Proposed Classification Scheme Research Articles

An Intelligent Computer-Aided Scheme for Classifying Multiple Skin Lesions

Skin diseases cases are increasing on a daily basis and are difficult to handle due to the global imbalance between skin disease patients and dermatologists. Skin diseases are among the top 5 leading cause of the worldwide disease burden. To reduce this burden, computer-aided diagnosis systems (CAD) are highly demanded. Single disease classification is the major shortcoming in the existing work. Due to the similar characteristics of skin diseases, classification of multiple skin lesions is very challenging. This research work is an extension of our existing work where a novel classification scheme is proposed for multi-class classification. The proposed classification framework can classify an input skin image into one of the six non-overlapping classes i.e., healthy, acne, eczema, psoriasis, benign and malignant melanoma. The proposed classification framework constitutes four steps, i.e., pre-processing, segmentation, feature extraction and classification. Different image processing and machine learning techniques are used to accomplish each step. 10-fold cross-validation is utilized, and experiments are performed on 1800 images. An accuracy of 94.74% was achieved using Quadratic Support Vector Machine. The proposed classification scheme can help patients in the early classification of skin lesions.

Tectorial membrane injury in adult and pediatric trauma patients: a retrospective review and proposed classification scheme.

Traumatic tectorial membrane injuries have different radiologic presentations in adult versus pediatric patients. The purpose of this study was to identify and classify the different types of tectorial membrane injuries that occur in the adult and pediatric populations. Patients who suffered tectorial membrane injury were identified retrospectively using the keywords 'tectorial membrane," "craniocervical ligament tear/injury," and "atlanto-occipital dissociation" included in radiology reports between 2012 and 2018 using Nuance mPower software. All relevant imaging studies were reviewed by two certificates of additional qualification-certified neuroradiologists. Detailed descriptions of injuries were recorded along with any relevant additional findings, including clinical history. Ten adults and six pediatric patients were identified with acute traumatic injuries of the tectorial membrane. Ninety percent of the adult patients sustained complete disruptions inferior to the clivus, or subclival, with 22% of tears at the level of the basion and 78% at the level of the odontoid tip. In contrast, 83% of pediatric patients suffered a stripping injury of the tectorial membrane located posterior to the clivus, or retroclival. Stretch injuries of the tectorial membrane were identified in 10% of adults and 17% of pediatric patients. The juvenile-type injury, which causes retroclival epidural hematoma, was determined to preferentially occur in patients less than or equal to 14years of age with a high level of statistical significance (p value = 0.0014). A classification system for tectorial membrane injuries is proposed based on this data: type 1-retroclival stripping injury (more common in pediatric patients); type 2a-subclival disruption at the basion and type 2b-subclival disruption at the odontoid (both more common in adult patients); and type 3-thinning of the tectorial membrane.

Fuzzy relevance vector machine based classification of lung nodules in computed tomography images

AbstractLung cancer is a critical disease with growing death rate, hence, the faster identification and treatment of lung cancer is essential. In medical image processing, the traditional methods like support vector machine, relevance vector machine for classifying cancer tissues are less sensitive to false data and required optimal improvement in classification accuracy. The proposed system of accurate lung cancer classification is obtained by a hybrid fuzzy relevance vector machine (FRVM) classifier with correlation negation ant colony optimization (CNACO) algorithm. This system provides enhanced accuracy and sensitivity by implementing two stages of feature extraction, image thresholding, and tumor segmentation, with a novel feature selection and tumor classification algorithm. The best features are selected by the proposed CNACO algorithm. The selected features are labeled and classified by FRVM classifier. The proposed classification scheme is validated on lung image database consortium and image database resource initiative public database and obtained accuracy of about 98.75%.

Mid-term outcome of severe tricuspid regurgitation: are there any differences according to mechanism and severity?

