Multi-class Classification Performance Research Articles

An accurate valvular heart disorders detection model based on a new dual symmetric tree pattern using stethoscope sounds

Background and purposeValvular heart disease (VHD) is an important cause of morbidity and mortality. Echocardiography is the reference standard for VHD diagnosis but is not universally accessible. Manual cardiac auscultation is inadequate for screening VHD. Many machine learning models using heart sounds acquired with an electronic stethoscope may improve the accuracy of VHD diagnosis. We aimed to develop an accurate sound classification model for VHD diagnosis. Materials and methodsA new large stethoscope sound dataset containing 10,366 heart sounds divided into ten categories (nine VHD and one healthy) were prospectively collected. We developed a handcrafted learning model that comprised multilevel feature extraction based on a dual symmetric tree pattern (DSTP) and multilevel discrete wavelet transform (DWT), feature selection, and classification. The multilevel DWT was used to create subbands to extract features at both high and low levels. Then, iterative neighborhood component analysis was used to select the most discriminative 512 features from among the extracted features in the generated feature vector. In the classification phase, a support vector machine (SVM) was used with 10-fold cross-validation (CV) and leave-one-subject-out (LOSO) CV. ResultsOur proposed DSTP-based model attained 99.58% and 99.84% classification accuracies using SVM classifier with 10-fold CV and LOSO CV, respectively. ConclusionsThe presented DSTP-based classification model attained excellent multiclass classification performance on a large prospective heart sound dataset at a low computational cost.

Deep Learning-Based Multilevel Classification of Alzheimer’s Disease Using Non-invasive Functional Near-Infrared Spectroscopy

The timely diagnosis of Alzheimer’s disease (AD) and its prodromal stages is critically important for the patients, who manifest different neurodegenerative severity and progression risks, to take intervention and early symptomatic treatments before the brain damage is shaped. As one of the promising techniques, functional near-infrared spectroscopy (fNIRS) has been widely employed to support early-stage AD diagnosis. This study aims to validate the capability of fNIRS coupled with Deep Learning (DL) models for AD multi-class classification. First, a comprehensive experimental design, including the resting, cognitive, memory, and verbal tasks was conducted. Second, to precisely evaluate the AD progression, we thoroughly examined the change of hemodynamic responses measured in the prefrontal cortex among four subject groups and among genders. Then, we adopted a set of DL architectures on an extremely imbalanced fNIRS dataset. The results indicated that the statistical difference between subject groups did exist during memory and verbal tasks. This presented the correlation of the level of hemoglobin activation and the degree of AD severity. There was also a gender effect on the hemoglobin changes due to the functional stimulation in our study. Moreover, we demonstrated the potential of distinguished DL models, which boosted the multi-class classification performance. The highest accuracy was achieved by Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) using the original dataset of three hemoglobin types (0.909 ± 0.012 on average). Compared to conventional machine learning algorithms, DL models produced a better classification performance. These findings demonstrated the capability of DL frameworks on the imbalanced class distribution analysis and validated the great potential of fNIRS-based approaches to be further contributed to the development of AD diagnosis systems.

Hierarchy-aware contrastive learning with late fusion for skin lesion classification

Background and ObjectiveThe incidence rate of skin cancers is increasing worldwide annually. Using machine learning and deep learning for skin lesion classification is one of the essential research topics. In this study, we formulate a major-type misclassification problem that previous studies did not consider in the multi-class skin lesion classification. Moreover, addressing the major-type misclassification problem is significant for real-world computer-aided diagnosis. MethodsThis study presents a novel method, namely Hierarchy-Aware Contrastive Learning with Late Fusion (HAC-LF), to improve the overall performance of multi-class skin classification. In HAC-LF, we design a new loss function, Hierarchy-Aware Contrastive Loss (HAC Loss), to reduce the impact of the major-type misclassification problem. The late fusion method is applied to balance the major-type and multi-class classification performance. ResultsWe conduct a series of experiments with the ISIC 2019 Challenges dataset, which consists of three skin lesion datasets, to verify the performance of our methods. The results show that our proposed method surpasses the representative deep learning methods for skin lesion classification in all evaluation metrics used in this study. HAC-LF achieves 0.871, 0.842, 0.889 for accuracy, sensitivity, and specificity in the major-type classification, respectively. With the imbalanced class distribution, HAC-LF outperforms the baseline model regarding the sensitivity of minority classes. ConclusionsThis research formulates a major-type misclassification problem. We propose HAC-LF to deal with it and boost the multi-class skin lesion classification performance. According to the results, the advantage of HAC-LF is that the proposed HAC Loss can beneficially reduce the impact of the major-type misclassification by decreasing the major-type error rate. Besides the medical field HAC-LF is promising to be applied to other domains possessing the data with the hierarchical structure.

