Automatic diagnosis of diabetes using adaptive neuro‐fuzzy inference systems

Abstract: A new approach based on an adaptive neuro‐fuzzy inference system (ANFIS) is presented for diagnosis of diabetes diseases. The Pima Indians diabetes data set contains records of patients with known diagnosis. The ANFIS classifiers learn how to differentiate a new case in the domain by being given a training set of such records. The ANFIS classifier is used to detect diabetes diseases when eight features defining diabetes indications are used as inputs. The proposed ANFIS model combines neural network adaptive capabilities and the fuzzy logic qualitative approach. The conclusions concerning the impacts of features on the diagnosis of diabetes disease are obtained through analysis of the ANFIS. The performance of the ANFIS model is evaluated in terms of training performances and classification accuracies and the results confirm that the proposed ANFIS model has potential in detecting diabetes diseases.     

Estimating electromyography responses using an adaptive neuro‐fuzzy inference system with subtractive clustering

This study aimed to develop an adaptive neuro‐fuzzy inference system (ANFIS) approach to estimate the normalized electromyography (NEMG) responses, where the independent variables are demographic variables including population, gender, ethnicity, age, height, weight, posture, and muscle groups. The study groups comprised 75 US‐based (54 males and 21 females) and 10 Japan‐based (all males) automobile assembly workers. A total of 65 inputs and 1 output reflecting the NEMG values were considered at the beginning. After correlating analysis results, a total of 35 significant predictors were considered for both ANFIS and regression models. The data were partitioned into two datasets, one for training (70% of all data) and one for validation (30% of all data). In addition to a soft‐computing approach, a multiple linear regression (MLR) analysis was also performed to evaluate whether or not the ANFIS approach showed superior predictive performance compared to a classical statistical approach. According to the performance comparison, ANFIS had better predictive accuracy than MLR, as demonstrated by the experimental results. Overall, this study demonstrates that the ANFIS approach can predict normalized EMG responses according to subjects’ demographic variables, posture, and muscle groups.     

Classification of congestive heart failure with different New York Heart Association functional classes based on heart rate variability indices and machine learning

This study aims to evaluate the effect of heart rate variability (HRV) indices on the New York Heart Association (NYHA) classification of patients with congestive heart failure and to test the effectiveness of different machine learning algorithms. Twenty‐nine long‐term RR interval recordings from subjects (aged 34 to 79) with congestive heart failure (NYHA classes I, II, and III) in MIT‐BIH Database were studied. We firstly removed the unreasonable RR intervals and segment the RR recordings with a 300‐RR interval length window. Then the multiple HRV indexes were calculated for each RR segment. Support vector machine (SVM) and classification and regression tree (CART) methods were then separately used to distinguish patients with different NYHA classes based on the selected HRV indices. Receiver operating characteristic curve analysis was finally employed as the evaluation indicator to compare the performance of the two classifiers. The SVM classifier achieved accuracy, sensitivity, and specificity of 84.0%, 71.2%, and 83.4%, respectively, whereas the CART classifier achieved 81.4%, 66.5%, and 81.6%, respectively. The area under the curve of receiver operating characteristic for the two classifiers was 86.4% and 84.7%, respectively. It is possible for accurately classifying the NYHA functional classes I, II, and III when using the combination of HRV indices and machine learning algorithms. The SVM classifier performed better in classification than the CART classifier using the same HRV indices.     

