A biomedical system based on hidden Markov model for diagnosis of the heart valve diseases

Uğuz, Harun; Arslan, Ahmet; Türkoğlu, İ̇brahim

doi:10.1016/j.patrec.2006.08.009

Cited by 55 publications

(23 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Main aim of this system was to aid the physicians to overcome above two deficiencies. Classification performance of our previous system was comparable to studies [6] and [7] which use ANN and HMM as a classifier and more reliable than them because of SVM characteristics. When the data set and values of LS-SVM training parameters are selected optimally, LS-SVM shows a successful and reliable classification performance.…”

Section: Introductionsupporting

confidence: 55%

“…These optimal parameter values for LS-SVM are found automatically by our Renyi's entropy based procedure. According to our applications on DHS data set, classification correctness of LS-SVM with parameter regularization procedure is comparable with ANN based study [6] and HMM based study [7], however many advantages of our system over these systems are available. As we mentioned in our previous study [8]; training time of LS-SVM based system is about 13 times shorter than ANN one and developed system can always converge to same solution while alternatives can not.…”

Section: Experimental Classification Resultsmentioning

confidence: 87%

“…Doppler Heart Sounds (DHS) are produced by blood flow and vibration of cardiovascular components [5]. The paper uses study [6] as a preprocessing and feature extraction steps as in [7] and our previous study [8]. Alternative studies have been applied to Doppler signal research area including [9] and [10]; however these studies are not satisfactory completely.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Biomedical Decision Support System Using LS-SVM Classifier with an Efficient and New Parameter Regularization Procedure for Diagnosis of Heart Valve Diseases

Çomak

Arslan

2010

J Med Syst

Self Cite

View full text Add to dashboard Cite

Classification success of Support Vector Machine (SVM) depends on the characteristic of given data set and some training parameters (C and σ). In literature, a few studies have been presented for regularization of these parameters which affects classification performance directly. This study proposes a new approach based on Renyi's entropy and Logistic regression methods for parameter regularization. Our regularization procedure runs at two steps. In the first step, optimal value of kernel parameter interval is found via Renyi's entropy method and optimal C value is found via logistic regression using exponential function in the next step. In addition to, this new decision support system is applied to biomedical research area via an application related to Doppler Heart Sounds (DHS). Experimental results show the efficiency of developed regularization procedure.

show abstract

Section: Introductionsupporting

confidence: 55%

Section: Experimental Classification Resultsmentioning

confidence: 87%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Biomedical Decision Support System Using LS-SVM Classifier with an Efficient and New Parameter Regularization Procedure for Diagnosis of Heart Valve Diseases

Çomak

Arslan

2010

J Med Syst

Self Cite

View full text Add to dashboard Cite

show abstract

“…A brief overview on related works Various classification algorithms have been employed on valvular heart disease data set and high classification accuracies have been reported in the last decade [7][8][9][10][11][12]. Valvular heart disease data set was obtained from Firat Medical Center.…”

Section: Introductionmentioning

confidence: 99%

Support Vector Machine Ensembles for Intelligent Diagnosis of Valvular Heart Disease

Şengür

2011

J Med Syst

View full text Add to dashboard Cite

In this work, we investigate the use of ensemble learning for improving Support vector machines (SVM) classifier which is one of the important direction in the current research of machine learning, and thereinto bagging, boosting and random subspace are three powerful and popular representatives. Researchers have so far shown that the ensemble methods are quite well in many practical classification problems. However, for valvular heart disease detection, there are almost no studies investigating their feasibilities. Thus, in this study we evaluate the performance of three popular ensemble methods for diagnosing of the valvular heart disorders. To evaluate the performance of investigated ensemble methodology, a comparative study is realized by using a data set containing 215 samples. To achieve a comprehensive comparison, we consider the previous results reported by earlier methods. Experimental results suggest the feasibilities of ensemble of SVM classification methods, and we also derive some valuable conclusions on the performance of ensemble methods for valvular heart disease detection.

show abstract

“…Based on the research of Kline and McClintock (1953) they estimated the errors resulting from the uncertainty in the decision making process parameters. While using a Hidden Markov Model (HMM) and a technique for pattern recognition in order to detect cardiac diseases, Uguz et al (2007) noticed that although their technique was highly performing, it exhibited a lack of robustness due to variations inherent to the sound signals used to detect any anomaly in the heart condition.…”

Section: Introductionmentioning

confidence: 99%

Accuracy and robustness of decision making techniques in condition based maintenance

2009

View full text Add to dashboard Cite

This paper presents a study on Type I and Type II errors that are inherent in every technique for decision making in condition based maintenance. Specifically, the study considers those errors that result when three methods are used for decision making: the control charts known as the Statistical Process Control, the Hidden Markov Model, and the Proportional Hazard Model. The objective is to study the accuracy and the robustness of these techniques to variations in the models' parameters. Monte Carlo simulation is used to simulate the performance of these techniques. The effects of parameter's variations are obtained through Taguchi's design of experiments, and the results are analyzed by the analysis of variance technique. Those results show that accuracy and robustness should be taken into consideration when maintenance decisions are taken. Optimization techniques can be used to optimize the values of the parameters in order to minimize the two types of errors and to increase the correct decisions.

show abstract

A biomedical system based on hidden Markov model for diagnosis of the heart valve diseases

Cited by 55 publications

References 23 publications

A Biomedical Decision Support System Using LS-SVM Classifier with an Efficient and New Parameter Regularization Procedure for Diagnosis of Heart Valve Diseases

A Biomedical Decision Support System Using LS-SVM Classifier with an Efficient and New Parameter Regularization Procedure for Diagnosis of Heart Valve Diseases

Support Vector Machine Ensembles for Intelligent Diagnosis of Valvular Heart Disease

Accuracy and robustness of decision making techniques in condition based maintenance

Contact Info

Product

Resources

About