear machine learning (ML) models of FFNN, support vector machine (SVR), adaptive neuro-fuzzy inference system (ANFIS), and linear autoregressive integrated moving average (ARIMA) model demonstrated the superiority of FFNN method for PP task over others. In the LUBE method, the best relationship between pollution parameters was obtained via the MI method. PICP and CWC values were obtained for BOD as 94.08 % and 0.152 and for COD as 86.41 % and 0.1769, respectively. The innovation of this research is to examine the uncertainties in neural network-based analysis of WWTP modeling.

Brain Computer Interface Based on Motor Imagery for Mechanical Arm Grasp Control

This paper puts forward a brain computer interface (BCI) system to realize the hand and wrist control using the ABB Mechanical Arm. This BCI system gathers four kinds of motor imaginary (MI) tasks (hand grasp, hand spread, wrist flexion and wrist extension) electroencephalogram (EEG) signals from 30 electrodes. It utilizes two fifth-order Butterworth Band-Pass Filter (BPF) with different bandwidths and normalization method to achieve the raw MI tasks EEG signals preprocessing. The main challenge of feature extraction is to extract enough representative features from MI tasks to classify them. This proposed BCI system extracts eleven kinds of features in time domain and time-frequency domain and uses mutual information method to reduce the large dimension of the extracted features. In addition, the BCI system applies a single convolutional layer Convolutional neural networks (CNN) with 30 filters to implement the quaternary classification of MI tasks. Compared with early researches, the classification accuracy of this BCI system is increased by about 35%. The actual mechanical arm grasping control experiments verifies that this BCI system has good adaptability.

An improvised coral reef optimization-based image registration using modified mutual information method

Computer visualisation and medical applications require image registration. It includes transforming collective picture data into the familiar coordinate scheme. Metaheuristic-based methods were developed to explain the issue and improve the efficiency and accuracy of conventional image registration techniques due to their limitations. We describe a hybrid medical image registration technique using bio-inspired meta-heuristic algorithms: Betteroffspring and multi-crossover strategies increase convergent time and solution quality with an improvised coral reef optimization with modified mutual information (ICOR-MMI) algorithm. This new optimization method (ICORMMI) proposes that coral reefs expand, compete for space, and reproduce. A linear weighted sum of image intensity and contour flow model intensity is added to the mutual information calculation. Including statistical and spatial image data improves image registration. The established technique has been tested and verified using multiple medical image data sets, some of which contain single-modality and mixed-modality images (CT, MRI). The registration validates the proposed model's accuracy and efficiency and shows posture's contribution by incorporating statistical and geographical image data. This strategy is adapted to the real-coding problem and tested for realtime issues. The ICOR-MMI algorithm-based hybrid approach outperforms current results in time efficiency and toughness.

Medium- and Long-Term Load Forecasting for Power Plants Based on Causal Inference and Informer

Accurate forecasting of power plant loads is critical for maintaining a stable power supply, minimizing grid fluctuations, and enhancing power market trading mechanisms. However, the data on power plant generation load (hereinafter abbreviated as load) are non-stationary. The focus of existing load forecasting methods has been on continuously improving the ability to capture the dependent coupling between outputs and inputs, while research on external factors, which are the causes of non-stationary load data, has been neglected. The identification of the causal relationship between external variables and load is a significant factor in accurately predicting load. In the present study, the causal effects of various external variables on load were identified and then quantitatively calculated using various methods. Based on the improved Informer model, a long-time series forecasting model, a hybrid forecasting method was proposed called causal inference-improved Informer (hereinafter abbreviated as Causal–Informer). In the present study, the mutual information method was used to remove insignificant external variables. Subsequently, external factors such as GDP, holidays, ambient temperature, wind speed, power plant maintenance status, and rainfall were selected as input features of the proposed forecasting model. Finally, the proposed Causal–Informer method was evaluated using the historical load of a power plant in East China. Compared with four popular forecasting models, measurements on Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) for the proposed method were reduced by 89.8 million kwh–672.3 million kwh, 56.8 million kwh–637.9 million kwh, and 5.1–25.4%. The proposed method achieved the most accurate and stable results. The MAPE reached 10.4% and 24.8% in 30 time steps ahead and 90 time steps ahead of forecasts, respectively.

