Medical Decision Support Systems

Ensemble classifiers provide an efficient method to deal with diverse set of applications in various domains. The proposed research signifies the effectiveness of ensemble classifier for computer-aided breast cancer diagnosis. A novel combination of five heterogeneous classifiers namely Naïve Bayes, Decision tree using Gini index, Decision tree using information gain, Support vector machine and Memory based learner are used to make an ensemble framework. Weighted voting technique is used to determine the final prediction where weights are assigned on the basis of classification accuracy. Four different breast cancer datasets are used from online data repositories. Feature selection and various preprocessing techniques are applied on the datasets to enhance the classification accuracy. The analyses of experimental results show that the proposed ensemble technique provided a significant improvement as compared to other classifiers. The best accuracy achieved by proposed ensemble is 97.42 % whereas the best precision and recall is 100 and 98.60 % respectively.

The medical diagnosis process can be interpreted as a decision making process during which the physician induces the diagnosis of a new and unknown case from an available set of clinical data and using his/her clinical experience. This process can be computerized in order to present medical diagnostic procedures in a rational, objective, accurate and efficient way. In the last few decades many researchers have focused on developing effective methods for intelligent heart disease prediction and decision support systems. For such a system, high accuracy of prediction is paramount. In this research, an ensemble classifier is proposed which uses majority vote based scheme for heart disease data classification and prediction. The five heterogeneous classifiers used to construct the ensemble model are; Naïve Bayes, Decision Tree based on Gini Index, Decision Tree based on Information Gain, Memory based Learner and Support Vector Machine. Five datasets from different data repositories are employed for testing the effectiveness of the ensemble model. Each dataset has different types of attributes, for instance binary, real, continuous, categorical etc. Experimental results with stratified cross validation show that the proposed MV5 framework deals with all the attribute types. MV5 has achieved an accuracy of 88.52% with 86.96% sensitivity, 90.83% specificity, and 88.85% f-measure. MV5 has achieved best performance results of accuracy 88.52%, sensitivity 86.96%, specificity 90.83%, and f-measure 88.85%. Comparison of proposed MV5 model with individual classifiers shows increase in average accuracy, sensitivity, specificity and f-measure of about 14%, 11%, 17% and 18% respectively.

An automated pulmonary nodule detection system is necessary to help radiologist to identify and detect the nodules at early stage. In this paper, a novel pulmonary nodule detection system is proposed using Artificial Neural Networks (ANN) based on hybrid features consist of 2D and 3D Geometric and Intensity based statistical features. The lung volume is segmented using thresholding, 3D connected component labeling, contour correction and morphological operators. The candidate nodules are extracted and pruned based on the rules that are built using characteristics of nodules. The 2D and 3D Geometric features and Intensity Based Statistical features are extracted and used to train a Neural Network. The proposed Computer-Aided Diagnostic (CAD) system is tested and validated using standard dataset of Lung Image Consortium Database (LIDC). The results obtained from proposed CAD system are good as compared to existing CAD systems. The sensitivity of 96.95% is achieved with accuracy of 96.68%.

To identify and detect the nodule at early stage, efficient pulmonary nodule detection system is required. A novel pulmonary nodule detection system using support vector machine (SVM) based in hybrid features is proposed in this paper. The lung volume is segmented using thresholding, initial label masking, background removal, connected component labeling, morphological operators and contour correction. The candidate nodules are extracted from the segmented lung volume. The 2-Dimensional (2D) and 3-Dimensionan (3D) Geometric and Intensity based statistical features are extracted. These features are used to train the support vector machine. The efficiency of proposed CAD (Computer Aided Diagnostic) system is tested and evaluated using Lung Image Consortium Database (LIDC) that is standard dataset used in Lung CAD Systems. The results achieved from proposed CAD system are excellent as compare to previous CAD systems. The sensitivity of 95.31% is achieved in proposed CAD system.

Conventional clinical decision support systems are based on individual classifiers or simple combination of these classifiers which tend to show moderate performance. This research paper presents a novel classifier ensemble framework based on enhanced bagging approach with multi-objective weighted voting scheme for prediction and analysis of heart disease. The proposed model overcomes the limitations of conventional performance by utilizing an ensemble of five heterogeneous classifiers: Naïve Bayes, linear regression, quadratic discriminant analysis, instance based learner and support vector machines. Five different datasets are used for experimentation, evaluation and validation. The datasets are obtained from publicly available data repositories. Effectiveness of the proposed ensemble is investigated by comparison of results with several classifiers. Prediction results of the proposed ensemble model are assessed by ten fold cross validation and ANOVA statistics. The experimental evaluation shows that the proposed framework deals with all type of attributes and achieved high diagnosis accuracy of 84.16 %, 93.29 % sensitivity, 96.70 % specificity, and 82.15 % f-measure. The f-ratio higher than f-critical and p value less than 0.05 for 95 % confidence interval indicate that the results are extremely statistically significant for most of the datasets.