Patients with significant tricuspid regurgitation (TR) addressed according the new classification in torrential TR may have different prognosis compared with just severe TR patients. We sought to determine distribution and mechanism of consecutive severe TR patients, in accordance with aetiology and severity by applying the new proposed classification scheme and their long-term outcomes. Between January and December 2013, 249 patients with significant TR referred to the cardiac imaging unit (mean age 79.9 ± 10.2 years; 29.8% female) were included. Patients were divided according to aetiology in six groups, and TR severity was reclassified into severe, massive, and torrential TR. The follow-up period was of 313 ± 103 days. When considering cardiovascular mortality, patients in the massive/torrential group showed the highest number of events (P < 0.007). Patients with TR due to pulmonary diseases had the worst prognosis according to different aetiology. Noteworthy, the best predictors for the combined endpoint [cardiovascular mortality and readmission admission for heart failure (HF)] were TR severity according to the new classification [hazard ratio (HR) 2.48, 95% confidence interval (CI) 1.25-4.93] and clinical scores such as New York Heart Association classification and congestive status (HR 1.78, 95% CI 1.28-2.49; HR 2.08, 95% CI 1.06-4.06, respectively). Patients with massive/torrential TR and patients with comorbidities, especially pulmonary disease, were identified as populations at higher risk of death and readmission for HF. New classification scheme and clinical assessment may establish who may benefit the most of intensive therapeutic treatments and intervention on the tricuspid valve.

Automated Recognition of Epileptic EEG States Using a Combination of Symlet Wavelet Processing, Gradient Boosting Machine, and Grid Search Optimizer.

Automatic recognition methods for non-stationary electroencephalogram (EEG) data collected from EEG sensors play an essential role in neurological detection. The integrated approaches proposed in this study consist of Symlet wavelet processing, a gradient boosting machine, and a grid search optimizer for a three-class classification scheme for normal subjects, intermittent epilepsy, and continuous epilepsy. Fourth-order Symlet wavelets are adopted to decompose the EEG data into five frequencies sub-bands, such as gamma, beta, alpha, theta, and delta, whose statistical features were computed and used as classification features. The grid search optimizer is used to automatically find the optimal parameters for training the classifier. The classification accuracy of the gradient boosting machine was compared with that of a conventional support vector machine and a random forest classifier constructed according to previous descriptions. Multiple performance indices were used to evaluate the proposed classification scheme, which provided better classification accuracy and detection effectiveness than has been recently reported in other studies on three-class classification of EEG data.

An effective biomedical document classification scheme in support of biocuration: addressing class imbalance.

Published literature is an important source of knowledge supporting biomedical research. Given the large and increasing number of publications, automated document classification plays an important role in biomedical research. Effective biomedical document classifiers are especially needed for bio-databases, in which the information stems from many thousands of biomedical publications that curators must read in detail and annotate. In addition, biomedical document classification often amounts to identifying a small subset of relevant publications within a much larger collection of available documents. As such, addressing class imbalance is essential to a practical classifier. We present here an effective classification scheme for automatically identifying papers among a large pool of biomedical publications that contain information relevant to a specific topic, which the curators are interested in annotating. The proposed scheme is based on a meta-classification framework using cluster-based under-sampling combined with named-entity recognition and statistical feature selection strategies. We examined the performance of our method over a large imbalanced data set that was originally manually curated by the Jackson Laboratory’s Gene Expression Database (GXD). The set consists of more than 90 000 PubMed abstracts, of which about 13 000 documents are labeled as relevant to GXD while the others are not relevant. Our results, 0.72 precision, 0.80 recall and 0.75 f-measure, demonstrate that our proposed classification scheme effectively categorizes such a large data set in the face of data imbalance.

