Binary Decision Tree Research Articles

Classification of Different Degrees of Disability Following Intracerebral Hemorrhage: A Decision Tree Analysis from VISTA-ICH Collaboration.

Prognostication following intracerebral hemorrhage (ICH) has focused on poor outcome at the expense of lumping together mild and moderate disability. We aimed to develop a novel approach at classifying a range of disability following ICH. The Virtual International Stroke Trial Archive collaboration database was searched for patients with ICH and known volume of ICH on baseline CT scans. Disability was partitioned into mild [modified Rankin Scale (mRS) at 90 days of 0-2], moderate (mRS = 3-4), and severe disabilities (mRS = 5-6). We used binary and trichotomy decision tree methodology. The data were randomly divided into training (2/3 of data) and validation (1/3 data) datasets. The area under the receiver operating characteristic curve (AUC) was used to calculate the accuracy of the decision tree model. We identified 957 patients, age 65.9 ± 12.3 years, 63.7% males, and ICH volume 22.6 ± 22.1 ml. The binary tree showed that lower ICH volume (<13.7 ml), age (<66.5 years), serum glucose (<8.95 mmol/l), and systolic blood pressure (<170 mm Hg) discriminate between mild versus moderate-to-severe disabilities with AUC of 0.79 (95% CI 0.73-0.85). Large ICH volume (>27.9 ml), older age (>69.5 years), and low Glasgow Coma Scale (<15) classify severe disability with AUC of 0.80 (95% CI 0.75-0.86). The trichotomy tree showed that ICH volume, age, and serum glucose can separate mild, moderate, and severe disability groups with AUC 0.79 (95% CI 0.71-0.87). Both the binary and trichotomy methods provide equivalent discrimination of disability outcome after ICH. The trichotomy method can classify three categories at once, whereas this action was not possible with the binary method. The trichotomy method may be of use to clinicians and trialists for classifying a range of disability in ICH.

Binary Decision Trees for Preoperative Periapical Cyst Screening Using Cone-beam Computed Tomography

IntroductionCone-beam computed tomographic (CBCT) analysis allows for 3-dimensional assessment of periradicular lesions and may facilitate preoperative periapical cyst screening. The purpose of this study was to develop and assess the predictive validity of a cyst screening method based on CBCT volumetric analysis alone or combined with designated radiologic criteria. MethodsThree independent examiners evaluated 118 presurgical CBCT scans from cases that underwent apicoectomies and had an accompanying gold standard histopathological diagnosis of either a cyst or granuloma. Lesion volume, density, and specific radiologic characteristics were assessed using specialized software. Logistic regression models with histopathological diagnosis as the dependent variable were constructed for cyst prediction, and receiver operating characteristic curves were used to assess the predictive validity of the models. A conditional inference binary decision tree based on a recursive partitioning algorithm was constructed to facilitate preoperative screening. ResultsInterobserver agreement was excellent for volume and density, but it varied from poor to good for the radiologic criteria. Volume and root displacement were strong predictors for cyst screening in all analyses. The binary decision tree classifier determined that if the volume of the lesion was >247 mm3, there was 80% probability of a cyst. If volume was <247 mm3 and root displacement was present, cyst probability was 60% (78% accuracy). ConclusionsThe good accuracy and high specificity of the decision tree classifier renders it a useful preoperative cyst screening tool that can aid in clinical decision making but not a substitute for definitive histopathological diagnosis after biopsy. Confirmatory studies are required to validate the present findings.

Quantitative distinction of the morphological characteristic of erythrocyte precursor cells with texture analysis using gray level co-occurrence matrix.