Residual Learning for Marine Mammal Classification

The passive acoustic monitoring of marine mammals is an essential tool for researchers tracking the populations of individual species in threatened environments. Given the large quantity of audio data generated by passive acoustic arrays, it is desirable to automate the process of identifying marine mammals present in the recordings. Utilizing acoustic data from the William A. Watkins Marine Mammal Sounds Database, we present an approach using residual learning networks (ResNets) for classifying the marine mammal vocalizations of up to 32 species. We first determine the optimal methods for converting acoustic recordings into discrete spectrograms suitable for input into neural networks. A series of configurations for spectrographic window functions, preprocessing augmentations, and multi-channel spectrogram generation are examined. Each configuration’s spectrographic output is used to train a residual learning network. Its multi-class classification performance is ranked using the harmonic mean of precision and recall to calculate a weighted F1-score. Configurations specifying 512×256 spectrograms created with a Hann window of 1024 and utilizing horizontal roll demonstrate superior performance. We use the top-performing configurations to generate training data as input for a series of single and multi-channel residual neural networks. These networks are trained to high precision before evaluating their multi-class classification performance. A single-channel network performed the best, obtaining an F1-score of 0.867 with an AUC of 0.9281 on a 32-class classification task. Our multi-channel configuration obtained an F1-score of 0.846 with an AUC of 0.9169. While we demonstrate that networks may learn more information from multi-channel spectrographic inputs, we find that single-channel spectrograms offer superior classification performance overall.

Analysis of multi-class classification performance metrics for remote sensing imagery imbalanced datasets

Remote sensing imaging datasets for classification generally present high levels of imbalance between classes of interest. This work presented a study of a set of performance evaluation metrics for an imbalance dataset. In this work, a support vector machine (SVM) was used to perform the classification of seven classes of interest in a popular dataset called Salinas-A. The performance evaluation of the classifier was performed using two types of metrics: 1) Metrics for multi-class classification, and 2) Metrics based on the binary confusion matrix. In the results, a comparison of the scores of each metric is developed, some being more optimistic than others due to the bias that they present given the imbalance. In addition, our case study helps to conclude that the Matthews correlation coefficient (MCC) presents the lowest bias in imbalanced cases and is regarded to be robust metric. These results can be extended to any imbalanced dataset taking into account the equations developed by Luque.

Robust and Dynamic Graph Convolutional Network For Multi-view Data Classification

AbstractSince graph learning could preserve the structure information of the samples to improve the learning ability, it has been widely applied in both shallow learning and deep learning. However, the current graph learning methods still suffer from the issues such as outlier influence and model robustness. In this paper, we propose a new dynamic graph neural network (DGCN) method to conduct semi-supervised classification on multi-view data by jointly conducting the graph learning and the classification task in a unified framework. Specifically, our method investigates three strategies to improve the quality of the graph before feeding it into the GCN model: (i) employing robust statistics to consider the sample importance for reducing the outlier influence, i.e. assigning every sample with soft weights so that the important samples are with large weights and outliers are with small or even zero weights; (ii) learning the common representation across all views to improve the quality of the graph for every view; and (iii) learning the complementary information from all initial graphs on multi-view data to further improve the learning of the graph for every view. As a result, each of the strategies could improve the robustness of the DGCN model. Moreover, they are complementary for reducing outlier influence from different aspects, i.e. the sample importance reduces the weights of the outliers, both the common representation and the complementary information improve the quality of the graph for every view. Experimental result on real data sets demonstrates the effectiveness of our method, compared to the comparison methods, in terms of multi-class classification performance.

Feasibility of Machine Learning Algorithms for Predicting the Deformation of Anodic Titanium Films by Modulating Anodization Processes.