The availability of large amount of medical data leads to the need of intelligent disease prediction and analysis tools to extract hidden information. Large number of data mining and statistical analysis tools are used for disease prediction. Single data mining techniques show acceptable level of accuracy for heart disease diagnosis. This research paper focuses on prediction and analysis of heart disease using weighted vote based classifier ensemble technique. The proposed ensemble model overcomes the limitations of conventional data mining techniques by employing the ensemble of five heterogeneous classifiers: Naïve Bayes, Decision Tree based on Gini Index, Decision Tree based on Information Gain, Instance based Learner and Support Vector Machines. We have used five benchmark heart disease datasets taken from UCI repository. Each dataset contains different set of feature space that ultimately leads to the prediction of heart disease. The effectiveness of proposed ensemble classifier is investigated by comparing the performance with several different researchers’ techniques. 10 fold cross validation is used to handle the class imbalance problem. Moreover, confusion matrices and ANOVA statistics are used to show the prediction results of all classifiers. The experimental results verify that the proposed ensemble classifier can deal with all type of attributes and it has achieved the high diagnosis accuracy of 87.37%, sensitivity 93.75%, specificity 92.86%, and f-measure 82.17%. The f-ratio higher than the f-critical and p-value less than 0.01 for 95% confidence interval indicates that the results are statistically significant for all the datasets.

Accuracy plays a vital role in the medical field as it concerns with the life of an individual. Extensive research has been conducted on disease classification and prediction using machine learning techniques. However, there is no agreement on which classifier produces the best results. Ensemble of classifiers has been proved to be an effective way to improve the classification accuracy. In this research we present an ensemble framework with multi-layer classification using enhanced bagging and optimized weighting. The proposed model called “HM-BagMoov” overcomes the limitations of conventional performance bottlenecks by utilizing an ensemble of seven heterogeneous classifiers. The framework is evaluated on five different heart disease datasets, four breast cancer datasets, two diabetes datasets, two liver disease datasets and one hepatitis dataset obtained from public repositories. The analysis of the results show that ensemble framework achieved the highest accuracy, sensitivity and F-Measure when compared with individual classifiers for all the diseases. In addition to this, the ensemble framework also achieved the highest accuracy when compared with the state of the art techniques. An application named “IntelliHealth” is also developed based on proposed model that may be used by hospitals/doctors for diagnostic advice.

Decision support is a crucial function for decision makers in many industries. Typically, Decision Support Systems (DSS) help decision-makers to gather and interpret information and build a foundation for decision-making. Medical Decision Support Systems (MDSS) play an increasingly important role in medical practice. By assisting doctors with making clinical decisions, DSS are expected to improve the quality of medical care. Conventional clinical decision support systems are based on individual classifiers or a simple combination of these classifiers which tend to show moderate performance. In this research, a multi-layer classifier ensemble framework is proposed based on the optimal combination of heterogeneous classifiers. The proposed model named “HMV” overcomes the limitations of conventional performance bottlenecks by utilizing an ensemble of seven heterogeneous classifiers. The framework is evaluated on two different heart disease datasets, two breast cancer datasets, two diabetes datasets, two liver disease datasets, one Parkinson’s disease dataset and one hepatitis dataset obtained from public repositories. Effectiveness of the proposed ensemble is investigated by comparison of results with several well-known classifiers as well as ensemble techniques. The experimental evaluation shows that the proposed framework dealt with all types of attributes and achieved high diagnosis accuracy. A case study is also presented based on a real time medical dataset in order to show the high performance and effectiveness of the proposed model.

Disease diagnosis at early stages can enable the physicians to overcome the complications and treat them properly. The diagnosis method plays an important role in disease diagnosis and accuracy of its treatment. A diagnosis expert system can help a great deal in identifying those diseases and describing methods of treatment to be carried out; taking into account the user capability in order to deal and interact with expert system easily and clearly. A good way to improve diagnosis accuracy of expert systems is use of ensemble classifiers. The proposed research presents an expert system using multi-layer classification with enhanced bagging and optimized weighting. The proposed method is named as “M2-BagWeight” which overcomes the limitations of individual as well as other ensemble classifiers. Evaluation of the proposed model is performed on two different liver disease datasets, chronic kidney disease dataset, heart disease dataset, diabetic retinopathy debrecen dataset, breast cancer dataset and primary tumor dataset obtained from UCI public repository. It is clear from the analysis of results that proposed expert system has achieved high classification and prediction accuracy when compared with individual as well as ensemble classifiers. Moreover, an application named “WebMAC” is also developed for practical implementation of proposed model in hospital for diagnostic advice.