Multiple Classifier Systems Research Articles

Understanding of BRCA VUS genetic results by breast cancer specialists.

BackgroundMainstreaming genetic medicine, increased media coverage and clinical trials for BRCA mutation carriers are leading oncologists into more patient discussions about BRCA genetic testing. BRCA variants of uncertain significance (VUS) occur in 10–20 % of tests. VUS detection introduces additional uncertainty for patient and potentially clinician. We aimed to explore the ability of breast cancer specialists (BCS) in the UK to correctly respond to a VUS report.MethodsA survey sent to 800 UK BCS collected demographics data, VUS general knowledge and interpretation and communication based on two genetics reports. A separate survey of UK clinical geneticists collected demographics data, laboratory reporting practice and methods used to clarify VUS pathogenicity including classification systems.ResultsOf the 155 BCS (22.5 %) who completed the survey, 12 % reported no genetics training. Ninety five percent referred patients for BRCA genetic tests, 71 % felt unsure about the clinical implications of the test reports presented here. A VUS report from a patient with a positive family history was interpreted and theoretically communicated correctly by 94 % but when presented with a different VUS report with no management guidance and negative family history, 39 % did not know how to communicate this result to the patient. Geneticists reported multiple VUS classification systems; the most commonly used was word-based in 32 %.ConclusionsA consistent and standardised format to report particularly VUS results across all diagnostic laboratories plus additional training of UK BCS will be necessary for effective mainstreaming of BRCA testing to the oncology clinic.Electronic supplementary materialThe online version of this article (doi:10.1186/s12885-015-1934-1) contains supplementary material, which is available to authorized users.

Breast cancer diagnosis using GA feature selection and Rotation Forest

Breast cancer is one of the primary causes of death among the women worldwide, and the accurate diagnosis is one of the most significant steps in breast cancer treatment. Data mining techniques can support doctors in diagnosis decision-making process. In this paper, we present different data mining techniques for diagnosis of breast cancer. Two different Wisconsin Breast Cancer datasets have been used to evaluate the system proposed in this study. The proposed system has two stages. In the first stage, in order to eliminate insignificant features, genetic algorithms are used for extraction of informative and significant features. This process reduces the computational complexity and speed up the data mining process. In the second stage, several data mining techniques are employed to make a decision for two different categories of subjects with or without breast cancer. Different individual and multiple classifier systems were used in the second stage in order to construct accurate system for breast cancer classification. The performance of the methods is evaluated using classification accuracy, area under receiver operating characteristic curves and F-measure. Results obtained with the Rotation Forest model with GA-based 14 features show the highest classification accuracy (99.48 %), and when compared with the previous works, the proposed approach reveals the enhancement in performances. Results obtained in this study have potential to open new opportunities in diagnosis of breast cancer.

Multiple comparator classifier framework for accelerometer-based fall detection and diagnostic

There are a significant number of high fall risk individuals who are susceptible to falling and sustaining severe injuries. An automatic fall detection and diagnostic system is critical for ensuring a quick response with effective medical aid based on relative information provided by the fall detection system. This article presents and evaluates an accelerometer-based multiple classifier fall detection and diagnostic system implemented on a single wearable Shimmer device for remote health monitoring. Various classifiers have been utilised within literature, however there is very little current work in combining classifiers to improve fall detection and diagnostic performance within accelerometer-based devices. The presented fall detection system utilises multiple classifiers with differing properties to significantly improve fall detection and diagnostic performance over any single classifier and majority voting system. Additionally, the presented multiple classifier system utilises comparator functions to ensure fall event consistency, where inconsistent events are outsourced to a supervisor classification function and discrimination power is considered where events with high discrimination power are evaluated to further improve the system response. The system demonstrated significant performance advantages in comparison to other classification methods, where the proposed system obtained over 99% metrics for fall detection recall, precision, accuracy and F-value responses.

Cancer classification in the genomic era: five contemporary problems.