Subtractive clustering attribute weighting (SCAW) to discriminate the traffic accidents on Konya-Afyonkarahisar highway in Turkey with the help of GIS: A case study

A case study including the discrimination of traffic accidents as accident free and accident cases on Konya-Afyonkarahisar highway in Turkey using the proposed hybrid method based on combining of a new data preprocessing method called subtractive clustering attribute weighting (SCAW) and classifier algorithms with the help of Geographical Information System (GIS) technology has been conducted. In order to improve the discrimination of classifier algorithms including artificial neural network (ANN), adaptive network based fuzzy inference system (ANFIS), support vector machine, and decision tree, using data preprocessing need in solution of these kinds of problems (traffic accident case study). So, we have proposed a novel data preprocessing method called subtractive clustering attribute weighting (SCAW) and combined with classifier algorithms. In this study, the experimental data has been obtained by means of using GIS. The obtained GIS attributes are day, temperature, humidity, weather conditions, and month of occurred accident. To evaluate the performance of the proposed hybrid method, the classification accuracy, sensitivity and specificity values have been used. The experimental obtained results are 53.93%, 52.25%, and 38.76% classification successes using alone ANN, ANFIS, and SVM with RBF kernel type, respectively. As for the proposed hybrid method, the classification accuracies of 67.98%, 70.22%, and 61.24% have been obtained using the combination of SCAW with ANN, the combination of SCAW with SVM (radial basis function (RBF) kernel type), and the combination of SCAW with ANFIS, respectively. The proposed SCAW method with the combination of classifier algorithms has been achieved the very promising results in the discrimination of traffic accidents.     

Security‐level classification for confidential documents by using adaptive neuro‐fuzzy inference systems

The security‐level detection of a confidential document is a vital task for organizations to protect their confidential information. Diverse classification rules and techniques are being applied by human experts. Increasing number of confidential information in organizations is making difficult to classify all the documents carefully with human effort. The recommended frameworks in this study classify the internal documents of TUBITAK UEKAE (National Research Institute of Electronics and Cryptology of Turkey) by using classification algorithms naïve Bayes, support vector machines (SVMs) and adaptive neuro‐fuzzy inference systems (ANFISs). A hybrid approach involving support vector classifiers and adaptive neuro‐fuzzy classifiers exposes the most successful accuracy rates of expert system classification. This study also states preprocessing tasks required for document classification with natural language processing. To represent term–document relations, a recommended metric TF‐IDF was chosen to construct a weight matrix. Agglutinative nature of Turkish documents is handled by Turkish stemming algorithms. At the end of the article, some experimental results and success metrics are projected with accuracy rates and receiver operating characteristic (ROC) curves.     

An improved approach to medical data sets classification: artificial immune recognition system with fuzzy resource allocation mechanism

Abstract: The artificial immune recognition system (AIRS) has been shown to be an efficient approach to tackling a variety of problems such as machine learning benchmark problems and medical classification problems. In this study, the resource allocation mechanism of AIRS was replaced with a new one based on fuzzy logic. The new system, named Fuzzy-AIRS, was used as a classifier in the classification of three well-known medical data sets, the Wisconsin breast cancer data set (WBCD), the Pima Indians diabetes data set and the ECG arrhythmia data set. The performance of the Fuzzy-AIRS algorithm was tested for classification accuracy, sensitivity and specificity values, confusion matrix, computation time and receiver operating characteristic curves. Also, the AIRS and Fuzzy-AIRS algorithms were compared with respect to the amount of resources required in the execution of the algorithm. The highest classification accuracy obtained from applying the AIRS and Fuzzy-AIRS algorithms using 10-fold cross-validation was, respectively, 98.53% and 99.00% for classification of WBCD; 79.22% and 84.42% for classification of the Pima Indians diabetes data set; and 100% and 92.86% for classification of the ECG arrhythmia data set. Hence, these results show that Fuzzy-AIRS can be used as an effective classifier for medical problems.     

An adaptive neuro‐fuzzy system for stock portfolio analysis

We propose an adaptive neuro‐fuzzy inference system (ANFIS) for stock portfolio return prediction. Previous work has shown that portfolio optimization can be improved by using predicted stock earnings rather than historical earnings. We show that predicted portfolio returns can be improved by using ANFIS and taking as input a variety of technical and fundamental attributes about various indices of the stock market. To generate membership functions, we use a robust noise rejection‐clustering algorithm. The neuro‐fuzzy model is tested on portfolios constituted from the Tehran Stock Exchange. In our experiments, the proposed method performs better in predicting the portfolio return than the classical Markowitz portfolio optimization method, a multiple regression, a neural network, and the Sugeno–Yasukawa method. © 2010 Wiley Periodicals, Inc.     