InfoCTM: A Mutual Information Maximization Perspective of Cross-Lingual Topic Modeling

Cross-lingual topic models have been prevalent for cross-lingual text analysis by revealing aligned latent topics. However, most existing methods suffer from producing repetitive topics that hinder further analysis and performance decline caused by low-coverage dictionaries. In this paper, we propose the Cross-lingual Topic Modeling with Mutual Information (InfoCTM). Instead of the direct alignment in previous work, we propose a topic alignment with mutual information method. This works as a regularization to properly align topics and prevent degenerate topic representations of words, which mitigates the repetitive topic issue. To address the low-coverage dictionary issue, we further propose a cross-lingual vocabulary linking method that finds more linked cross-lingual words for topic alignment beyond the translations of a given dictionary. Extensive experiments on English, Chinese, and Japanese datasets demonstrate that our method outperforms state-of-the-art baselines, producing more coherent, diverse, and well-aligned topics and showing better transferability for cross-lingual classification tasks.

Characterization of seawater intrusion based on machine learning and implications for offshore management under shared socioeconomic paths

With the increase of climate change and the risk of human activities, seawater intrusion has become an essential threat to human health and sustainable economic development in coastal areas. Therefore, identifying the critical features of seawater intrusion is significantly important, but the existing research on it is insufficient. Moreover, most of the work cannot reflect the direct impact of different shared socioeconomic paths on groundwater salinity in the future. To solve this problem, this study first established a numerical model of variable density flow for the Dagu River Basin based on SEAWAT to analyze the main aquifers' current flow field and concentration field distribution. Then two machine learning methods, random forest and mutual information methods, are introduced to identify critical features of seawater intrusion for the first time. Finally, taking the characteristics affected by climate are used as variables for selecting GCMs, the seawater intrusion in the study area is predicted by six GCMs under shared socioeconomic paths (ssp1-2.6 and ssp5-8.5) from 2020.12.30 to 2030.12.30. The findings indicate that regional precipitation and groundwater extraction exert significant influence on seawater intrusion, with the hydraulic conductivity of the primary aquifer also demonstrating crucial importance. Moreover, it is observed that the potential impact of different shared socioeconomic paths on future variations of seawater intrusion is negligible. This study verifies the effectiveness of machine learning in seawater intrusion analysis to a certain extent and has specific reference significance for the optimal management of water resources in coastal areas and the effective mitigation of seawater intrusion.

Lupus nephritis or not? A simple and clinically friendly machine learning pipeline to help diagnosis of lupus nephritis.

Diagnosis of lupus nephritis (LN) is a complex process, which usually requires renal biopsy. We aim to establish a machine learning pipeline to help diagnosis of LN. A cohort of 681 systemic lupus erythematosus (SLE) patients without LN and 786 SLE patients with LN was established, and a total of 95 clinical, laboratory data and 17 meteorological indicators were collected. After tenfold cross-validation, the patients were divided into training set and test set. The features selected by collective feature selection method of mutual information (MI) and multisurf were used to construct the models of logistic regression, decision tree, random forest, naive Bayes, support vector machine (SVM), light gradient boosting (LGB), extreme gradient boosting (XGB), and artificial neural network (ANN), the models were compared and verified in post-analysis. Collective feature selection method screens out antistreptolysin (ASO), retinol binding protein (RBP), lupus anticoagulant 1 (LA1), LA2, proteinuria and other features, and the hyperparameter optimized XGB (ROC: AUC = 0.995; PRC: AUC = 1.000, APS = 1.000; balance accuracy: 0.990) has the best performance, followed by LGB (ROC: AUC = 0.992; PRC: AUC = 0.997, APS = 0.977; balance accuracy: 0.957). The worst performance is naive Bayes model (ROC: AUC = 0.799; PRC: AUC = 0.822, APS = 0.823; balance accuracy: 0.693). In the composite feature importance bar plots, ASO, RF, Up/Ucr, and other features play important roles in LN. We developed and validated a new and simple machine learning pathway for diagnosis of LN, especially the XGB model based on ASO, LA1, LA2, proteinuria, and other features screened out by collective feature selection.