Two-layer malicious network flow detection system with sparse linear model based feature selection

The amount of malicious network traffic of enterprise systems has increased due to spreading of botnets, fuzzers, shellcodes or exploits. This malicious threatens the everyday operation of enterprises. Building classification models from this malicious traffic is an important issue. Classification models can help us to discover new types of attacks based on previously built predictive models. The most prominent attacks on accessibility in the CIA Triad are distributed denial-of-service attacks. By using denial-of-service attacks targeted at the availability of CIA triad, it is intended to block access to services for legitimate users who need to be connected to the service. Just like the Mirai cyber-attack, major service providers like Twitter and Reddit can become inaccessible by simply attacking their DNS servers. The fact that distributed denial-of-service, a rather old type of attack, is still valid today. This paper describes two-stage filtering based network traffic identification based on network flow patterns. The paper shows that the predictive performance of the malicious traffic classification model increases with the filtering of network flow. We use L1 -norm based sparse linear models for feature selection to find an optimal feature set and determine the effect of different features. We demonstrate the effectiveness and performance of the proposed scheme with graphics and tables. Simulation results validate the effectiveness of the proposed classification scheme.

Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural Networks

The interpretation of land use and land cover (LULC) is an important issue in the fields of high-resolution remote sensing (RS) image processing and land resource management. Fully training a new or existing convolutional neural network (CNN) architecture for LULC classification requires a large amount of remote sensing images. Thus, fine-tuning a pre-trained CNN for LULC detection is required. To improve the classification accuracy for high resolution remote sensing images, it is necessary to use another feature descriptor and to adopt a classifier for post-processing. A fully connected conditional random fields (FC-CRF), to use the fine-tuned CNN layers, spectral features, and fully connected pairwise potentials, is proposed for image classification of high-resolution remote sensing images. First, an existing CNN model is adopted, and the parameters of CNN are fine-tuned by training datasets. Then, the probabilities of image pixels belong to each class type are calculated. Second, we consider the spectral features and digital surface model (DSM) and combined with a support vector machine (SVM) classifier, the probabilities belong to each LULC class type are determined. Combined with the probabilities achieved by the fine-tuned CNN, new feature descriptors are built. Finally, FC-CRF are introduced to produce the classification results, whereas the unary potentials are achieved by the new feature descriptors and SVM classifier, and the pairwise potentials are achieved by the three-band RS imagery and DSM. Experimental results show that the proposed classification scheme achieves good performance when the total accuracy is about 85%.

Collaborative representation with k-nearest classes for classification

Sparse representation based classification (SRC) and collaborative representation based classification (CRC), have been widely used in many fields such as face recognition and hyperspectral image classification. All known pattern classes of the training samples are used by SRC and CRC to collaborative represent a test sample, which may be time-consuming and unnecessary. This paper proposes the use of only a part of known pattern classes to collaboratively represent a test sample. Specifically, only the first k (a pre-given constant) classes that are nearest to the sample to be tested are used. With these k-nearest classes (KNCs), SRC and CRC, each can be generalized as KNCs based SRC (KNCSRC) and KNCs based CRC(KNCCRC), and conversely, SRC and CRC can be seen as special cases of KNCSRC and KNCCRC, respectively. Weighted SRC (WSRC) and weighted CRC (WCRC) also can be generalized as KNCs based WSRC (KNCWSRC) and KNCs based WCRC (KNCWSRC), respectively. Competitive classification experiments are carried out on five databases to evaluate the proposed classification scheme. The experimental results show that, in all cases, the proposed KNCSRC and KNCCRC outperform SRC and CRC, respectively.The results also prove that the proposed KNCWCRC outperforms WCRC. In conclusion, collaborative representation with KNCs can further improve collaborative representation based classification.

Classifications in Brief: Sanders Classification of Intraarticular Fractures of the Calcaneus.

Classifications in Brief: Sanders Classification of Intraarticular Fractures of the Calcaneus.

An Inconvenient Monophyly: An Update on the Taxonomy of the Groupers (Epinephelidae)

Groupers (Epinephelidae) are a species-rich assemblage of marine reef fishes whose classification has undergone many changes across various taxonomic levels. Over the past two decades, a number of molecular phylogenetic studies have contributed significantly to clarifying the relationships among groupers; however, usage of the relationships recovered by these studies in the classification of groupers has been inconsistent. In this paper, we build upon the most comprehensive phylogeny for groupers and their allies and use it as a framework for a phylogenetic classification of the group. We correct previous errors or omissions in earlier works and hope to lay to rest any doubts regarding previously proposed classification schemes for groupers. The resulting classification scheme includes nine monophyletic genera and the subsummation of all but one previously monotypic genus.