Comparison of different classifier algorithms for diagnosing macular and optic nerve diseases

Abstract: The aim of this research was to compare classifier algorithms including the C4.5 decision tree classifier, the least squares support vector machine (LS-SVM) and the artificial immune recognition system (AIRS) for diagnosing macular and optic nerve diseases from pattern electroretinography signals. The pattern electroretinography signals were obtained by electrophysiological testing devices from 106 subjects who were optic nerve and macular disease subjects. In order to show the test performance of the classifier algorithms, the classification accuracy, receiver operating characteristic curves, sensitivity and specificity values, confusion matrix and 10-fold cross-validation have been used. The classification results obtained are 85.9%, 100% and 81.82% for the C4.5 decision tree classifier, the LS-SVM classifier and the AIRS classifier respectively using 10-fold cross-validation. It is shown that the LS-SVM classifier is a robust and effective classifier system for the determination of macular and optic nerve diseases.     

Automatic determination of traffic accidents based on KMC-based attribute weighting

In this study, the traffic accidents recognizing risk factors related to the environmental (climatological) conditions that are associated with motor vehicles accidents on the Konya-Afyonkarahisar highway with the aid of Geographical Information Systems (GIS) have been determined using the combination of K-means clustering (KMC)-based attribute weighting (KMCAW) and classifier algorithms including artificial neural network (ANN) and adaptive network-based fuzzy inference system (ANFIS). The dynamic segmentation process in ArcGIS9.0 from the traffic accident reports recorded by District Traffic Agency has identified the locations of the motor vehicle accidents. The attributes obtained from this system are day, temperature, humidity, weather conditions, and month of occurred traffic accidents. The traffic accident dataset comprises five attributes (day, temperature, humidity, weather conditions, and month of occurred traffic accidents) and 358 observations including 179 without accident and 179 with accident. The proposed comprises two stages. In the first stage, the all attributes of dataset have been weighted using KMCAW method. The aims of this weighting method are both to increase the classification performance of used classifier algorithm and to transform from linearly non-separable traffic accidents dataset to a linearly separable dataset. In the second stage, after weighting process, ANN and ANFIS classifier algorithms have been separately used to determine the case of traffic accidents as with accident or without accident. In order to evaluate the performance of proposed method, the classification accuracy, sensitivity, specificity and area under the ROC (Receiver Operating Characteristic) curves (AUC) values have been used. While ANN and ANFIS classifiers obtained the overall prediction accuracies of 53.93 and 38.76%, respectively, the combination of KMCAW and ANN and the combination of KMCAW and ANFIS achieved the overall prediction accuracies of 74.15 and 55.06% on the prediction of traffic accidents. The experimental results have demonstrated that the proposed attribute weighting method called KMCAW is a robust and effective data pre-processing method in the prediction of traffic accidents on Konya-Afyonkarahisar highway in Turkey.     

Interval type-2 fuzzy logic based multiclass ANFIS algorithm for real-time EEG based movement control of a robot arm

Brain–computer interfacing is an emerging field of research where signals extracted from the human brain are used for decision making and generation of control signals. Selection of the right classifier to detect the mental states from electroencephalography (EEG) signal is an open area of research because of the signal’s non-stationary and Ergodic nature. Though neural network based classifiers, like Adaptive Neural Fuzzy Inference System (ANFIS), act efficiently, to deal with the uncertainties involved in EEG signals, we have introduced interval type-2 fuzzy system in the fray to improve its uncertainty handling. Also, real-time scenarios require a classifier to detect more than two mental states. Thus, a multi-class discriminating algorithm based on the fusion of interval type-2 fuzzy logic and ANFIS, is introduced in this paper. Two variants of this algorithm have been developed on the basis of One-Vs-All and One-Vs-One methods. Both the variants have been tested on an experiment involving the real-time control of robot arm, where both the variants of the proposed classifier, produces an average success rate of reaching a target to 65% and 70% respectively. The result shows the competitiveness of our algorithm over other standard ones in the domain of non-stationary and uncertain signal data classification.     