Application of Metaheuristic Algorithms and ANN Model for Univariate Water Level Forecasting

With the rapid development of machine learning (ML) models, the artificial neural network (ANN) is being increasingly applied for forecasting hydrological processes. However, researchers have not treated hybrid ML models in much detail. To address these issues, this study herein suggests a novel methodology to forecast the monthly water level (WL) based on multiple lags of the Tigris River in Al-Kut, Iraq, over ten years. The methodology includes preprocessing data methods, and the ANN model optimises with a marine predator algorithm (MPA). In the optimisation procedure, to decrease uncertainty and expand the predicting range, the slime mould algorithm (SMA-ANN), constriction coefficient-based particle swarm optimisation and chaotic gravitational search algorithms (CPSOCGSA-ANN), and particle swarm optimisation (PSO-ANN) are applied to compare and validate the MPA-ANN model performance. Analysis of results revealed that the data pretreatment methods improved the original data quality and selected the ideal predictors’ scenario by singular spectrum analysis and mutual information methods, respectively. For example, the correlation coefficient of the first lag improved from 0.648 to 0.938. Depending on various evaluation metrics, MPA-ANN tends to forecast WL better than SMA-ANN, PSO-ANN, and CPSOCGSA-ANN algorithms with coefficients of determination of 0.94, 0.81, 0.85, and 0.90, respectively. Evidence shows that the proposed methodology yields excellent results, with a scatter index equal to 0.002. The research outcomes represent an additional step towards evolving various hybrid ML techniques, which are valuable to practitioners wishing to forecast WL data and the management of water resources in light of environmental shifts.

A Hybrid Forecasting Model to Simulate the Runoff of the Upper Heihe River

River runoff simulation and prediction are important for controlling the water volume and ensuring the optimal allocation of water resources in river basins. However, the instability of medium- and long-term runoff series increases the difficulty of runoff forecasting work. In order to improve the prediction accuracy, this research establishes a hybrid deep learning model framework based on variational mode decomposition (VMD), the mutual information method (MI), and a long short-term memory network (LSTM), namely, VMD-LSTM. First, the original runoff data are decomposed into a number of intrinsic mode functions (IMFs) using VMD. Then, for each IMF, a long short-term memory (LSTM) network is applied to establish the prediction model, and the MI method is used to determine the data input lag time. Finally, the prediction results of each subsequence are reconstructed to obtain the final forecast result. We explored the predictive performance of the model with regard to monthly runoff in the upper Heihe River Basin, China, and compared its performance with other single and hybrid models. The results show that the proposed model has obvious advantages in terms of the performance of point prediction and interval prediction compared to several comparative models. The Nash–Sutcliffe efficiency coefficient (NSE) of the prediction results reached 0.96, and the coverage of the interval prediction reached 0.967 and 0.908 at 95% and 90% confidence intervals, respectively. Therefore, the proposed model is feasible for simulating the monthly runoff of this watershed.

Prediction of the pore structure by machine learning techniques in the carbonate reservoirs in Iraq H oilfield

Pore structure impacts the capability of seepage pattern of subsurface fluid mineral and mineral exploitation efficiency. Because of the strong heterogeneity in carbonate reservoirs, the pore structure is nonlinearly varying and complex in reservoirs. It is necessary to establish a method for pore structure type (PST) prediction. Machine learning provides an efficient prediction method by finding the relationship between core test data and well‐logging data. In this paper, a reservoir identification method based on Gradient Boost Decision Trees (GBDT) is proposed for the pore structure characteristics of carbonate reservoirs integrating core test and logging data. Core testing data are utilized to form the training set for PST prediction. Then, we adopt the mutual information method to optimize the logging data as the input of the machine learning model. GBDT hyperparameters are optimized using the learning curve to make it have optimal performance. Finally, we compare the predicted result of GBDT with K‐Nearest Neighbor and support vector machines, demonstrating the superior performance of GBDT. The results indicate that the four PS are different in some properties, such as permeability, pore radius and shape of pore and throat. 20%, 10 and 0.15 are recommended as the optimal parameter values for Maximum feature, iteration number and Sub‐sample number to improve GBDT accuracy in carbonate reservoir PST prediction. Evaluation indicators of machine learning, including the Kappa coefficient, Heming distance, Jaccard similarity coefficient and Confusion matrices, indicate that GBDT has better performance. This research proves the high accuracy and applicability of machine learning technology in the carbonate reservoir and also provides a meaningful reference for the identification of the carbonate reservoirs in the central area of Iraq.

