Kenneth Revett | Champlain College (original) (raw)

Papers by Kenneth Revett

Research paper thumbnail of A rough sets based breast cancer decision support system

CSREA Press eBooks, 2005

Abstract Breast cancer remains the dominant form of cancer in women across most of the western wo... more Abstract Breast cancer remains the dominant form of cancer in women across most of the western world. Early diagnosis is a critical determinant of the successful outcome in terms of diagnostic accuracy and clinical outcome for the patient. In this preliminary study, we ...

Research paper thumbnail of Fiducial feature reduction analysis for electrocardiogram (ECG) based biometric recognition

Journal of Intelligent Information Systems, Jul 20, 2012

Although the electrocardiogram (ECG) has been a reliable diagnostic tool for decades, its deploym... more Although the electrocardiogram (ECG) has been a reliable diagnostic tool for decades, its deployment in the context of biometrics is relatively recent. Its robustness to falsification, the evidence it carries about aliveness and its rich feature space has rendered the deployment of ECG based biometrics an interesting prospect. The rich feature space contains fiducial based information such as characteristic peaks which reflect the underlying physiological properties of the heart. The principal goal of this study is to quantitatively evaluate the information content of the fiducial based feature set in terms of their effect on subject and heart beat classification accuracy (ECG data acquired from the PhysioNet ECG repository). To this end, a comprehensive set of fiducial based features was extracted from a collection of ECG records. This feature set was subsequently reduced using a variety of feature extraction/selection methods such as principle component analysis (PCA), linear discriminant analysis (LDA), information-gain ratio (IGR), and rough sets (in conjunction with the PASH algorithm). The performance of the reduced feature set was examined and the results evaluated with respect to the full feature set in terms of the overall classification accuracy and false (acceptance/rejection) ratios (FAR/FRR). The results of this study indicate that the PASH algorithm, deployed within the context of rough sets, reduced the dimensionality of the feature space maximally, while maintaining maximal classification accuracy.

Research paper thumbnail of Data mining the pima dataset using rough set theory with a special emphasis on rule reduction

8th International Multitopic Conference, 2004. Proceedings of INMIC 2004., Aug 10, 2005

Data Mining is a technique employed to extract non-trivial information from large datasets. There... more Data Mining is a technique employed to extract non-trivial information from large datasets. There are litera!ly thousands of medical databases throughout the world, each providing valuable information. Clinicians and medical researchers should be able to examine (data mining) ...

Research paper thumbnail of QT correction for fiducial ECG features based biometric systems

ABSTRACT In this paper, an electrocardiogram (ECG) based biometric system is proposed. A QT corre... more ABSTRACT In this paper, an electrocardiogram (ECG) based biometric system is proposed. A QT correction step is introduced to obviate the impact of heart rate variability, instead of just normalizing the features by the corresponding RR duration. Consequently, both approaches were examined in this work. Two sets of fiducial features were investigated: a super set of 36 features and a reduced version of it. Radial basis functions neural network is used as a classifier. The evaluation of the system was performed on the basis of subject identification (SI) accuracy and heartbeat recognition (HR) accuracy. The experiments were conducted using a 50-subject database and the results revealed the superiority of the QT correction approach, especially over time.

Research paper thumbnail of A novel feature set for deployment in ECG based biometrics

Research paper thumbnail of An Analysis of a Lymphoma/Leukaemia Dataset Using Rough Sets and Neural Networks

Lecture Notes in Computer Science, 2007

In this paper, we describe a rough sets approach to classification and attribute extraction of a ... more In this paper, we describe a rough sets approach to classification and attribute extraction of a lymphoma cancer dataset. We verify the classification accuracy of the results obtained from rough sets with a two artificial neural network based classifiers (ANNs). Our primary goal was to produce a classifier and a set of rules that could be used in a predictive manner. The dataset consisted of a number of relevant clinical variables obtained from patients that were suspected of having some form of blood based cancer (lymphoma or leukaemia). Of the 18 attributes that were collected for this patient cohort, seven were useful with respect to outcome prediction. In addition, this study was able to predict with a high degree of accuracy whether or not the disease would undergo metastases.