Classification is an everyday instinct as well as a full-fledged scientific discipline. Throughout the history of medicine, disease classification is central to how we develop knowledge, make diagnosis, and assign treatment. Here, we discuss the classification of cancer and the process of categorizing cancer subtypes based on their observed clinical and biological features. Traditionally, cancer nomenclature is primarily based on organ location, e.g., “lung cancer” designates a tumor originating in lung structures. Within each organ-specific major type, finer subgroups can be defined based on patient age, cell type, histological grades, and sometimes molecular markers, e.g., hormonal receptor status in breast cancer or microsatellite instability in colorectal cancer. In the past 15+ years, high-throughput technologies have generated rich new data regarding somatic variations in DNA, RNA, protein, or epigenomic features for many cancers. These data, collected for increasingly large tumor cohorts, have provided not only new insights into the biological diversity of human cancers but also exciting opportunities to discover previously unrecognized cancer subtypes. Meanwhile, the unprecedented volume and complexity of these data pose significant challenges for biostatisticians, cancer biologists, and clinicians alike. Here, we review five related issues that represent contemporary problems in cancer taxonomy and interpretation. (1) How many cancer subtypes are there? (2) How can we evaluate the robustness of a new classification system? (3) How are classification systems affected by intratumor heterogeneity and tumor evolution? (4) How should we interpret cancer subtypes? (5) Can multiple classification systems co-exist? While related issues have existed for a long time, we will focus on those aspects that have been magnified by the recent influx of complex multi-omics data. Exploration of these problems is essential for data-driven refinement of cancer classification and the successful application of these concepts in precision medicine.

Multiple spectral similarity metrics for surface materials identification using hyperspectral data

Modern hyperspectral imaging and non-imaging spectroradiometer has the capability to acquire high-resolution spectral reflectance data required for surface materials identification and mapping. Spectral similarity metrics, due to their mathematical simplicity and insensitiveness to the number of reference labelled spectra, have been increasingly used for material mapping by labelling reflectance spectra in hyperspectral data labelling. For a particular hyperspectral data set, the accuracy of spectral labelling depends considerably upon the degree of unambiguous spectral matching achieved by the spectral similarity metric used. In this work, we propose a new methodology for quantifying spectral similarity for hyperspectral data labelling for surface materials identification. Developed adopting the multiple classifier system architecture, the proposed methodology unifies into a single framework the differential performances of eight different spectral similarity metrics for the quantification of spectral matching for surface materials. The proposed methodology has been implemented on two types of hyperspectral data viz. image (airborne hyperspectral images) and non-image (library spectra) for numerous surface materials identification. Further, the performance of the proposed methodology has been compared with the support vector machines (SVM) approach, and with all the base spectral similarity metrics. The results indicate that, for the hyperspectral images, the performance of the proposed methodology is comparable with that of the SVM. For the library spectra, the proposed methodology shows a consistently higher (increase of about 30% when compared to SVM) classification accuracy. The proposed methodology has the potential to serve as a general library search method for materials identification using hyperspectral data.

Mitigating Mimicry Attacks Against the Session Initiation Protocol

The U.S. National Academies of Science's Board on Science, Technology and Economic Policy estimates that the Internet and voice-over-IP (VoIP) communications infrastructure generates 10% of U.S. economic growth. As market forces move increasingly towards Internet and VoIP communications, there is proportional increase in telephony denial of service (TDoS) attacks. Like denial of service (DoS) attacks, TDoS attacks seek to disrupt business and commerce by directing a flood of anomalous traffic towards key communication servers. In this work, we focus on a new class of anomalous traffic that exhibits a mimicry TDoS attack. Such an attack can be launched by crafting malformed messages with small changes from normal ones. We show that such malicious messages easily bypass intrusion detection systems (IDS) and degrade the goodput of the server drastically by forcing it to parse the message looking for the needed token. Our approach is not to parse at all; instead, we use multiple classifier systems (MCS) to exploit the strength of multiple learners to predict the true class of a message with high probability (98.50% ≤ p ≤ 99.12%). We proceed systematically by first formulating an optimization problem of picking the minimum number of classifiers such that their combination yields the optimal classification performance. Next, we analytically bound the maximum performance of such a system and empirically demonstrate that it is possible to attain close to the maximum theoretical performance across varied datasets. Finally, guided by our analysis we construct an MCS appliance that demonstrates superior classification accuracy with $O(1)$ runtime complexity across varied datasets.