Fuzzy cluster based neural network classifier for classifying breast tumors in ultrasound images

The performance of supervised classification algorithms is highly dependent on the quality of training data. Ambiguous training patterns may misguide the classifier leading to poor classification performance. Further, the manual exploration of class labels is an expensive and time consuming process. An automatic method is needed to identify noisy samples in the training data to improve the decision making process. This article presents a new classification technique by combining an unsupervised learning technique (i.e. fuzzy c-means clustering (FCM)) and supervised learning technique (i.e. back-propagation artificial neural network (BPANN)) to categorize benign and malignant tumors in breast ultrasound images. Unsupervised learning is employed to identify ambiguous examples in the training data. Experiments were conducted on 178 B-mode breast ultrasound images containing 88 benign and 90 malignant cases on MATLAB® software platform. A total of 457 features were extracted from ultrasound images followed by feature selection to determine the most significant features. Accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUC) and Mathew's correlation coefficient (MCC) were used to access the performance of different classifiers. The result shows that the proposed approach achieves classification accuracy of 95.862% when all the 457 features were used for classification. However, the accuracy is reduced to 94.138% when only 19 most relevant features selected by multi-criterion feature selection approach were used for classification. The results were discussed in light of some recently reported studies. The empirical results suggest that eliminating doubtful training examples can improve the decision making performance of expert systems. The proposed approach show promising results and need further evaluation in other applications of expert and intelligent systems.     

An evolutionary‐based adaptive neuro‐fuzzy inference system for intelligent short‐term load forecasting

The continuing growth in size and complexity of electric power systems requires the development of applicable load forecasting models to estimate the future electrical energy demands accurately. This paper presents a novel load forecasting approach called genetic‐based adaptive neuro‐fuzzy inference system (GBANFIS) to construct short‐term load forecasting expert systems and controllers. At the first stage, all records of data are searched by a novel genetic algorithm (GA) to find the most suitable feature of inputs to construct the model. Then, determined inputs are fed into the adaptive neuro‐fuzzy inference system to evolve the initial knowledge‐base of the expert system. Finally, the initial knowledge‐base is searched by another robust GA to induce a better cooperation among the rules by rule weight derivation and rule selection mechanisms. We show the superiority and applicability of our approach by applying it to the Iranian monthly electrical energy demand problem and comparing it with the most frequently adopted approaches in this field. Results indicate that GBANFIS outperforms its rival approaches and is a promising tool for dealing with short‐term load forecasting problems.     

A comprehensive search for expert classification methods in disease diagnosis and prediction

Healthcare data analysis is currently a challenging and crucial research issue for the development of a robust disease diagnosis and prediction system. Many specific and a few common methods have been discussed in the literature for healthcare data classification. The present study implements 32 classification methods of six categories (Bayes, function‐based, lazy, meta, rule‐based, and tree‐based) with the objective of searching the best and common categories and methods in healthcare data mining. The performance of each classification method has been evaluated based on analysis time, classification accuracy, precision, recall, F‐measure, area under the receiver operating characteristic curve, root mean square error, kappa coefficient, Kulczynski's measure, and Fowlkes–Mallows index and compared with more than 90 classification methods used in past studies. Seventeen healthcare datasets related to thyroid, cancer, skin disease, heart disease, hepatitis, lymphography, audiology, diabetes, surgery, arrhythmia, postsurvival, liver, and tumour have been used in the performance assessment of the classification methods. The tree‐based classification methods have a better performance (with an average classification accuracy of 79.92% and maximum accuracy of 99.50%; an analysis time of 3.91 s for the logistic model tree classifier) than the other methods. Furthermore, the association of datasets and classification methods has been discussed.     