Auxiliary Verbs in Serbo-Croatian, French, Persian, Spanish and English: A Cognitive-Semantic Approach to the Auxiliary Verb Usage and Passive Voice

Auxiliary verbs have an important influence in the way languages connect with the cognitive processes. In this study, we investigate the role of auxiliary verbs in the formation of the semantic picture we get from their usage. Furthermore, the semantic notion and its interaction with the cognitive processing are taken into account. For our goal to be more tangible and testable, we took Serbo-Croatian, Persian, Spanish, French and English for an in-depth analysis, wherefrom we proposed a classification scheme for all languages based on the behavior of their auxiliary verbs. Based on the proposed model, we investigate furthermore the passive voice in English and propose a strong explanation for the cognitive-semantic sense of the passive in English based on the cognitive duality principle. Importance of Croatian in the way that it forms an extreme pole in the proposed classification scheme is further discussed. Furthermore, it is demonstrated that Persian has a syntactic incorporation in its simple past and present perfect.

Image Segmentation Parameter Selection and Ant Colony Optimization for Date Palm Tree Detection and Mapping from Very-High-Spatial-Resolution Aerial Imagery

Accurate mapping of date palm trees is essential for their sustainable management, yield estimation, and environmental studies. In this study, we integrated geographic object-based image analysis, class-specific accuracy measures, fractional factorial design, metaheuristic feature-selection technique, and rule-based classification to detect and map date palm trees from very-high-spatial-resolution (VHSR) aerial images of two study areas. First, multiresolution segmentation was optimized through the synergy of the F1-score accuracy measure and the robust Taguchi design. Second, ant colony optimization (ACO) was adopted to select the most significant features. Out of 31 features, only 12 significant color invariants and textural features were selected. Third, based on the selected features, the rule-based classification with the aid of a decision tree algorithm was applied to extract date palm trees. The proposed methodology was developed on a subset of the first study area, and ultimately applied to the second study area to investigate its efficiency and transferability. To evaluate the proposed classification scheme, various supervised object-based algorithms, namely random forest (RF), support vector machine (SVM), and k-nearest neighbor (k-NN), were applied to the first study area. The result of image segmentation optimization demonstrated that segmentation optimization based on an integrated F1-score class-specific accuracy measure and Taguchi statistical design showed improvement compared with objective function, along with the Taguchi design. Moreover, the result of the feature selection by ACO outperformed, with almost 88% overall accuracy, several feature-selection techniques, such as chi-square, correlation-based feature selection, gain ratio, information gain, support vector machine, and principal component analysis. The integrated framework for palm tree detection outperformed RF, SVM, and k-NN classification algorithms with an overall accuracy of 91.88% and 87.03%, date palm class-specific accuracies of 0.91 and 0.89, and kappa coefficients of 0.90 and 0.85 for the first and second study areas, respectively. The proposed integrated methodology demonstrated a highly efficient and promising tool to detect and map date palm trees from VHSR aerial images.

Deep Convolutional Neural Network for Complex Wetland Classification Using Optical Remote Sensing Imagery