Research paper thumbnail of EEG Signal Classification Using Wavelet Feature Extraction and Neural Networks

Decision Support Systems have been utilised since 1960, providing physicians with fast and accura... more Decision Support Systems have been utilised since 1960, providing physicians with fast and accurate means towards more accurate diagnoses and increased tolerance when handling missing or incomplete data. This paper describes the application of neural network models for classification of electroencephalogram (EEG) signals. Decision making was performed in two stages: initially, a feature extraction scheme using the wavelet transform (WT) has been applied and then a learning-based algorithm classifier performed the classification. The performance of the neural model was evaluated in terms of training performance and classification accuracies and the results confirmed that the proposed scheme has potential in classifying the EEG signals.

Research paper thumbnail of Data Mining an EEG Dataset With an Emphasis on Dimensionality Reduction

The human brain is obviously a complex system, and exhibits rich spatiotemporal dynamics. Among t... more The human brain is obviously a complex system, and exhibits rich spatiotemporal dynamics. Among the noninvasive techniques for probing human brain dynamics, electroencephalography (EEG) provides a direct measure of cortical activity with millisecond temporal resolution. Early attempts to analyse EEG data relied on visual inspection of EEG records. Since the introduction of EEG recordings, the volume of data generated from a study involving a single patient has increased exponentially. Therefore, automation based on pattern classification techniques have been applied with considerable success. In this study, a multi-step approach for the classification of EEG signal has been adopted. We have analysed sets of EEG time series recording from healthy volunteers with open eyes and intracranial EEG recordings from patients with epilepsy during ictal (seizure) periods. In the present work, we have employed a discrete wavelet transform to the EEG data in order to extract temporal information in the form of changes in the frequency domain over time-that is they are able to extract non-stationary signals embedded in the noisy background of the human brain. Principal Components Analysis (PCA) and Rough Sets have been used to reduce the data dimensionality. A multi-classifier scheme consists of LVQ2.1 neural networks have been developed for the classification task. The experimental results validated the proposed methodology. Index Terms⎯ Discrete wavelet transform (DWT), electroencephalogram (EEG), neural networks, principal component analysis, and rough sets

Research paper thumbnail of A rule based approach to classification of EEG datasets: A comparison between ANFIS and rough sets

Research paper thumbnail of On the Deployment of Artificial Immune Systems for Biometrics

Artificial immune systems (AIS) are a computational metaphor based on biological implementations ... more Artificial immune systems (AIS) are a computational metaphor based on biological implementations of immune systems. Natural immune systems are capable of performing computation based on several properties that they possess. Immune systems are capable of adapting to new stimuli-they respond appropriately to novel stimuli, and they can remember previous encounters with stimuli. The processes which natural immune systems utilise are a combination of cellular and humoral responses-which act independently and in concert to perform stimulus identification and eradication, with minimal impact on the host. This provides an overview of artificial immune systems-which attempt to implement the basic functionality of natural systems. The basic properties and their interrelations are described in this paper-which is a prelude to their application in the context of biometrics. It will be demonstrated that the AIS approach is both a natural and potentially very effective approach to providing biometric security within a range of modalities.

Research paper thumbnail of A machine learning approach to artefact extraction from independent components derived from EEG datasets

Artefact selection from EEG datasets is still a task remanded to domain experts. In this study, a... more Artefact selection from EEG datasets is still a task remanded to domain experts. In this study, a working memory task dataset is used to examine automated methods for artefact removal. Such artefacts include eye blinks, muscle movements, and blood flow changes that do not reflect actual physiological responses to presented stimuli. In this work, a set of attributes were extracted from a 33 channel EEG recording. The attributes related predominantly to the independent components that were generated from the epoched data. In conjunction with expert analysis of the components, the attributes were used to produce an automated component artefact removal system. The result is an artefact removal system that performs at essentially 94% accuracy.

Research paper thumbnail of Machine learning techniques in electrocardiogram diagnosis

Advances in the area of computer sciences algorithms and artificial intelligence-based machine le... more Advances in the area of computer sciences algorithms and artificial intelligence-based machine learning techniques have greatly enhanced the electrocardiogram (ECG) signal classification and contributed to correct diagnosis. Recently, machine learning techniques have proved to be useful in electrocardiogram (ECG) diagnosis. This paper presents and discusses the current machine learning techniques and approaches used in ECG signal classification.