Current review of juvenile osteochondritis dissecans of the knee

Juvenile osteochondritis dissecans (OCD) is a common source of knee pain in the skeletally immature patient. The etiology of this common lesion is still not agreed upon, but is thought to represent repetitive microtrauma to a focal area in the knee. It can be present anywhere in the knee, but usually is found in the lateral aspect of the medial femoral condyle. It most commonly presents with pain, activity limitation, and occasional swelling. In severe cases, mechanical symptoms may be present as well. Multiple classification systems have been described. Broadly, these lesions can be graded as stable or unstable based on the level of subchondral bone involvement and the integrity of the overlying cartilage. The majority of these lesions can initially be treated nonoperatively. Surgery may be indicated with either progression of symptoms in stable lesions or with unstable lesions. Many operative procedures have shown good results. This review describes the etiology, incidence, diagnosis, imaging modalities, classification, nonoperative, and operative management of juvenile OCD lesions of the knee with a focus on the most recent literature.

A new multiple classifier system for the prediction of protein's contacts map

In this paper, we introduce FoDT, a new algorithm for the prediction of proteins contact map, one of the great challengers of the Bioinformatics. The need of more accurate predictions, aims to combining classifiers, beyond complexity increase. The proposed methodology can be considered as a set of cooperative classifiers, which employs a not trainable combination method and coverage optimization. The robustness and predictivity evaluation, with the dataset of 9th Critical Assessment of Techniques for Protein Structure Prediction, demonstrates that our algorithm can assign contacts with an average accuracy up to 58%. It performs similarly to free servers as SVM-SEQ, NNcon and LRcon, and overcoming FragHMMent. The FoDT main advantage is their capability to break down the complex process of protein folding into a collection of simple decision rules, providing a more easy and interpretable solution for the prediction of contact's maps.

Classifier and feature set ensembles for web page classification

Web page classification is an important research direction on web mining. The abundant amount of data available on the web makes it essential to develop efficient and robust models for web mining tasks. Web page classification is the process of assigning a web page to a particular predefined category based on labelled data. It serves for several other web mining tasks, such as focused web crawling, web link analysis and contextual advertising. Machine learning and data mining methods have been successfully applied for several web mining tasks, including web page classification. Multiple classifier systems are a promising research direction in machine learning, which aims to combine several classifiers by differentiating base classifiers and/or dataset distributions so that more robust classification models can be built. This paper presents a comparative analysis of four different feature selections (correlation, consistency, information gain and chi-square-based feature selection) and four different ensemble learning methods (Boosting, Bagging, Dagging and Random Subspace) based on four different base learners (naive Bayes, K-nearest neighbour algorithm, C4.5 algorithm and FURIA algorithm). The article examines the predictive performance of ensemble methods for web page classification. The experimental results indicate that feature selection and ensemble learning can enhance the predictive performance of classifiers in web page classification. For the DMOZ-50 dataset, the highest average predictive performance (88.1%) is obtained with the combination of consistency-based feature selection with AdaBoost and naive Bayes algorithms, which is a promising result for web page classification. Experimental results indicate that Bagging and Random Subspace ensemble methods and correlation-based and consistency-based feature selection methods obtain better results in terms of accuracy rates.