The synergistic use of spatial features with spectral properties of satellite images enhances thematic land cover information, which is of great significance for complex land cover mapping. Incorporating spatial features within the classification scheme have been mainly carried out by applying just low-level features, which have shown improvement in the classification result. By contrast, the application of high-level spatial features for classification of satellite imagery has been underrepresented. This study aims to address the lack of high-level features by proposing a classification framework based on convolutional neural network (CNN) to learn deep spatial features for wetland mapping using optical remote sensing data. Designing a fully trained new convolutional network is infeasible due to the limited amount of training data in most remote sensing studies. Thus, we applied fine tuning of a pre-existing CNN. Specifically, AlexNet was used for this purpose. The classification results obtained by the deep CNN were compared with those based on well-known ensemble classifiers, namely random forest (RF), to evaluate the efficiency of CNN. Experimental results demonstrated that CNN was superior to RF for complex wetland mapping even by incorporating the small number of input features (i.e., three features) for CNN compared to RF (i.e., eight features). The proposed classification scheme is the first attempt, investigating the potential of fine-tuning pre-existing CNN, for land cover mapping. It also serves as a baseline framework to facilitate further scientific research using the latest state-of-art machine learning tools for processing remote sensing data.

A hybrid cascade method for EEG classification

The classification of EEG signals is an essential step in the design of brain–computer interface. In order to optimize this step, a method for EEG classification using a cascade technique is proposed. This classification process is implemented in two steps. In the first stage, the feature extraction is accomplished using a discrete wavelet transform (DWT), followed by a feature selection via a filter method in order to select pertinent features from the original feature set. Therefore, reduced feature vectors are presented as input to the classifier. In the second step, features are extracted from the misclassified trials resulting from the first step. Then, the most relevant features are selected. Eventually, the feature vectors are classified. Two configurations of cascade classifiers are tested. In each of them, the classifier used in the first step of the classification scheme is different from the one applied in the second step. Moreover, regarding feature extraction, two cases are studied. In the first case, the same wavelet function is applied in the two steps of the classification process, whereas in the second case, two different wavelet functions are applied successively in the two steps of the classification procedure. The proposed classification scheme was applied on EEG signals of motor imagery dataset III available in BCI competition II, leading to good classification results compared to the existent methods.

Reproducibility of importance extraction methods in neural network based fMRI classification

Recent advances in machine learning allow faster training, improved performance and increased interpretability of classification techniques. Consequently, their application in neuroscience is rapidly increasing. While classification approaches have proved useful in functional magnetic resonance imaging (fMRI) studies, there are concerns regarding extraction, reproducibility and visualization of brain regions that contribute most significantly to the classification. We addressed these issues using an fMRI classification scheme based on neural networks and compared a set of methods for extraction of category-related voxel importances in three simulated and two empirical datasets. The simulation data revealed that the proposed scheme successfully detects spatially distributed and overlapping activation patterns upon successful classification. Application of the proposed classification scheme to two previously published empirical fMRI datasets revealed robust importance maps that extensively overlap with univariate maps but also provide complementary information. Our results demonstrate increased statistical power of importance maps compared to univariate approaches for both detection of overlapping patterns and patterns with weak univariate information.

Genetic and phenotypic features defining industrial relevant Lactococcus lactis, L. cremoris and L. lactis biovar. diacetylactis strains

Lactococcus lactis strains constitute one of the most important starter cultures for cheese production. In this study, a genome-wide analysis was performed including 68 available genomes of L. lactis group strains showing the existence of two species (L. lactis and L. cremoris) and two biovars (L. lactis biovar. diacetylactis and L. cremoris biovar. lactis). The proposed classification scheme revealed coherency among phenotypic (through in silico and in vivo bacterial function profiling), phylogenomic (through maximum likelihood trees) and genomic (using overall genome sequence-based parameters) approaches. Strain biodiversity for the industrial biovar. diacetylactis was also analyzed, finding they are formed by at least three variants with the CC1 clonal complex as the only one distributed worldwide. These findings and methodologies will help improve the selection of L. lactis group strains for industrial use as well as facilitate the interpretation of previous or future research studies on this diverse group of bacteria.

Interleukin-33 overexpression reflects less aggressive tumour features in large-duct type cholangiocarcinomas.