Research paper thumbnail of A rule-based approach to processing SPECT imaging for the diagnosis of heart disease

Research paper thumbnail of The Future of Behavioral Biometrics

John Wiley & Sons, Ltd eBooks, Nov 24, 2008

Research paper thumbnail of A Rough Sets Based Classifier for Primary Biliary Cirrhosis

In this paper, a decision support system is presented based on the machine learning approach of r... more In this paper, a decision support system is presented based on the machine learning approach of rough sets. The resulting decision support system was able to reduce the dimensionality of the data, produce a highly accurate classifier, and generate a rule based classifier that is readily understood by a domain expert. These preliminary results indicate that the rough sets machine learning approach can be successfully applied to biomedical datasets that contain a variety of attribute types, missing values and multiple decision classes.

Research paper thumbnail of Cognitive Biometrics

IGI Global eBooks, Oct 4, 2011

Cognitive biometrics is a new authentication scheme that utilises the cognitive, emotional, and c... more Cognitive biometrics is a new authentication scheme that utilises the cognitive, emotional, and conative state of an individual as the basis of user authentication and/or identification. These states of mind (and their derivatives) are extracted by recording various biosignals such as the EEG, ECG, and electrodermal response (EDR) of the individual in response to the presentation of the authentication stimulus. Stimuli are selected which elicit characteristic changes within the acquired biosignal(s) that represent unique responses from the individual. These characteristic changes are processed using a variety of machine learning algorithms, resulting in a unique signature that identifies or authenticates the individual. This approach can be applied in both static mode (single point of authentication), or in continuous mode, either alone, or in a multi-modal approach. The data suggest that the classification accuracy can reach

Research paper thumbnail of User Authentication via Keystroke Dynamics: An Artificial Immune System Based Approach

Keystroke dynamics is a behavioral biometric that is based on how a user enters their login detai... more Keystroke dynamics is a behavioral biometric that is based on how a user enters their login details. In this study, a set of eight attributes were extracted during the course of entering login details. This collection of attributes was used to form a reference signature (a biometric identification record) for subsequent authentication requests. The algorithm for the authentication process entails the deployment of an artificial immune based approach. The approach uses self-reactivity to discriminate self from non-self from the enrollment data. During the classification task, the system relies on deploying a pool of non-self reactive antibodies to perform a very general classification task. The results of this study indicate that the error rate is less than 5% in many cases.

Research paper thumbnail of On the Use of Rough Sets for Artefact Extraction from EEG Datasets

High-density electroencephalography produces large volumes of data. The analysis of EEG data is c... more High-density electroencephalography produces large volumes of data. The analysis of EEG data is confounded by the existed of a number of different artefacts such as eye blink and, muscle movement which impede the analysis of the data. Typically, artefacts are removed by visual inspection - an arduous task for high-density recordings. In addition, different researchers use Consistency across different laboratories is often difficult, and in addition, the task has to be repeated for each study. An automated method for artefact identification and removal would be a very useful tool for data processing in this domain. In this study, rough sets is employed as a means of automating artefact identification and removal within the context of EEG analysis using the EEGLAB analysis system. The results from this preliminary study indicate that artefacts can be identified and removed with approximately 85% accuracy.

Research paper thumbnail of Classification using adaptive fuzzy inference neural networks

ACTA Press eBooks, 2008

The architecture and learning scheme of a novel fuzzy logic system implemented in the framework o... more The architecture and learning scheme of a novel fuzzy logic system implemented in the framework of a neural network is proposed. The network constructs its initial rules by clustering while the final fuzzy rule base is determined by competitive learning. Both error backpropagation and recursive least squares estimation, are applied to the learning scheme. The convergence of training is faster because the least-squares algorithm is applied to the estimation of consequence parameters of the system and backpropagation is applied only to the estimation of the premise parameters. Using the proposed scheme, high-dimensional fuzzy systems can be realized with fewer rules than a typical Takagi-Sugeno fuzzy system. A number of simulations demonstrate the performance of the proposed system.

Research paper thumbnail of Data Mining a Prostate Cancer Dataset Using Neural Networks

Prostate cancer remains one of the leading causes of cancer death worldwide, with a reported inci... more Prostate cancer remains one of the leading causes of cancer death worldwide, with a reported incidence rate of 650,000 cases per annum worldwide. The causal factors of prostate cancer still remain to be determined. In this paper, we investigate a medical dataset containing clinical information on 502 prostate cancer patients using the machine learning technique of rough sets and radial basis function neural network.. Our preliminary results yield a classification accuracy of 90%, with high sensitivity and specificity (both at approximately 91%). Our results yield a predictive positive value (PPN) of 81% and a predictive negative value (PNV) of 95%.