Morphological characteristics of blood cells are still qualitatively defined. So a texture analysis (Tx) method using gray level co-occurrence matrices (GLCMs; CM-Tx method) was applied to images of erythrocyte precursor cells (EPCs) for quantitatively distinguishing four types of EPC stages: proerythroblast, basophilic erythroblast, polychromatic erythroblast, and orthochromatic erythroblast. Fifty-five images of four types of EPCs were downloaded from an atlas uploaded by the Blood Cell Morphology Standardization Subcommittee (BCMSS) of the Japanese Society of Laboratory Hematology (JSLH). Using in-house programs, two types of GLCMs-(R: d=1, θ=0°) and (U: d=1, θ=270°)-and nine types of texture distinction index (TDI) were calculated with images removed outer part of cell. Three binary decision trees were sequentially divided among four types of EPC with the sum average of GLCM (U), the contrast of GLCM (R), and the sum average of GLCM (U). The average concordance rate (sensitivity) of CM-Tx method with the judgments of eleven experts in the BCMSS of the JSLH was 95.8% (87.5-100.0), and the average specificity was 97.6% (92.5-100.0). The CM-Tx method is an effective tool for quantitative distinction of EPC with their morphological features.

Clustering nominal data using unsupervised binary decision trees: Comparisons with the state of the art methods

In this work, we propose an extension of CUBT (clustering using unsupervised binary trees) to nominal data. For this purpose, we primarily use heterogeneity criteria and dissimilarity measures based on mutual information, entropy and Hamming distance. We show that for this type of data, CUBT outperforms most of the existing methods. We also provide and justify some guidelines and heuristics to tune the parameters in CUBT. Extensive comparisons are done with other well known approaches using simulations, and two examples of real datasets applications are given.

Sistema de apoyo para la toma de decisiones entermografía de glándulas mamarias

A decision tree based system with heuristic weight factors oriented to diagnosis by thermography was developed. The proposed clinical decision support system (CDSS) includes binary decision trees for statistical classification of some diseases of the mammary gland. These include weight constant factors found by the correlation of the variables involved in the development of breast disorders according to the statistical attachment of the thermal features, as well as complementary clinical information (medical history, physical parameters, genetic load and habits, among others), regarding its diagnostic incidence. The certainty of the system for inflammatory disorders is 96%, for fibrocystic mastopathy is 78 %, for fat necrosis is 62% and for breast cancer is 86%. The 14% error is associated to microcalcifications that with the reported thermographic technique it is not possible to differentiate. Thus, it is possible to catalog the risk of a patient having a condition according to the results obtained after conducting a thermal test. Then, with the CDSS the possibility of applying early detection tests in the population suffering from contraindications to perform the others is improved, in addition it allows a comprehensive diagnosis for other patients.

Classification of Broadcast News Audio Data Employing Binary Decision Architecture

A novel binary decision architecture (BDA) for broadcast news audio classification task is presented in this paper. The idea of developing such architecture came from the fact that the appropriate combination of multiple binary classifiers for two-class discrimination problem can reduce a miss-classification error without rapid increase in computational complexity. The core element of classification architecture is represented by a binary decision (BD) algorithm that performs discrimination between each pair of acoustic classes, utilizing two types of decision functions. The first one is represented by a simple rule-based approach in which the final decision is made according to the value of selected discrimination parameter. The main advantage of this solution is relatively low processing time needed for classification of all acoustic classes. The cost for that is low classification accuracy. The second one employs support vector machine (SVM) classifier. In this case, the overall classification accuracy is conditioned by finding the optimal parameters for decision function resulting in higher computational complexity and better classification performance. The final form of proposed BDA is created by combining four BD discriminators supplemented by decision table. The effectiveness of proposed BDA, utilizing rule-based approach and the SVM classifier, is compared with two most popular strategies for multiclass classification, namely the binary decision trees (BDT) and the One-Against-One SVM (OAOSVM). Experimental results show that the proposed classification architecture can decrease the overall classification error in comparison with the BDT architecture. On the contrary, an optimization technique for selecting the optimal set of training data is needed in order to overcome the OAOSVM.