This study aims to demonstrate the feasibility of applying eight machine learning algorithms to predict the classification of the surface characteristics of titanium oxide (TiO2) nanostructures with different anodization processes. We produced a total of 100 samples, and we assessed changes in TiO2 nanostructures’ thicknesses by performing anodization. We successfully grew TiO2 films with different thicknesses by one-step anodization in ethylene glycol containing NH4F and H2O at applied voltage differences ranging from 10 V to 100 V at various anodization durations. We found that the thicknesses of TiO2 nanostructures are dependent on anodization voltages under time differences. Therefore, we tested the feasibility of applying machine learning algorithms to predict the deformation of TiO2. As the characteristics of TiO2 changed based on the different experimental conditions, we classified its surface pore structure into two categories and four groups. For the classification based on granularity, we assessed layer creation, roughness, pore creation, and pore height. We applied eight machine learning techniques to predict classification for binary and multiclass classification. For binary classification, random forest and gradient boosting algorithm had relatively high performance. However, all eight algorithms had scores higher than 0.93, which signifies high prediction on estimating the presence of pore. In contrast, decision tree and three ensemble methods had a relatively higher performance for multiclass classification, with an accuracy rate greater than 0.79. The weakest algorithm used was k-nearest neighbors for both binary and multiclass classifications. We believe that these results show that we can apply machine learning techniques to predict surface quality improvement, leading to smart manufacturing technology to better control color appearance, super-hydrophobicity, super-hydrophilicity or batter efficiency.

Classification of Biomedical Texts for Cardiovascular Diseases with Deep Neural Network Using a Weighted Feature Representation Method.

This study aims to improve the performance of multiclass classification of biomedical texts for cardiovascular diseases by combining two different feature representation methods, i.e., bag-of-words (BoW) and word embeddings (WE). To hybridize the two feature representations, we investigated a set of possible statistical weighting schemes to combine with each element of WE vectors, which were term frequency (TF), inverse document frequency (IDF) and class probability (CP) methods. Thus, we built a multiclass classification model using a bidirectional long short-term memory (BLSTM) with deep neural networks for all investigated operations of feature vector combinations. We used MIMIC III and the PubMed dataset for the developing language model. To evaluate the performance of our weighted feature representation approaches, we conducted a set of experiments for examining multiclass classification performance with the deep neural network model and other state-of-the-art machine learning (ML) approaches. In all experiments, we used the OHSUMED-400 dataset, which includes PubMed abstracts related with specifically one class over 23 cardiovascular disease categories. Afterwards, we presented the results obtained from experiments and provided a comparison with related research in the literature. The results of the experiment showed that our BLSTM model with the weighting techniques outperformed the baseline and other machine learning approaches in terms of validation accuracy. Finally, our model outperformed the scores of related studies in the literature. This study shows that weighted feature representation improves the performance of the multiclass classification.

Automatic A-Phase Detection of Cyclic Alternating Patterns in Sleep Using Dynamic Temporal Information.

The identification of recurrent, transient perturbations in brain activity during sleep, so called cyclic alternating patterns (CAP), is of significant interest as they have been linked to neurological pathologies. CAP sequences comprise multiple, consecutive cycles of phasic activation (A-phases). Here, we propose a novel, automated system exploiting the dynamical, temporal information in electroencephalography (EEG) recordings for the classification of A-phases and their subtypes. Using recurrent neural networks (RNN), crucial information in the temporal behavior of the EEG is extracted. The automatic classification system is equipped to deal with the biasing issue of imbalanced data sets and uses state-of-the-art signal processing methods to reduce inter-subject variation. To evaluate our system, we applied recordings from the publicly available CAP Sleep Database on Physionet. Our results show that the RNN improved the detection accuracy by 3-5% and the F1-score by approximately 7% on two data sets compared to a normal feed-forward neural network. Our system achieves a sensitivity of approximately 76-78% and F1-score between 63-68%, significantly outperforming existing technologies. Moreover, its sensitivity for subtype classification of 60-63% (A1), 42-45% (A2), and 71-74% (A3) indicates superior multi-class classification performance for CAP detection. In conclusion, we have developed a fully automated high performance CAP scoring system that includes A-phase subtype classification. RNN classifiers yield a significant improvement in accuracy and sensitivity compared to previously proposed systems.