Dynamic Ensemble Selection Approach for Hyperspectral Image Classification With Joint Spectral and Spatial Information

Accurate generation of a land cover map using hyperspectral data is an important application of remote sensing. Multiple classifier system (MCS) is an effective tool for hyperspectral image classification. However, most of the research in MCS addressed the problem of classifier combination, while the potential of selecting classifiers dynamically is least explored for hyperspectral image classification. The goal of this paper is to assess the potential of dynamic classifier selection/dynamic ensemble selection (DCS/DES) for classification of hyperspectral images, which consists in selecting the best (subset of) optimal classifier(s) relative to each input pixel by exploiting the local information content of the image pixel. In order to have an accurate as well as computationally fast DCS/DES, we proposed a new DCS/DES framework based on extreme learning machine (ELM) regression and a new spectral-spatial classification model, which incorporates the spatial contextual information by using the Markov random field (MRF) with the proposed DES method. The proposed classification framework can be considered as a unified model to exploit the full spectral and spatial information. Classification experiments carried out on two different airborne hyperspectral images demonstrate that the proposed method yields a significant increase in the accuracy when compared to the state-of-the-art approaches.

Computing a rodent’s diary

Rodent monitoring in biomedical laboratories is a time consuming and tedious task. Several automatic solutions that rely on different types of sensors have been proposed. Computer vision provides a significantly more universal and less intrusive solution. In this article we propose a new method to detect and classify three behaviors in rodents: exploring, rearing, and static. The method uses motion history images and a multiple classifier system to detect the three behaviors under typical laboratory conditions. It is independent of the color of the rodent and of the background. The method performs equally well on short and long video sequences, achieving a success rate of 87 %.

INFFC: An iterative class noise filter based on the fusion of classifiers with noise sensitivity control

In classification, noise may deteriorate the system performance and increase the complexity of the models built. In order to mitigate its consequences, several approaches have been proposed in the literature. Among them, noise filtering, which removes noisy examples from the training data, is one of the most used techniques. This paper proposes a new noise filtering method that combines several filtering strategies in order to increase the accuracy of the classification algorithms used after the filtering process. The filtering is based on the fusion of the predictions of several classifiers used to detect the presence of noise. We translate the idea behind multiple classifier systems, where the information gathered from different models is combined, to noise filtering. In this way, we consider the combination of classifiers instead of using only one to detect noise. Additionally, the proposed method follows an iterative noise filtering scheme that allows us to avoid the usage of detected noisy examples in each new iteration of the filtering process. Finally, we introduce a noisy score to control the filtering sensitivity, in such a way that the amount of noisy examples removed in each iteration can be adapted to the necessities of the practitioner. The first two strategies (use of multiple classifiers and iterative filtering) are used to improve the filtering accuracy, whereas the last one (the noisy score) controls the level of conservation of the filter removing potentially noisy examples. The validity of the proposed method is studied in an exhaustive experimental study. We compare the new filtering method against several state-of-the-art methods to deal with datasets with class noise and study their efficacy in three classifiers with different sensitivity to noise.

Object-oriented semantic labelling of spectral–spatial LiDAR point cloud for urban land cover classification and buildings detection

The urban land cover mapping and automated extraction of building boundaries is a crucial step in generating three-dimensional city models. This study proposes an object-based point cloud labelling technique to semantically label light detection and ranging (LiDAR) data captured over an urban scene. Spectral data from multispectral images are also used to complement the geometrical information from LiDAR data. Initial object primitives are created using a modified colour-based region growing technique. Multiple classifier system is then applied on the features extracted from the segments for classification and also for reducing the subjectivity involved in the selection of classifier and improving the precision of the results. The proposed methodology produces two outputs: (i) urban land cover classes and (ii) buildings masks which are further reconstructed and vectorized into three-dimensional buildings footprints. Experiments carried out on three airborne LiDAR datasets show that the proposed technique successfully discriminates urban land covers and detect urban buildings.