Research paper thumbnail of A rough sets based breast cancer decision support system

CSREA Press eBooks, 2005

Abstract Breast cancer remains the dominant form of cancer in women across most of the western wo... more Abstract Breast cancer remains the dominant form of cancer in women across most of the western world. Early diagnosis is a critical determinant of the successful outcome in terms of diagnostic accuracy and clinical outcome for the patient. In this preliminary study, we ...

Research paper thumbnail of Fiducial feature reduction analysis for electrocardiogram (ECG) based biometric recognition

Journal of Intelligent Information Systems, Jul 20, 2012

Although the electrocardiogram (ECG) has been a reliable diagnostic tool for decades, its deploym... more Although the electrocardiogram (ECG) has been a reliable diagnostic tool for decades, its deployment in the context of biometrics is relatively recent. Its robustness to falsification, the evidence it carries about aliveness and its rich feature space has rendered the deployment of ECG based biometrics an interesting prospect. The rich feature space contains fiducial based information such as characteristic peaks which reflect the underlying physiological properties of the heart. The principal goal of this study is to quantitatively evaluate the information content of the fiducial based feature set in terms of their effect on subject and heart beat classification accuracy (ECG data acquired from the PhysioNet ECG repository). To this end, a comprehensive set of fiducial based features was extracted from a collection of ECG records. This feature set was subsequently reduced using a variety of feature extraction/selection methods such as principle component analysis (PCA), linear discriminant analysis (LDA), information-gain ratio (IGR), and rough sets (in conjunction with the PASH algorithm). The performance of the reduced feature set was examined and the results evaluated with respect to the full feature set in terms of the overall classification accuracy and false (acceptance/rejection) ratios (FAR/FRR). The results of this study indicate that the PASH algorithm, deployed within the context of rough sets, reduced the dimensionality of the feature space maximally, while maintaining maximal classification accuracy.

Research paper thumbnail of Data mining the pima dataset using rough set theory with a special emphasis on rule reduction

8th International Multitopic Conference, 2004. Proceedings of INMIC 2004., Aug 10, 2005

Data Mining is a technique employed to extract non-trivial information from large datasets. There... more Data Mining is a technique employed to extract non-trivial information from large datasets. There are litera!ly thousands of medical databases throughout the world, each providing valuable information. Clinicians and medical researchers should be able to examine (data mining) ...

Research paper thumbnail of QT correction for fiducial ECG features based biometric systems

ABSTRACT In this paper, an electrocardiogram (ECG) based biometric system is proposed. A QT corre... more ABSTRACT In this paper, an electrocardiogram (ECG) based biometric system is proposed. A QT correction step is introduced to obviate the impact of heart rate variability, instead of just normalizing the features by the corresponding RR duration. Consequently, both approaches were examined in this work. Two sets of fiducial features were investigated: a super set of 36 features and a reduced version of it. Radial basis functions neural network is used as a classifier. The evaluation of the system was performed on the basis of subject identification (SI) accuracy and heartbeat recognition (HR) accuracy. The experiments were conducted using a 50-subject database and the results revealed the superiority of the QT correction approach, especially over time.

Research paper thumbnail of A novel feature set for deployment in ECG based biometrics

Research paper thumbnail of An Analysis of a Lymphoma/Leukaemia Dataset Using Rough Sets and Neural Networks

Lecture Notes in Computer Science, 2007

In this paper, we describe a rough sets approach to classification and attribute extraction of a ... more In this paper, we describe a rough sets approach to classification and attribute extraction of a lymphoma cancer dataset. We verify the classification accuracy of the results obtained from rough sets with a two artificial neural network based classifiers (ANNs). Our primary goal was to produce a classifier and a set of rules that could be used in a predictive manner. The dataset consisted of a number of relevant clinical variables obtained from patients that were suspected of having some form of blood based cancer (lymphoma or leukaemia). Of the 18 attributes that were collected for this patient cohort, seven were useful with respect to outcome prediction. In addition, this study was able to predict with a high degree of accuracy whether or not the disease would undergo metastases.