Using Ensemble Technique to Improve Multiclass Classification

Many real world applications inevitably contain datasets that have multiclass structure characterized by imbalance classes, redundant and irrelevant features that degrade performance of classifiers. Minority classes in the datasets are treated as outliers’ classes. The research aimed at establishing the role of ensemble technique in improving performance of multiclass classification. Multiclass datasets were transformed to binary and the datasets resampled using Synthetic minority oversampling technique (SMOTE) algorithm. Relevant features of the datasets were selected by use of an ensemble filter method developed using Correlation, Information Gain, Gain-Ratio and ReliefF filter selection methods. Adaboost and Random subspace learning algorithms were combined using Voting methodology utilizing random forest as the base classifier. The classifiers were evaluated using 10 fold stratified cross validation. The model showed better performance in terms of outlier detection and classification prediction for multiclass problem. The model outperformed other well-known existing classification and outlier detection algorithms such as Naive bayes, KNN, Bagging, JRipper, Decision trees, RandomTree and Random forest. The study findings established that ensemble technique, resampling datasets and decomposing multiclass results in an improved classification performance as well as enhanced detection of minority outlier (rare) classes. Keywords: Multiclass, Classification, Outliers, Ensemble, Learning Algorithm DOI : 10.7176/JIEA/9-5-04 Publication date : August 31 st 2019

CAESNet: Convolutional AutoEncoder based Semi-supervised Network for improving multiclass classification of endomicroscopic images.

This article presents a novel method of semisupervised learning using convolutional autoencoders for optical endomicroscopic images. Optical endomicroscopy (OE) is a newly emerged biomedical imaging modality that can support real-time clinical decisions for the grade of dysplasia. To enable real-time decision making, computer-aided diagnosis (CAD) is essential for its high speed and objectivity. However, traditional supervised CAD requires a large amount of training data. Compared with the limited number of labeled images, we can collect a larger number of unlabeled images. To utilize these unlabeled images, we have developed a Convolutional AutoEncoder based Semi-supervised Network (CAESNet) for improving the classification performance. We applied our method to an OE dataset collected from patients undergoing endoscope-based confocal laser endomicroscopy procedures for Barrett's esophagus at Emory Hospital, which consists of 429 labeled images and 2826 unlabeled images. Our CAESNet consists of an encoder with 5 convolutional layers, a decoder with 5 transposed convolutional layers, and a classification network with 2 fully connected layers and a softmax layer. In the unsupervised stage, we first update the encoder and decoder with both labeled and unlabeled images to learn an efficient feature representation. In the supervised stage, we further update the encoder and the classification network with only labeled images for multiclass classification of the OE images. Our proposed semisupervised method CAESNet achieves the best average performance for multiclass classification of OE images, which surpasses the performance of supervised methods including standard convolutional networks and convolutional autoencoder network. Our semisupervised CAESNet can efficiently utilize the unlabeled OE images, which improves the diagnosis and decision making for patients with Barrett's esophagus.

Multi class disorder detection of magnetic resonance brain images using composite features and neural network.

Brain disorder recognition has becoming a promising area of study. In reality, somedisorders share similar features and signs, making the task ofdiagnosis and treatment challenging. This paper presents a rigorous and robust computer aided diagnosis system for the detection of multiple brain abnormalities which can assist physicians in the diagnosis and treatment of brain diseases. In this system, we used energy of wavelet sub bands, textural features of gray level co-occurrence matrix and intensity feature of MR brain images. These features are ranked using Wilcoxon test. The composite features are classified using back propagation neural network. Bayesian regulation is adopted to find the optimal weights of neural network. The experimentation is carried out on datasets DS-90 and DS-310 of Harvard Medical School. To enhance the generalization capability of the network, fivefold stratified cross validation technique is used. The proposed system yields multi class disease classification accuracy of 100% in differentiating 90 MR brain images into 18 classes and 97.81% in differentiating 310 MR brain images into 6 classes. The experimental results reveal that the composite features along with BPNN classifier create a competent and reliable system for the identification of multiple brain disorders which can be used in clinical applications. The Wilcoxon test outcome demonstrates that standard deviation feature along with energies of approximate and vertical sub bands of level 7 contribute the most in achieving enhanced multi class classification performance results.