Computer-aided detection of lung nodules based on decision fusion techniques

We adopted decision fusion techniques to develop a computer-aided detection (CAD) system for automatic detection of pulmonary nodules in low-dose CT images. Two distinct phases, aimed, respectively, at detecting volumes of interests (VOIs) within the CT scan, and at classifying VOIs into nodules and non-nodules, were considered. Three algorithms, namely thresholding, region growing and robust fuzzy clustering, were used as VOI detectors. For the classification phase, we built multi-classifier systems, which aggregate the decisions of three statistical classifiers, a neural network and a decision tree. Finally, the receiver operating characteristic convex hull method was used to build the final classifier, which results to be the aggregation of the best local behaviors of both classifiers and combiners. All the CAD modules were tested on CT scans analyzed by two expert radiologists. In the experiments, we achieved a sensitivity of 92.5% against a specificity of 83.5%.     

Identifying mortality risk factors amongst acute coronary syndrome patients admitted to Arabian Gulf hospitals using machine‐learning methods

Acute coronary syndrome (ACS) is a leading cause of mortality and morbidity in the Arabian Gulf. In this study, the in‐hospital mortality amongst patients admitted with ACS to Arabian Gulf hospitals is predicted using a comprehensive modelling framework that combines powerful machine‐learning methods such as support‐vector machine (SVM), Naïve Bayes (NB), artificial neural networks (NN), and decision trees (DT). The performance of the machine‐learning methods is compared with that of the performance of a commonly used statistical method, namely, logistic regression (LR). The study follows the current practise of computing mortality risk using risk scores such as the Global Registry of Acute Coronary Events (GRACE) score, which has not been validated for Arabian Gulf patients. Cardiac registry data of 7,000 patients from 65 hospitals located in Arabian Gulf countries are used for the study. This study is unique as it uses a contemporary data analytics framework. A k‐fold (k = 10) cross‐validation is utilized to generate training and validation samples from the GRACE dataset. The machine‐learning‐based predictive models often incur prejudgments for imbalanced training data patterns. To mitigate the data imbalance due to scarce observations for in‐hospital mortalities, we have utilized specialized methods such as random undersampling (RUS) and synthetic minority over sampling technique (SMOTE). A detailed simulation experimentation is carried out to build models with each of the five predictive methods (LR, NN, NB, SVM, and DT) for the each of the three datasets k‐fold subsamples generated. The predictive models are developed under three schemes of the k‐fold samples that include no data imbalance, RUS, and SMOTE. We have implemented an information fusion method rooted in computing weighted impact scores obtain for an individual medical history attributes from each of the predictive models simulated for a collective recommendation based on an impact score specific to a predictor. Finally, we grouped the predictors using fuzzy c‐mean clustering method into three categories, high‐, medium‐, and low‐risk factors for in‐hospital mortality due to ACS. Our study revealed that patients with medical history related to the presences of peripheral artery disease, congestive heart failure, cardiovascular transient ischemic attack valvular disease, and coronary artery bypass grafting amongst others have the most risk for in‐hospital mortality.     

Automated breast cancer detection by thermography : performance goal and diagnostic feature identification

This paper presents the methodology used for establishing a performance goal and identifying the diagnostic features in a program to develop an automated system for breast cancer detection based on thermographic principles. The receiver operating characteristic (ROC) curve approach is used to evaluate both observer classification and classification rules based on an observer's evaluation of diagnostic features. The multivariate logistic function is applied to two sets of observer evaluated feature sets using 623 normal and 122 breast cancer diagnosed subjects. It is shown that the observer outperforms the multivariate logistic function classifier based on the diagnostic features.     