Optimal construction of one-against-one classifier based on meta-learning

A commonly used strategy for solving a multi-class classification problem is to decompose the original problem into several binary subproblems. The recently proposed method, diversified one-against-one (DOAO), constructs a one-against-one classifier by selecting the best classifier for each class pair from the set of heterogeneous base classifiers. It was found to yield better classification accuracy than other one-against-one classifiers that are based on individual classification algorithms. This paper presents a novel method, called optimally diversified one-against-one (ODOAO) which is an improvement of DOAO. ODOAO is based on meta-learning, and seeks to construct a multiple classifier system where a meta-classifier effectively combines the outputs from all the heterogeneous base classifiers that are trained using various classification algorithms for every class pair. Experimental results show that ODOAO outperforms DOAO and other one-against-one based methods with statistical significance.

On the identification of sleep stages in mouse electroencephalography time-series

The automatic identification of sleep stages in electroencephalography (EEG) time-series is a long desired goal for researchers concerned with the study of sleep disorders. This paper presents advances towards achieving this goal, with particular application to EEG time-series recorded from mice. Approaches in the literature apply supervised learning classifiers, however, these do not reach the performance levels required for use within a laboratory. In this paper, detection reliability is increased, most notably in the case of REM stage identification, by naturally decomposing the problem and applying a support vector machine (SVM) based classifier to each of the EEG channels. Their outputs are integrated within a multiple classifier system. Furthermore, there exists no general consensus on the ideal choice of parameter values in such systems. Therefore, an investigation into the effects upon the classification performance is presented by varying parameters such as the epoch length; features size; number of training samples; and the method for calculating the power spectral density estimate. Finally, the results of these investigations are brought together to demonstrate the performance of the proposed classification algorithm in two cases: intra-animal classification and inter-animal classification. It is shown that, within a dataset of 10 EEG recordings, and using less than 1% of an EEG as training data, a mean classification errors of Awake 6.45%, NREM 5.82%, and REM 6.65% (with standard deviations less than 0.6%) are achieved in intra-animal analysis and, when using the equivalent of 7% of one EEG as training data, Awake 10.19%, NREM 7.75%, and REM 17.43% are achieved in inter-animal analysis (with mean standard deviations of 6.42%, 2.89%, and 9.69% respectively). A software package implementing the proposed approach will be made available through Cybula Ltd.

Fusion of hyperspectral and LIDAR data using decision template-based fuzzy multiple classifier system

Regarding to the limitations and benefits of remote sensing sensors, fusion of remote sensing data from multiple sensors such as hyperspectral and LIDAR (light detection and ranging) is effective at land cover classification. Hyperspectral images (HSI) provide a detailed description of the spectral signatures of classes, whereas LIDAR data give height detailed information. However, because of the more complexities and mixed information in LIDAR and HSI, traditional crisp classification methods could not be more efficient. In this situation, fuzzy classifiers could deliver more satisfactory results than crisp classification approaches. Also, referring to the limitation of single classifiers, multiple classifier system (MCS) may exhibit better performance in the field of multi-sensor fusion. This paper presents a fuzzy multiple classifier system for fusions of HSI and LIDAR data based on decision template (DT). After feature extraction and feature selection on each data, all selected features of both data are applied on a cube. Then classifications were performed by fuzzy k-nearest neighbour (FKNN) and fuzzy maximum likelihood (FML) on cube of features. Finally, a fuzzy decision fusion method is utilized to fuse the results of fuzzy classifiers. In order to assess fuzzy MCS proposed method, a crisp MCS based on support vector machine (SVM), KNN and maximum likelihood (ML) as crisp classifiers and naive Bayes (NB) as crisp classifier fusion method is applied on selected cube feature. A co-registered HSI and LIDAR data set from Houston of USA was available to examine the effect of proposed MCSs. Fuzzy MCS on HSI and LIDAR data provide interesting conclusions on the effectiveness and potentialities of the joint use of these two data.