New Insights into Multiclass Damage Classification of Tsunami-Induced Building Damage from SAR Images

The fine resolution of synthetic aperture radar (SAR) images enables the rapid detection of severely damaged areas in the case of natural disasters. Developing an optimal model for detecting damage in multitemporal SAR intensity images has been a focus of research. Recent studies have shown that computing changes over a moving window that clusters neighboring pixels is effective in identifying damaged buildings. Unfortunately, classifying tsunami-induced building damage into detailed damage classes remains a challenge. The purpose of this paper is to present a novel multiclass classification model that considers a high-dimensional feature space derived from several sizes of pixel windows and to provide guidance on how to define a multiclass classification scheme for detecting tsunami-induced damage. The proposed model uses a support vector machine (SVM) to determine the parameters of the discriminant function. The generalization ability of the model was tested on the field survey of the 2011 Great East Japan Earthquake and Tsunami and on a pair of TerraSAR-X images. The results show that the combination of different sizes of pixel windows has better performance for multiclass classification using SAR images. In addition, we discuss the limitations and potential use of multiclass building damage classification based on performance and various classification schemes. Notably, our findings suggest that the detectable classes for tsunami damage appear to differ from the detectable classes for earthquake damage. For earthquake damage, it is well known that a lower damage grade can rarely be distinguished in SAR images. However, such a damage grade is apparently easy to identify from tsunami-induced damage grades in SAR images. Taking this characteristic into consideration, we have successfully defined a detectable three-class classification scheme.

Multiclass classification of obstructive sleep apnea/hypopnea based on a convolutional neural network from a single-lead electrocardiogram

Objective: In this paper, we propose a convolutional neural network (CNN)-based deep learning architecture for multiclass classification of obstructive sleep apnea and hypopnea (OSAH) using single-lead electrocardiogram (ECG) recordings. OSAH is the most common sleep-related breathing disorder. Many subjects who suffer from OSAH remain undiagnosed; thus, early detection of OSAH is important. Approach: In this study, automatic classification of three classes—normal, hypopnea, and apnea—based on a CNN is performed. An optimal six-layer CNN model is trained on a training dataset (45 096 events) and evaluated on a test dataset (11 274 events). The training set (69 subjects) and test set (17 subjects) were collected from 86 subjects with length of approximately 6 h and segmented into 10 s durations. Main results: The proposed CNN model reaches a mean -score of 93.0 for the training dataset and 87.0 for the test dataset. Significance: Thus, proposed deep learning architecture achieved a high performance for multiclass classification of OSAH using single-lead ECG recordings. The proposed method can be employed in screening of patients suspected of having OSAH.

SVM-based multi-state-mapping approach for multi-class classification

Traditional SVM-based multi-class classification algorithms mainly adopt the strategy of mapping the data set with all classes into a single feature space via a kernel function, in which SVM is constructed for each decomposed binary classification problem. However, it is not always possible to find an appropriate kernel function to render all the classes distinguishable in a single feature space, since each class is always derived from different data distributions. Consequently, the performance is not always as good as expected. To improve the performance of multi-class classification, this paper proposes an improved approach, called multi-state-mapping (MSM) with SVM based on hierarchical architecture, which maps the data set with all classes into different feature spaces at the different states of the decomposition of a multi-class classification problem in terms of a binary tree architecture. We prove that the computational complexity of MSM at its worst lies between that of the one-against-all scheme and one-against-one scheme. Substantial experiments have been conducted on sixteen UCI data sets to show the performance of our method. The statistical results show that MSM outperforms state-of-the-art methods in terms of accuracy and standard deviation.

Side-channel attacks and learning-vector quantization

The security of cryptographic systems is a major concern for cryptosystem designers, even though cryptography algorithms have been improved. Side-channel attacks, by taking advantage of physical vulnerabilities of cryptosystems, aim to gain secret information. Several approaches have been proposed to analyze side-channel information, among which machine learning is known as a promising method. Machine learning in terms of neural networks learns the signature (power consumption and electromagnetic emission) of an instruction, and then recognizes it automatically. In this paper, a novel experimental investigation was conducted on field-programmable gate array (FPGA) implementation of elliptic curve cryptography (ECC), to explore the efficiency of side-channel information characterization based on a learning vector quantization (LVQ) neural network. The main characteristics of LVQ as a multi-class classifier are that it has the ability to learn complex non-linear input-output relationships, use sequential training procedures, and adapt to the data. Experimental results show the performance of multi-class classification based on LVQ as a powerful and promising approach of side-channel data characterization.