Surface Mounted Production Line Capacity Prediction Based on Data Driven

Capacity prediction of surface mount production lines is important for improving productivity, dynamic production scheduling, and optimizing production structure. An integrated mutual information-cuckoo search-extreme learning machine (MI-CS-ELM) prediction model is proposed for the impact of nonlinear and high-dimensional features of surface-mounted production lines on capacity. Firstly, the features affecting the production capacity are ranked based on the mutual information method, and the predictive input data set is reconstructed. Secondly, the input parameters of the prediction model are adjusted by the cuckoo search algorithm and the capacity prediction model is built based on an extreme limit learning machine. Finally, the rationality of the proposed method is verified by an example to guide capacity planning and production scheduling optimization.

Individualized Recommendation Method of Multimedia Network Teaching Resources Based on Classification Algorithm in a Smart University

Abstract Current recommendation methods of multimedia network teaching resources cannot classify the resources with high precision, which leads to large deviation of recommendation results and low recall. Therefore, a personalized recommendation method of multimedia network teaching resources based on classification algorithm is proposed. Adaptive sliding window mutual information method is used to process historical and incremental data of teaching resources and extract the characteristics of multimedia network teaching resources. Two-step clustering algorithm combined with business identifier code criteria are used to complete resource classification. Based on structured query language (SQL) multimedia network teaching resources data recommendation database, document query and processing, combined with SQL structure, a personalized recommendation of multimedia network teaching resources is completed. The experimental results show that the recall of the classification results is higher than 99.58 %, and the highest F1 value is 97.28 %. The precision of personalized recommendation of multimedia network teaching resources is always higher than 90 %, and the recall rate is also higher.

Hydraulic fracturing performance analysis by the mutual information and Gaussian process regression methods

In shale gas production rate prediction, data are commonly acquired through simulations, which offer benefits but cannot fully predict the outcome of real engineering projects. This paper applied a Gaussian process regression (GPR) model, which was trained on the geological and field data from 175 wells in the Fuling gas field, China. GPR was chosen because this method can be suitably applied to a small dataset. To improve the accuracy, reduce the computational consumption, and increase the interpretability of the model, the mutual information method was used for feature selection after Pearson coefficient analysis. Based on comparison of the GPR, random forest and artificial neural network models, GPR achieved the best performance with the mutual information feature selection method, exhibiting R2 values of 0.71 and 0.68 for the training and test datasets, respectively. After model training, the model applied the Shapley additive explanation (SHAP) value method. The results indicate that geological factors dominate the gas production rate in Fuling, China. The volume of fluid loss could exert a negative impact on production rates and increase the consumption of 40–70 mesh proppant could increase gas production in Fuling.

Fine-grained aspect-based opinion mining on online course reviews for feedback analysis

ABSTRACT Online learning has developed rapidly and billions of learners have participated in various courses. However, the high dropout rate is universal and learning performance is not satisfactory. Fortunately, learners have posted a large number of reviews which express their feedback opinions. The fine-grained aspects and opinions existing in reviews provide valuable information for educators to improve learning performance. Thus, it is necessary to make fine-grained aspect-based opinion mining on course reviews for feedback analysis. First, the latent Dirichlet allocation is applied to generate topics and the topic related words from course reviews. Each word belongs to a topic and these words are considered as the core aspects. Second, a series of rules and an algorithm are designed based on dependency syntax analysis to extract the fine-grained aspects and opinions. The aspect-based opinion candidate set is generated. Then the corresponding opinions of the core aspects are selected from the candidate set. Third, the sentiment score of each opinion is calculated combining the dictionary-based and pointwise mutual information method to identify the polarity of the opinion. The extracted aspects and opinions with sentiment polarity provide fine-grained feedback information from several topic perspectives.