A hybrid particle swarm optimization based fuzzy expert system for the diagnosis of coronary artery disease

This paper presents a particle swarm optimization (PSO)-based fuzzy expert system for the diagnosis of coronary artery disease (CAD). The designed system is based on the Cleveland and Hungarian Heart Disease datasets. Since the datasets consist of many input attributes, decision tree (DT) was used to unravel the attributes that contribute towards the diagnosis. The output of the DT was converted into crisp if–then rules and then transformed into fuzzy rule base. PSO was employed to tune the fuzzy membership functions (MFs). Having applied the optimized MFs, the generated fuzzy expert system has yielded 93.27% classification accuracy. The major advantage of this approach is the ability to interpret the decisions made from the created fuzzy expert system, when compared with other approaches.     

Application of a neuro-fuzzy model to landslide-susceptibility mapping for shallow landslides in a tropical hilly area

This paper presents landslide-susceptibility mapping using an adaptive neuro-fuzzy inference system (ANFIS) using a geographic information system (GIS) environment. In the first stage, landslide locations from the study area were identified by interpreting aerial photographs and supported by an extensive field survey. In the second stage, landslide-related conditioning factors such as altitude, slope angle, plan curvature, distance to drainage, distance to road, soil texture and stream power index (SPI) were extracted from the topographic and soil maps. Then, landslide-susceptible areas were analyzed by the ANFIS approach and mapped using landslide-conditioning factors. In particular, various membership functions (MFs) were applied for the landslide-susceptibility mapping and their results were compared with the field-verified landslide locations. Additionally, the receiver operating characteristics (ROC) curve for all landslide susceptibility maps were drawn and the areas under curve values were calculated. The ROC curve technique is based on the plotting of model sensitivity — true positive fraction values calculated for different threshold values, versus model specificity — true negative fraction values, on a graph. Landslide test locations that were not used during the ANFIS modeling purpose were used to validate the landslide susceptibility maps. The validation results revealed that the susceptibility maps constructed by the ANFIS predictive models using triangular, trapezoidal, generalized bell and polynomial MFs produced reasonable results (84.39%), which can be used for preliminary land-use planning. Finally, the authors concluded that ANFIS is a very useful and an effective tool in regional landslide susceptibility assessment.     

An intelligent diagnosis system for diabetes on Linear Discriminant Analysis and Adaptive Network Based Fuzzy Inference System: LDA-ANFIS

In this study, an intelligent diagnosis system for diabetes on Linear Discriminant Analysis (LDA) and Adaptive Network Based Fuzzy Inference System (ANFIS): LDA-ANFIS is presented. The structure of this LDA-ANFIS intelligent system for diagnosis of diabetes is composed by two phases: The Linear Discriminant Analysis (LDA) phase and classificiation by using ANFIS classifier phase. In first phase, Linear Discriminant Analysis (LDA) is used to separate features variables between healthy and patient (diabetes) data. In second phase, the healthy and patient (diabetes) features obtained in first phase are given to inputs of ANFIS classifier. The correct diagnosis performance of the LDA-ANFIS intelligent system is calculated by using sensitivity and specificity analysis, classification accuracy and confusion matrix respectively. The classification accuracy of this LDA-ANFIS intelligent system was obtained about 84.61%.     

New machine-learning algorithms for prediction of Parkinson's disease

This article presents an enhanced prediction accuracy of diagnosis of Parkinson's disease (PD) to prevent the delay and misdiagnosis of patients using the proposed robust inference system. New machine-learning methods are proposed and performance comparisons are based on specificity, sensitivity, accuracy and other measurable parameters. The robust methods of treating Parkinson's disease (PD) includes sparse multinomial logistic regression, rotation forest ensemble with support vector machines and principal components analysis, artificial neural networks, boosting methods. A new ensemble method comprising of the Bayesian network optimised by Tabu search algorithm as classifier and Haar wavelets as projection filter is used for relevant feature selection and ranking. The highest accuracy obtained by linear logistic regression and sparse multinomial logistic regression is 100% and sensitivity, specificity of 0.983 and 0.996, respectively. All the experiments are conducted over 95% and 99% confidence levels and establish the results with corrected t-tests. This work shows a high degree of advancement in software reliability and quality of the computer-aided diagnosis system and experimentally shows best results with supportive statistical inference.