The aim of the present study was to elucidate the clinicopathological significance of interleukin (IL)-6 and IL-33 expression in intrahepatic cholangiocarcinomas (iCCAs) and perihilar cholangiocarcinomas (pCCAs). IL-6 and IL-33 mRNA expression levels were examined in iCCAs (n = 55) and pCCAs (n = 32) by the use of quantitative real-time polymerase chain reaction and a highly sensitive in-situ hybridisation protocol (RNAscope), and expression levels were correlated with clinicopathological features. According to a recently proposed classification scheme, iCCAs were separated into small-duct (n = 33) and large-duct (n = 22) types. IL-6 and IL-33 expression levels were higher in large-duct iCCAs and pCCAs than in small-duct iCCAs, and there was a positive correlation between the expression levels of these cytokines. Double in-situ hybridisation/immunostaining showed that IL-6 mRNA was expressed in actin-positive (myo)fibroblasts, whereas IL-33 mRNA was mainly produced by CD31-positive endothelial cells. With the average expression level as a cut-off point, cases were classified as IL-6high and IL-6low or IL-33high and IL-33low . In the combined cohort of large-duct iCCAs and pCCAs, IL-6high and IL-6low cholangiocarcinomas shared many features, whereas IL-33high cases had less aggressive characteristics than IL-33low cases, as shown by lower tumour marker concentrations, smaller tumour sizes, less common vascular invasion, lower pT stages, and higher lymphocyte/monocyte ratios in blood. KRAS mutations were slightly less common in IL-33high cases than in IL-33low cases (9% versus 29%; P = 0.061). The strong expression of IL-33 in tissue appeared to be an independent favourable prognostic factor. IL-33high cholangiocarcinomas may represent a unique, less aggressive carcinogenetic process of the large bile ducts.

Analysing and classifying knowledge management publications – a proposed classification scheme

PurposeThe growing number of publications on knowledge management (KM) has addressed heterogeneous topics that lack integration and classification. This article closes the classification gap by presenting a classification scheme, providing an integrated overview of KM publications.Design/methodology/approachThe development of the classification scheme follows a multistep approach. By applying a taxonomy development method, the results of a previous content analysis of 4,290 publications were processed to integrate 3,780 keywords into a classification scheme.FindingsThe classification scheme consists of 13 main categories and subcategories with six levels of detail. The scheme covers not only KM-specific keywords but also keywords from related disciplines, indicating a strong interdependence with related research domains.Research limitations/implicationsThe scheme provides a starting point for ongoing collaboration within the KM community with the aim of improving the classification results and refining the scheme to manifest the core identity.Practical implicationsThe scheme is helpful in understanding whether KM implementation activities in organisations are aligned with overall research activities and topics covered by publications.Originality/valueDeveloping a scheme based on a prior content analysis turns out to be a unique and innovative approach that has never before been done in the KM domain.

Land cover mapping in urban environments using hyperspectral APEX data: A study case in Baden, Switzerland

High spatial resolution hyperspectral imagery has shown considerable potential for deriving accurate land cover maps in urban areas. In this paper, a new classification framework for mapping land cover in urban environments using high spatial resolution hyperspectral data was proposed. The proposed classification scheme was applied to map urban land cover using APEX data in the city of Baden, Switzerland. We first used the NDWI and NDVI indices to separate the land cover in the scene into three main classes: water, vegetation and non-vegetated surface. Then we partitioned the scene into many superpixels and applied classification using a SVM separately on the vegetation and non-vegetated surfaces. Soil was classified both in vegetation and non-vegetated surface, and these two soil results were merged in the final classification map. Shadows were initially classified in shaded vegetation surfaces and shaded non-vegetated surfaces, and then they were further classified into meaningful land cover categories. Our experimental results demonstrate that the proposed classification framework is well suited for mapping land cover in urban environments using high resolution hyperspectral data. Although the proposed method performs better than traditional methods in terms of soil classification accuracy, our findings emphasize that the soil class should be interpreted with caution in urban land cover maps derived from remote sensing data, even when high spatial resolution hyperspectral data are used. Results from this study also demonstrate that although shaded surfaces are generally classified as a single category in urban environments, in high resolution hyperspectral data, the shadows can be further classified into meaningful land cover classes with an acceptable accuracy.