Hyperspectral Image Classification With Canonical Correlation Forests

Multiple classifier systems or ensemble learning is an effective tool for providing accurate classification results of hyperspectral remote sensing images. Two well-known ensemble learning classifiers for hyperspectral data are random forest (RF) and rotation forest (RoF). In this paper, we proposed to use a novel decision tree (DT) ensemble method, namely, canonical correlation forest (CCF). More specifically, several individual canonical correlation trees (CCTs) that are binary DTs, which use canonical correlation components for the hyperplane splitting, are used to construct the CCF. Additionally, we adopt the projection bootstrap technique in CCF, in which the full spectral bands are retained for split selection in the projected space. The techniques aforementioned allow the CCF to improve the accuracy of member classifiers and diversity within the ensemble. Furthermore, the CCF is extended to the spectral-spatial frameworks that incorporate Markov random fields, extended multiattribute profiles (EMAPs), and the ensemble of independent component analysis and rolling guidance filter (E-ICA-RGF). Experimental results on six hyperspectral data sets are used to indicate the comparative effectiveness of the proposed method, in terms of accuracy and computational complexity, compared with RF and RoF, and it turns out that CCF is a promising approach for hyperspectral image classification not only with spectral information but also in the spectral-spatial frameworks.

Classification Tree Extraction from Trained Artificial Neural Networks

Recent advances in neural networks design and training provoked the 2nd artificial neural networks (ANN) renaissance. In many cases classification decision made by trained fully connected neural nets is better than that acquired by models like C4.5 or C5.01,2. But in contrast to decision trees, ANN models are “black boxes”, i.e., it is impossible to understand how classification decision is made. In many areas it is critical and even obligate to understand how a model performs classification thus rendering ANN usage as obsolete. Recently, some researchers have proposed and described separate steps that would allow extracting knowledge from a trained multi-layered fully connected sigmoidal neural network. This process involves several steps such as trained network training, pruning and knowledge extraction. This paper provides an overview of all the aforementioned steps, as well as describes how a knowledge extraction system can be built. We describe our Neural Network Knowledge eXtraction (NNKX) system and provide experimental results of rule extraction from the trained multi-layered feed-forward sigmoidal artificial neural network in the form of binary classification decision trees. The results obtained suggest that extracted decision trees have good classification accuracy and sizes comparable to C4.5 trees and even overcoming them in some cases. Thus the proposed system can be successfully applied to better understand and validate ANN models. We provide link to source code repository with the implementation of described system.

Expert system in medicine and its application on pulmonary diseases

Objective: Our aim is to develop a medical expert system for pulmonary diseases providing practitioners and medical students with the advantages of improving their ability, minimizing the error and cost in diagnosing and developing their medical knowledge. Material and Methods: CLIPS is chosen as a programming environment for this work. A respiratory disease binary decision tree which helps us to create the system database which includes twenty-eight diseases is formed for the inference engine of this program. Results: The evaluation of this program is based on hundred and eighty-nine patients’ data each is classified into three data types. These are patient’s history and physical findings, radiological data and laboratory data. The combination of them shapes four different data sets for each patient. The diagnosing result for each data set of each patient is compared with diagnosing of gold standards. If both results indicates the same disease this operation of the program is assumed as “accurate”, otherwise as “error”. These operations for the considered each data set are repeated for all patients’ data. The total number of accurate diagnosis is divided by the number of all patients and these accuracy rates are respectively 64.02%, 71.43%, 82.54% and 96.83%. Conclusion: We can conclude that the accuracy of the system is enhanced with the increasing total number and type of data for each patient. Finally, further improvement on the performance and accuracy of the system may be obtained by designing the program with the self-learning ability.

Expert system in medicine and its application for pulmonary diseases

Objective:&nbsp;Our aim is to develop a medical expert system for pulmonary diseases providing practitioners and medical students with the advantages of improving their ability, minimizing the error and cost in diagnosing and developing their medical knowledge. Material and Methods:&nbsp;CLIPS&nbsp; has been chosen as a programming environment for this study. A respiratory disease binary decision tree which helps us to create the system database which includes twenty-eight diseases is formed for the inference engine of this program. Results:&nbsp;The evaluation of this program is based on hundred and eighty-nine patients’ data each is classified into three data types. These are patient’s history and physical findings, radiological data and laboratory data. The combination of them shapes four different data sets for each patient. The diagnosing result for each data set of each patient is compared with diagnosing of gold standards. If both results indicates the same disease this operation of the program is assumed as “accurate”, otherwise as “error”. These operations for the considered each data set are repeated for all patients’ data. The total number of accurate diagnosis is divided by the number of all patients and these accuracy rates are respectively 64.02%, 71.43%, 82.54% and 96.83%. Conclusion:&nbsp;We can conclude that the accuracy of the system is enhanced with the increasing total number and type of data for each patient. Finally, further improvement on the performance and accuracy of the system may be obtained by designing the program with the self-learning ability.