Research paper thumbnail of EEG Signal Classification Using Wavelet Feature Extraction and Neural Networks

Decision Support Systems have been utilised since 1960, providing physicians with fast and accura... more Decision Support Systems have been utilised since 1960, providing physicians with fast and accurate means towards more accurate diagnoses and increased tolerance when handling missing or incomplete data. This paper describes the application of neural network models for classification of electroencephalogram (EEG) signals. Decision making was performed in two stages: initially, a feature extraction scheme using the wavelet transform (WT) has been applied and then a learning-based algorithm classifier performed the classification. The performance of the neural model was evaluated in terms of training performance and classification accuracies and the results confirmed that the proposed scheme has potential in classifying the EEG signals.

Research paper thumbnail of Data Mining an EEG Dataset With an Emphasis on Dimensionality Reduction

The human brain is obviously a complex system, and exhibits rich spatiotemporal dynamics. Among t... more The human brain is obviously a complex system, and exhibits rich spatiotemporal dynamics. Among the noninvasive techniques for probing human brain dynamics, electroencephalography (EEG) provides a direct measure of cortical activity with millisecond temporal resolution. Early attempts to analyse EEG data relied on visual inspection of EEG records. Since the introduction of EEG recordings, the volume of data generated from a study involving a single patient has increased exponentially. Therefore, automation based on pattern classification techniques have been applied with considerable success. In this study, a multi-step approach for the classification of EEG signal has been adopted. We have analysed sets of EEG time series recording from healthy volunteers with open eyes and intracranial EEG recordings from patients with epilepsy during ictal (seizure) periods. In the present work, we have employed a discrete wavelet transform to the EEG data in order to extract temporal information in the form of changes in the frequency domain over time-that is they are able to extract non-stationary signals embedded in the noisy background of the human brain. Principal Components Analysis (PCA) and Rough Sets have been used to reduce the data dimensionality. A multi-classifier scheme consists of LVQ2.1 neural networks have been developed for the classification task. The experimental results validated the proposed methodology. Index Terms⎯ Discrete wavelet transform (DWT), electroencephalogram (EEG), neural networks, principal component analysis, and rough sets

Research paper thumbnail of A rule based approach to classification of EEG datasets: A comparison between ANFIS and rough sets

Research paper thumbnail of On the Deployment of Artificial Immune Systems for Biometrics

Artificial immune systems (AIS) are a computational metaphor based on biological implementations ... more Artificial immune systems (AIS) are a computational metaphor based on biological implementations of immune systems. Natural immune systems are capable of performing computation based on several properties that they possess. Immune systems are capable of adapting to new stimuli-they respond appropriately to novel stimuli, and they can remember previous encounters with stimuli. The processes which natural immune systems utilise are a combination of cellular and humoral responses-which act independently and in concert to perform stimulus identification and eradication, with minimal impact on the host. This provides an overview of artificial immune systems-which attempt to implement the basic functionality of natural systems. The basic properties and their interrelations are described in this paper-which is a prelude to their application in the context of biometrics. It will be demonstrated that the AIS approach is both a natural and potentially very effective approach to providing biometric security within a range of modalities.

Research paper thumbnail of A machine learning approach to artefact extraction from independent components derived from EEG datasets

Artefact selection from EEG datasets is still a task remanded to domain experts. In this study, a... more Artefact selection from EEG datasets is still a task remanded to domain experts. In this study, a working memory task dataset is used to examine automated methods for artefact removal. Such artefacts include eye blinks, muscle movements, and blood flow changes that do not reflect actual physiological responses to presented stimuli. In this work, a set of attributes were extracted from a 33 channel EEG recording. The attributes related predominantly to the independent components that were generated from the epoched data. In conjunction with expert analysis of the components, the attributes were used to produce an automated component artefact removal system. The result is an artefact removal system that performs at essentially 94% accuracy.

Research paper thumbnail of Machine learning techniques in electrocardiogram diagnosis

Advances in the area of computer sciences algorithms and artificial intelligence-based machine le... more Advances in the area of computer sciences algorithms and artificial intelligence-based machine learning techniques have greatly enhanced the electrocardiogram (ECG) signal classification and contributed to correct diagnosis. Recently, machine learning techniques have proved to be useful in electrocardiogram (ECG) diagnosis. This paper presents and discusses the current machine learning techniques and approaches used in ECG signal classification.