Deep belief network-based drug identification using near infrared spectroscopy

Near infrared spectroscopy (NIRS) analysis technology, combined with chemometrics, can be effectively used in quick and nondestructive analysis of quality and category. In this paper, an effective drug identification method by using deep belief network (DBN) with dropout mechanism (dropout-DBN) to model NIRS is introduced, in which dropout is employed to overcome the overfitting problem coming from the small sample. This paper tests proposed method under datasets of different sizes with the example of near infrared diffuse reflectance spectroscopy of erythromycin ethylsuccinate drugs and other drugs, aluminum and nonaluminum packaged. Meanwhile, it gives experiments to compare the proposed method’s performance with back propagation (BP) neural network, support vector machines (SVMs) and sparse denoising auto-encoder (SDAE). The results show that for both binary classification and multi-classification, dropout mechanism can improve the classification accuracy, and dropout-DBN can achieve best classification accuracy in almost all cases. SDAE is similar to dropout-DBN in the aspects of classification accuracy and algorithm stability, which are higher than that of BP neural network and SVM methods. In terms of training time, dropout-DBN model is superior to SDAE model, but inferior to BP neural network and SVM methods. Therefore, dropout-DBN can be used as a modeling tool with effective binary and multi-class classification performance on a spectrum sample set of small size.

Improving Performance of Multiclass Classification by Inducing Class Hierarchies

In the last decades, one issue that has received a lot of attention in classification problems is how to obtain better classifications. This problem becomes even more complicated when the number of classes is high. In this multiclass scenario, it is assumed that the class labels are independent of each other, and thus, most techniques and methods proposed to improve the performance of the classifiers rely on it. An alternative way to address the multiclass problem is to hierarchically distribute the classes in a collection of multiclass subproblems by reducing the number of classes involved in each local subproblem. In this paper, we propose a new method for inducing a class hierarchy from the confusion matrix of a multiclass classifier. Then, we use the class hierarchy to learn a tree-like hierarchy of classifiers for solving the original multiclass problem in a similar way as the top-down hierarchical classification approach does for working with hierarchical domains. We experimentally evaluate the proposal on a collection of multiclass datasets showing that, in general, the generated hierarchies not only outperforms the original (flat) classification but also hierarchical approaches based on other alternative ways of constructing the class hierarchy.

Improving Multi-class Classification for Endomicroscopic Images by Semi-supervised Learning.

Optical Endomicroscopy (OE) is a newly-emerged biomedical imaging modality that can help physicians make real-time clinical decisions about patients' grade of dysplasia. However, the performance of applying medical imaging classification for computer-aided diagnosis is primarily limited by the lack of labeled images. To improve the classification performance, we propose a semi-supervised learning algorithm that can incorporate large sets of unlabeled images. Our real-world endo-microscopic imaging datasets consist of 425 labeled images and 2,826 unlabeled ones. With semi-supervised learning algorithms, we improved multi-class classification performance over supervised learning algorithms by around 10% in all evaluation metrics, namely precision, recall, F1 score and Cohen-Kappa statistics.

Iterative ensemble feature selection for multiclass classification of imbalanced microarray data.

BackgroundMicroarray technology allows biologists to monitor expression levels of thousands of genes among various tumor tissues. Identifying relevant genes for sample classification of various tumor types is beneficial to clinical studies. One of the most widely used classification strategies for multiclass classification data is the One-Versus-All (OVA) schema that divides the original problem into multiple binary classification of one class against the rest. Nevertheless, multiclass microarray data tend to suffer from imbalanced class distribution between majority and minority classes, which inevitably deteriorates the performance of the OVA classification.ResultsIn this study, we propose a novel iterative ensemble feature selection (IEFS) framework for multiclass classification of imbalanced microarray data. In particular, filter feature selection and balanced sampling are performed iteratively and alternatively to boost the performance of each binary classification in the OVA schema. The proposed framework is tested and compared with other representative state-of-the-art filter feature selection methods using six benchmark multiclass microarray data sets. The experimental results show that IEFS framework provides superior or comparable performance to the other methods in terms of both classification accuracy and area under receiver operating characteristic curve. The more number of classes the data have, the better performance of IEFS framework achieves.ConclusionsBalanced sampling and feature selection together work well in improving the performance of multiclass classification of imbalanced microarray data. The IEFS framework is readily applicable to other biological data analysis tasks facing the same problem.