Real-Time Classification of Patients with Balance Disorders vs. Normal Subjects Using a Low-Cost Small Wireless Wearable Gait Sensor.

Gait analysis using wearable wireless sensors can be an economical, convenient and effective way to provide diagnostic and clinical information for various health-related issues. In this work, our custom designed low-cost wireless gait analysis sensor that contains a basic inertial measurement unit (IMU) was used to collect the gait data for four patients diagnosed with balance disorders and additionally three normal subjects, each performing the Dynamic Gait Index (DGI) tests while wearing the custom wireless gait analysis sensor (WGAS). The small WGAS includes a tri-axial accelerometer integrated circuit (IC), two gyroscopes ICs and a Texas Instruments (TI) MSP430 microcontroller and is worn by each subject at the T4 position during the DGI tests. The raw gait data are wirelessly transmitted from the WGAS to a near-by PC for real-time gait data collection and analysis. In order to perform successful classification of patients vs. normal subjects, we used several different classification algorithms, such as the back propagation artificial neural network (BP-ANN), support vector machine (SVM), k-nearest neighbors (KNN) and binary decision trees (BDT), based on features extracted from the raw gait data of the gyroscopes and accelerometers. When the range was used as the input feature, the overall classification accuracy obtained is 100% with BP-ANN, 98% with SVM, 96% with KNN and 94% using BDT. Similar high classification accuracy results were also achieved when the standard deviation or other values were used as input features to these classifiers. These results show that gait data collected from our very low-cost wearable wireless gait sensor can effectively differentiate patients with balance disorders from normal subjects in real time using various classifiers, the success of which may eventually lead to accurate and objective diagnosis of abnormal human gaits and their underlying etiologies in the future, as more patient data are being collected.

Automatic detection and mapping of double potentials in intracardiac electrograms

Abstract Catheter ablation has become a very efficient strategy to terminate sustained cardiac arrhythmias like atrial flutter (AFlut). Identification of the optimal ablation spot, however, often proves difficult when scar from previous ablations is present. Although the application of electro-anatomical mapping systems allows to record thousands of intracardiac electrograms (EGMs) from each atrium, state-of-the-art techniques provide limited options for automatic signal processing. Goal of the presented research was the development of an algorithm to detect EGMs that present double potentials (DPs), as these often indicate functional or anatomical lines of block for cardiac excitation. Using an annotated database, we developed several features based on the morphological descriptors of DPs. These were used to train a binary decision tree which was able to detect DPs with a correct rate of over 90%.

SHIELD: Scalable Homomorphic Implementation of Encrypted Data-Classifiers

Homomorphic encryption (HE) systems enable computations on encrypted data, without decrypting and without knowledge of the secret key. In this work, we describe an optimized Ring Learning With Errors (RLWE) based implementation of a variant of the HE system recently proposed by Gentry, Sahai and Waters (GSW). Although this system was widely believed to be less efficient than its contemporaries, we demonstrate quite the opposite behavior for a large class of applications. We first highlight and carefully exploit the algebraic features of the system to achieve significant speedup over the state-of-the-art HE implementation, namely the IBM homomorphic encryption library (HElib). We introduce several optimizations on top of our HE implementation, and use the resulting scheme to construct a homomorphic Bayesian spam filter, secure multiple keyword search, and a homomorphic evaluator for binary decision trees. Our results show a factor of $10\times$ improvement in performance (under the same security settings and CPU platforms) compared to IBM HElib for these applications. Our system is built to be easily portable to GPUs (unlike IBM HElib) which results in an additional speedup of up to a factor of $103.5\times$ to offer an overall speedup of $1{,}035\times$ .