Circuit Breaker Fault Diagnosis Method Based on Coil Current Time Sequence Phase Trajectory Characteristics

The traditional circuit breaker fault diagnosis method suffers from insufficient feature information extraction and is easily affected by abnormal signal acquisition. To address this, this paper introduces the phase space reconstruction algorithm to reconstruct the current signal for fault diagnosis based on phase trajectory features. The proposed method uses a first-order forward differencing method and mutual information method to process abnormal data and select the parameters of the reconstruction, then extract overall and local inflection point features to construct a fault feature set. The support vector machine algorithm-based model is trained and tested using actual samples, and the results show that the proposed method can adaptively sample anomalous signals, exhibit strong robustness, and significantly improve the accuracy of fault classification.

Predicting postoperative delirium after hip arthroplasty for elderly patients using machine learning

Postoperative delirium (POD) is a common and severe complication in elderly hip-arthroplasty patients. This study aims to develop and validate a machine learning (ML) model that determines essential features related to POD and predicts POD for elderly hip-arthroplasty patients. The electronic record data of elderly patients who received hip-arthroplasty surgery between January 2017 and April 2021 were enrolled as the dataset. The Confusion Assessment Method (CAM) was administered to the patients during their perioperative period. The feature section method was employed as a filter to determine leading features. The classical machine learning algorithms were trained in cross-validation processing, and the model with the best performance was built in predicting the POD. Metrics of the area under the curve (AUC), accuracy (ACC), sensitivity, specificity, and F1-score were calculated to evaluate the predictive performance. 476 Arthroplasty elderly patients with general anesthesia were included in this study, and the final model combined feature selection method mutual information (MI) and linear binary classifier using logistic regression (LR) achieved an encouraging performance (AUC = 0.94, ACC = 0.88, sensitivity = 0.85, specificity = 0.90, F1-score = 0.87) on a balanced test dataset. The model could predict POD with satisfying accuracy and reveal important features of suffering POD such as age, Cystatin C, GFR, CHE, CRP, LDH, monocyte count, history of mental illness or psychotropic drug use and intraoperative blood loss. Proper preoperative interventions for these factors could reduce the incidence of POD among elderly patients.

A multi-objective based radiomics feature selection method for response prediction following radiotherapy

Objective. Radiomics contains a large amount of mineable information extracted from medical images, which has important significance in treatment response prediction for personalized treatment. Radiomics analyses generally involve high dimensions and redundant features, feature selection is essential for construction of prediction models. Approach. We proposed a novel multi-objective based radiomics feature selection method (MRMOPSO), where the number of features, sensitivity, and specificity are jointly considered as optimization objectives in feature selection. The MRMOPSO innovated in the following three aspects: (1) Fisher score to initialize the population to speed up the convergence; (2) Min-redundancy particle generation operations to reduce the redundancy between radiomics features, a truncation strategy was introduced to further reduce the number of features effectively; (3) Particle selection operations guided by elitism strategies to improve local search ability of the algorithm. We evaluated the effectiveness of the MRMOPSO by using a multi-institution oropharyngeal cancer dataset from The Cancer Imaging Archive. 357 patients were used for model training and cross validation, an additional 64 patients were used for evaluation. Main results. The area under the curve (AUC) of our method achieved AUCs of 0.82 and 0.84 for cross validation and independent dataset, respectively. Compared with classical feature selection methods, the AUC of MRMOPSO is significantly higher than the Lasso (AUC = 0.74, p-value = 0.02), minimal-redundancy-maximal-relevance criterion (mRMR) (AUC = 0.73, p-value = 0.05), F-score (AUC = 0.48, p-value < 0.01), and mutual information (AUC = 0.69, p-value < 0.01) methods. Compared to single-objective methods, the AUC of MRMOPSO is 12% higher than those of the genetic algorithm (GA) (AUC = 0.68, p-value = 0.02) and particle swarm optimization algorithm (AUC = 0.72, p-value = 0.05) methods. Compared to other multi-objective feature selection methods, the AUC of MRMOPSO is 14% higher than those of multiple objective particle swarm optimization (MOPSO) (AUC = 0.68, p-value = 0.02) and nondominated sorting genetic algorithm II (NSGA2) (AUC = 0.70, p-value = 0.03). Significance. We proposed a multi-objective based radiomics feature selection method. Compared to conventional feature reduction algorithms, the proposed algorithm effectively reduced feature dimension, and achieved superior performance, with improved sensitivity and specificity, for response prediction in radiotherapy.