Adaboost Ensemble Classifiers for Corporate Default Prediction

This study aims to show a substitute technique to corporate default prediction. Data mining techniques have been extensively applied for this task, due to its ability to notice non-linear relationships and show a good performance in presence of noisy information, as it usually happens in corporate default prediction problems. In spite of several progressive methods that have widely been proposed, this area of research is not out dated and still needs further examination. In this study, the performance of multiple classifier systems is assessed in terms of their capability to appropriately classify default and non-default Malaysian firms listed in Bursa Malaysia. Multi-stage combination classifiers provided significant improvements over the single classifiers. In addition, Adaboost shows improvement in performance over the single classifiers.

Classifier Subset Selection to construct multi-classifiers by means of estimation of distribution algorithms

This paper proposes a novel approach to select the individual classifiers to take part in a Multiple-Classifier System. Individual classifier selection is a key step in the development of multi-classifiers. Several works have shown the benefits of fusing complementary classifiers. Nevertheless, the selection of the base classifiers to be used is still an open question, and different approaches have been proposed in the literature. This work is based on the selection of the appropriate single classifiers by means of an evolutionary algorithm. Different base classifiers, which have been chosen from different classifier families, are used as candidates in order to obtain variability in the classifications given. Experimental results carried out with 20 databases from the UCI Repository show how adequate the proposed approach is; Stacked Generalization multi-classifier has been selected to perform the experimental comparisons.

A hybrid feature selection approach for the early diagnosis of Alzheimer’s disease

Objective. Recently, significant advances have been made in the early diagnosis of Alzheimer’s disease (AD) from electroencephalography (EEG). However, choosing suitable measures is a challenging task. Among other measures, frequency relative power (RP) and loss of complexity have been used with promising results. In the present study we investigate the early diagnosis of AD using synchrony measures and frequency RP on EEG signals, examining the changes found in different frequency ranges. Approach. We first explore the use of a single feature for computing the classification rate (CR), looking for the best frequency range. Then, we present a multiple feature classification system that outperforms all previous results using a feature selection strategy. These two approaches are tested in two different databases, one containing mild cognitive impairment (MCI) and healthy subjects (patients age: 71.9 ± 10.2, healthy subjects age: 71.7 ± 8.3), and the other containing Mild AD and healthy subjects (patients age: 77.6 ± 10.0; healthy subjects age: 69.4 ± 11.5). Main results. Using a single feature to compute CRs we achieve a performance of 78.33% for the MCI data set and of 97.56% for Mild AD. Results are clearly improved using the multiple feature classification, where a CR of 95% is found for the MCI data set using 11 features, and 100% for the Mild AD data set using four features. Significance. The new features selection method described in this work may be a reliable tool that could help to design a realistic system that does not require prior knowledge of a patient's status. With that aim, we explore the standardization of features for MCI and Mild AD data sets with promising results.

Magnetic resonance imaging criteria for the assessment of the rotator cuff after repair: a systematic review.

The purpose of this systematic review was to detect the reliability of the currently available magnetic resonance imaging measurements used in the evaluation of repaired rotator cuff. Search was performed using major electronic databases from their inception to February 2014. All studies reporting post-operative magnetic resonance assessment after rotator cuff repair were included. After the identification of available magnetic resonance criteria, reliability studies were further analysed. Descriptive statistics were used to summarize findings. Methodological quality was assessed using the Quality Appraisal of Reliability Studies checklist. One hundred and twenty studies were included in the review. Twenty-six different criteria were identified. Ten studies reported inter-observer reliability, and only two assessed intra-observer reliability of some of the identified criteria. Structural integrity was the most investigated criterion. The dichotomized Sugaya's classification showed the highest reliability (k = 0.80-0.91). All other criteria showed moderate to low inter-observer reliability. Tendon signal intensity and footprint coverage showed a complete discordance. Intra-observer reliability was high for the presence of structural integrity, and moderate to low for all other criteria. Methodological quality was high only for one study and moderate for three studies. Twenty-six different criteria described by multiple classification systems have been identified for the magnetic resonance assessment of rotator cuff after repair. Reliability of most of them has not been analysed yet. With the data available, only the presence of structural integrity showed good intra- and inter-observer agreement. Systematic review of descriptive and qualitative studies, Level IV.