Statistical classification techniques in high energy physics (SDDT algorithm)

We present our proposal of the supervised binary divergence decision tree with nested separation method based on the generalized linear models. A key insight we provide is the clustering driven only by a few selected physical variables. The proper selection consists of the variables achieving the maximal divergence measure between two different classes. Further, we apply our method to Monte Carlo simulations of physics processes corresponding to a data sample of top quark-antiquark pair candidate events in the lepton+jets decay channel. The data sample is produced in pp̅ collisions at √S = 1.96 TeV. It corresponds to an integrated luminosity of 9.7 fb-1 recorded with the D0 detector during Run II of the Fermilab Tevatron Collider. The efficiency of our algorithm achieves 90% AUC in separating signal from background. We also briefly deal with the modification of statistical tests applicable to weighted data sets in order to test homogeneity of the Monte Carlo simulations and measured data. The justification of these modified tests is proposed through the divergence tests.

Pixel-based skin segmentation in psoriasis images.

In this paper, we present a detailed comparison study of skin segmentation methods for psoriasis images. Different techniques are modified and then applied to a set of psoriasis images acquired from the Royal Melbourne Hospital, Melbourne, Australia, with aim of finding the best technique suited for application to psoriasis images. We investigate the effect of different colour transformations on skin detection performance. In this respect, explicit skin thresholding is evaluated with three different decision boundaries (CbCr, HS and rgHSV). Histogram-based Bayesian classifier is applied to extract skin probability maps (SPMs) for different colour channels. This is then followed by using different approaches to find a binary skin map (SM) image from the SPMs. The approaches used include binary decision tree (DT) and Otsu's thresholding. Finally, a set of morphological operations are implemented to refine the resulted SM image. The paper provides detailed analysis and comparison of the performance of the Bayesian classifier in five different colour spaces (YCbCr, HSV, RGB, XYZ and CIELab). The results show that histogram-based Bayesian classifier is more effective than explicit thresholding, when applied to psoriasis images. It is also found that decision boundary CbCr outperforms HS and rgHSV. Another finding is that the SPMs of Cb, Cr, H and B-CIELab colour bands yield the best SMs for psoriasis images. In this study, we used a set of 100 psoriasis images for training and testing the presented methods. True Positive (TP) and True Negative (TN) are used as statistical evaluation measures.

Signal classification and software–hardware implementation of digital filter banks based on field-programmable gate arrays and compute unified device architecture

The paper is devoted to handling wideband monitoring tasks by discrete Fourier transform (DFT) modulated filter banks. Filter bank implementation is considered using CPU (Central Processing Unit) and CUDA (Compute Unified Device Architecture) based on GPUs (Graphics Processing Units). We show that CUDA is more efficient for big signal sets due to low temporal and computational costs. The paper also discusses signal classification in filter bank channels for different signal-to-noise ratios using binary decision trees (with the iterative Adaboost procedure) and neural networks. The total classification error in our experiments does not exceed 10%. The results can be extended and applied to hydroacoustic monitoring tasks.

Scalable Multi-Match Packet Classification Using TCAM and SRAM

Packet classification is an enabling technology for various network services. Fast single-match packet classification can be achieved by using ternary content addressable memory (TCAM) because of the superior speed performance. TCAM has some drawbacks including incapability to store arbitrary ranges, confined TCAM capacity and limited choices of entry lengths. Moreover, TCAM only reports the first matching entry to impose a limitation on supporting multi-match packet classification, which requires all matching rules. The existing algorithms deal with the issues of TCAM-based multi-match packet classification by burdening TCAM with extra entries and/or accesses. In this work, we offload the overhead of TCAM to static random access memory (SRAM) to achieve efficient multi-match packet classification. Our scheme synthesizes TCAM compatible entries by using binary decision trees and employs SRAM for further comparisons. Each synthesized entry can be stored in one TCAM entry to significantly reduce TCAM consumption and fulfill low power consumption. The experimental results show that our scheme can lower the demand of TCAM to improve both search latency and energy efficiency. The scalability of TCAM-based multi-match packet classification can thus be improved drastically.