Research paper thumbnail of A rule-based approach to processing SPECT imaging for the diagnosis of heart disease

Research paper thumbnail of The Future of Behavioral Biometrics

John Wiley & Sons, Ltd eBooks, Nov 24, 2008

Research paper thumbnail of A Rough Sets Based Classifier for Primary Biliary Cirrhosis

In this paper, a decision support system is presented based on the machine learning approach of r... more In this paper, a decision support system is presented based on the machine learning approach of rough sets. The resulting decision support system was able to reduce the dimensionality of the data, produce a highly accurate classifier, and generate a rule based classifier that is readily understood by a domain expert. These preliminary results indicate that the rough sets machine learning approach can be successfully applied to biomedical datasets that contain a variety of attribute types, missing values and multiple decision classes.

Research paper thumbnail of Cognitive Biometrics

IGI Global eBooks, Oct 4, 2011

Cognitive biometrics is a new authentication scheme that utilises the cognitive, emotional, and c... more Cognitive biometrics is a new authentication scheme that utilises the cognitive, emotional, and conative state of an individual as the basis of user authentication and/or identification. These states of mind (and their derivatives) are extracted by recording various biosignals such as the EEG, ECG, and electrodermal response (EDR) of the individual in response to the presentation of the authentication stimulus. Stimuli are selected which elicit characteristic changes within the acquired biosignal(s) that represent unique responses from the individual. These characteristic changes are processed using a variety of machine learning algorithms, resulting in a unique signature that identifies or authenticates the individual. This approach can be applied in both static mode (single point of authentication), or in continuous mode, either alone, or in a multi-modal approach. The data suggest that the classification accuracy can reach

Research paper thumbnail of User Authentication via Keystroke Dynamics: An Artificial Immune System Based Approach

Keystroke dynamics is a behavioral biometric that is based on how a user enters their login detai... more Keystroke dynamics is a behavioral biometric that is based on how a user enters their login details. In this study, a set of eight attributes were extracted during the course of entering login details. This collection of attributes was used to form a reference signature (a biometric identification record) for subsequent authentication requests. The algorithm for the authentication process entails the deployment of an artificial immune based approach. The approach uses self-reactivity to discriminate self from non-self from the enrollment data. During the classification task, the system relies on deploying a pool of non-self reactive antibodies to perform a very general classification task. The results of this study indicate that the error rate is less than 5% in many cases.

Research paper thumbnail of On the Use of Rough Sets for Artefact Extraction from EEG Datasets

High-density electroencephalography produces large volumes of data. The analysis of EEG data is c... more High-density electroencephalography produces large volumes of data. The analysis of EEG data is confounded by the existed of a number of different artefacts such as eye blink and, muscle movement which impede the analysis of the data. Typically, artefacts are removed by visual inspection - an arduous task for high-density recordings. In addition, different researchers use Consistency across different laboratories is often difficult, and in addition, the task has to be repeated for each study. An automated method for artefact identification and removal would be a very useful tool for data processing in this domain. In this study, rough sets is employed as a means of automating artefact identification and removal within the context of EEG analysis using the EEGLAB analysis system. The results from this preliminary study indicate that artefacts can be identified and removed with approximately 85% accuracy.

Research paper thumbnail of Classification using adaptive fuzzy inference neural networks

ACTA Press eBooks, 2008

The architecture and learning scheme of a novel fuzzy logic system implemented in the framework o... more The architecture and learning scheme of a novel fuzzy logic system implemented in the framework of a neural network is proposed. The network constructs its initial rules by clustering while the final fuzzy rule base is determined by competitive learning. Both error backpropagation and recursive least squares estimation, are applied to the learning scheme. The convergence of training is faster because the least-squares algorithm is applied to the estimation of consequence parameters of the system and backpropagation is applied only to the estimation of the premise parameters. Using the proposed scheme, high-dimensional fuzzy systems can be realized with fewer rules than a typical Takagi-Sugeno fuzzy system. A number of simulations demonstrate the performance of the proposed system.

Research paper thumbnail of Data Mining a Prostate Cancer Dataset Using Neural Networks

Prostate cancer remains one of the leading causes of cancer death worldwide, with a reported inci... more Prostate cancer remains one of the leading causes of cancer death worldwide, with a reported incidence rate of 650,000 cases per annum worldwide. The causal factors of prostate cancer still remain to be determined. In this paper, we investigate a medical dataset containing clinical information on 502 prostate cancer patients using the machine learning technique of rough sets and radial basis function neural network.. Our preliminary results yield a classification accuracy of 90%, with high sensitivity and specificity (both at approximately 91%). Our results yield a predictive positive value (PPN) of 81% and a predictive negative value (PNV) of 95%.