Iran’s Medical Tourism Development Model in Cardiac Surgery

Introduction: Since medical tourism is considered as an incremental activity in this sector and proper infrastructures in country to make medical tourism are lacking, announcement by authorities to provide perquisites of medical tourism to make the first clinic hotel and health town are necessary for the purpose of developing medical tourism in Iran, all of these side issues should be investigated. Methods: Cardiovascular diseases are very common because of pollution and industrial development. In this research, by extracting related studies on medical tourism and its localization, using Hospitals data bank and questionnaire of 640 medical tourists, only 528 of them were approved and finally by using IBM SPSS modular 14.1 software decision tree in data algorithm, efficiency and purity level were obtained. The method for data preprocess step is utilized to extract the best model. Two preprocess steps are deleting useless and correlated features, because data should be prepared until model has the least error. Results: Among the examined algorithm including C&RT, CHAID and Multiple Linear Regression, it was shown that C&RT has the optimal results. The results obtained from this research indicate that C&RT binary decision tree has the smallest error value (0.078) and the greatest accuracy value (0.922). Conclusion: In this research, first effective factors on heart medical tourism were investigated with the help of experts in this field, then C&RT and CHASID models using Clementine software and multiple linear regression variables were compared and ranked. With respect to this algorithm, personnel behavior, social security and communication variables are respectively the most important factors for medical tourist attraction.

Quantitative Structure–Property Relationship Models for Recognizing Metal Organic Frameworks (MOFs) with High CO2 Working Capacity and CO2/CH4 Selectivity for Methane Purification

Metal‐organic frameworks (MOFs) can theoretically yield a nearly infinite number of nanoporous materials, which represents a combinatorial design challenge that demands computational tools rather than experimental trial‐and‐error. Here we report Quantitative Structure–Property Relationship (QSPR) models to identify high‐performing MOFs for methane purification solely using geometrical features. The CO2 working capacity and CO2/CH4 selectivity of ca. 320,000 hypothetical MOF structures was computed at conditions relevant to natural gas purification using grand canonical Monte‐Carlo (GCMC) simulations. Using 32,500 MOF structures we calibrated binary decision tree (DT) and support vector machine (SVM) models that can accurately identify high‐performing MOFs based on their pore size, void fraction and surface area. DT models yielded guidelines of pore size, void fraction and surface area for designing high‐performing materials. The SVM machine learning classifiers could be used to quickly pre‐screen MOFs, such that compute intensive GCMC simulations are not performed on all structures. The SVM classifiers were tested on ca. 290,000 MOFs that were not part of the training set and could correctly identify up to 90 % of high‐performing MOFs while only flagging a fraction of the MOFs for more rigorous screening. QSPR models constitute efficient computational tools for the virtual screening of large structural libraries and provide rational design rules for the discovery of sorbents for methane purification.

Negating a Generalized Cut Sequence: Bridging the Gap Between Dynamic Fault Trees Quantification and Sum of Disjoint Products Methods

Dynamic fault trees (DFTs) are powerful tools to model industrial systems having dynamic failure mechanisms, such as sequence- and function-dependent failure behaviors. Yet for large and complex DFTs, their quantitative analyses are still of great challenges. Up to now, many researchers have presented several approaches to deal with this problem, and among which, the sum of disjoint products (SDP) methods, such as dynamic binary decision tree, sequential binary decision diagram (SBDD), and improved SBDD, have proven to be an efficient way. In SDP methods, negating a generalized cut sequence is an unavoidable task. Yet, for a complex cut sequence expression where normal, cold and warm spares basic events coexist, its negating operation is still difficult and needs to be further studied. In this paper, based on De Morgan theorem, improved explicit formulas for negating a generalized cut sequences are presented. The new concept of universal set of basic event and its operating rules are proposed to deduce the simplified expressions of general enforcing occurring cut sequences and warm spares occurring cut sequences. To validate the presented approaches, a typical system DFT is analyzed. The results indicate the reasonability and effectiveness of the improved negating formulas. Copyright © 2016 John Wiley & Sons, Ltd.