This paper presents a study of the application of autonomously learning multiple neural network systems to medical pattern classification tasks. In our earlier work, a hybrid neural network architecture has been developed for on-line learning and probability estimation tasks. The network has been shown to be capable of asymptotically achieving the Bayes optimal classification rates, on-line, in a number of benchmark classification experiments. In the context of pattern classification, however, the concept of multiple classifier systems has been proposed to improve the performance of a single classifier. Thus, three decision combination algorithms have been implemented to produce a multiple neural network classifier system. Here the applicability of the system is assessed using patient records in two medical domains. The first task is the prognosis of patients admitted to coronary care units; whereas the second is the prediction of survival in trauma patients. The results are compared with those from logistic regression models, and implications of the system as a useful clinical diagnostic tool are discussed.
In this paper we suggest that the healthcare enterprise needs to be more conscious of its vast knowledge resources vis-à-vis the exploitation of knowledge management techniques to efficiently manage its knowledge. The development of healthcare enterprise memory is suggested as a solution, together with a novel approach advocating the operationalisation of healthcare enterprise memories leading to the modelling of healthcare processes for strategic planning. As an example, we present a simulation of Service Delivery Time in a hospital's OPD.
The 21st century promises to usher in an era of Internet based healthcare services--Tele-Healthcare. Such services augur well with the on-going paradigm shift in healthcare delivery patterns, i.e. patient centred services as opposed to provider centred services and wellness maintenance as opposed to illness management. This paper presents a Tele-Healthcare info-structure TIDE--an 'intelligent' wellness-oriented healthcare delivery environment. TIDE incorporates two WWW-based healthcare systems: (1) AIMS (Automated Health Monitoring System) for wellness maintenance and (2) IDEAS (Illness Diagnostic & Advisory System) for illness management. Our proposal comes from an attempt to rethink the sources of possible leverage in improving healthcare; vis-à-vis the provision of a continuum of personalised home-based healthcare services that emphasise the role of the individual in self health maintenance.
Presently, there is a growing demand from the healthcare community to leverage upon and transform the vast quantities of healthcare data into value-added, 'decision-quality' knowledge, vis-à-vis, strategic knowledge services oriented towards healthcare management and planning. To meet this end, we present a Strategic Knowledge Services Info-structure that leverages on existing healthcare knowledge/data bases to derive decision-quality knowledge-knowledge that is extracted from healthcare data through services akin to knowledge discovery in databases and data mining.
The healthcare enterprise requires a great deal of knowledge to maintain premium efficiency in the delivery of quality healthcare. We employ Knowledge Management based knowledge acquisition strategies to procure 'tacit' healthcare knowledge from experienced healthcare practitioners. Situational, problem-specific Scenarios are proposed as viable knowledge acquisition and representation constructs. We present a healthcare Tacit Knowledge Acquisition Info-structure (TKAI) that allows remote healthcare practitioners to record their tacit knowledge. TKAI employs (a) ontologies for standardisation of tacit knowledge and (b) XML to represent scenario instances for their transfer over the Internet to the server-side Scenario-Base and for the global sharing of acquired tacit healthcare knowledge.
Electronic patient records (EPR) can be regarded as an implicit source of clinical behaviour and problem-solving knowledge, systematically compiled by clinicians. We present an approach, together with its computational implementation, to pro-actively transform XML-based EPR into specialised Clinical Cases (CC) in the realm of Medical Case Base Systems. The 'correct' transformation of EPR to CC involves structural, terminological and conceptual standardisation, which is achieved by a confluence of techniques and resources, such as XML, UMLS (meta-thesaurus) and medical knowledge ontologies. We present below the functional architecture of a Medical Case-Base Reasoning Info-Structure (MCRIS) that features two distinct, yet related, functionalities: (1) a generic medical case-based reasoning system for decision-support activities; and (2) an EPR-CC transformation system to transform typical EPR's to CC.
The abundance and transient nature to healthcare knowledge has rendered it difficult to acquire with traditional knowledge acquisition methods. In this paper, we propose a Knowledge Management approach, through the use of scenarios, as a mean to acquire and represent tacit healthcare knowledge. This proposition is based on the premise that tacit knowledge is best manifested in atypical situations. We also provide an overview of the representational scheme and novel acquisition mechanism of scenarios.
A quantitative structure-human intestinal absorption relationship was developed using artificial neural network (ANN) modeling. A set of 86 drug compounds and their experimentally-derived intestinal absorption values used in this study was gathered from the literature and a total of 57 global molecular descriptors, including constitutional, topological, chemical, geometrical and quantum chemical descriptors, calculated for each compound. A supervised network with radial basis transfer function was used to correlate calculated molecular descriptors with experimentally-derived measures of human intestinal absorption. A genetic algorithm was then used to select important molecular descriptors. Intestinal absorption values (IA%) were used as the ANN's output and calculated molecular descriptors as the inputs. The best genetic neural network (GNN) model with 15 input descriptors was chosen, and the significance of the selected descriptors for intestinal absorption examined. Results obtained with the model that was developed indicate that lipophilicity, conformational stability and inter-molecular interactions (polarity, and hydrogen bonding) have the largest impact on intestinal absorption.
Worldwide healthcare delivery trends are undergoing a subtle paradigm shift--patient centered services as opposed to provider centered services and wellness maintenance as opposed to illness management. In this paper we present a Tele-Healthcare project TIDE--Tele-Healthcare Information and Diagnostic Environment. TIDE manifests an 'intelligent' healthcare environment that aims to ensure lifelong coverage of person-specific health maintenance decision-support services--i.e., both wellness maintenance and illness management services--ubiquitously available via the Internet/WWW. Taking on an all-encompassing health maintenance role--spanning from wellness to illness issues--the functionality of TIDE involves the generation and delivery of (a) Personalized, Pro-active, Persistent, Perpetual, and Present wellness maintenance services, and (b) remote diagnostic services for managing noncritical illnesses. Technically, TIDE is an amalgamation of diverse computer technologies--Artificial Intelligence, Internet, Multimedia, Databases, and Medical Informatics--to implement a sophisticated healthcare delivery infostructure.
Quantitative structure-retention relationship(QSRR) method was used to model reversed-phase high-performance liquid chromatography (RP-HPLC) separation of 18 selected amino acids. Retention data for phenylthiocarbamyl (PTC) amino acids derivatives were obtained using gradient elution on ODS column with mobile phase of varying acetonitrile, acetate buffer and containing 0.5 ml/l of triethylamine (TEA). Molecular structure of each amino acid was encoded with 36 calculated molecular descriptors. The correlation between the molecular descriptors and the retention time of the compounds in the calibration set was established using the genetic neural network method. A genetic algorithm (GA) was used to select important molecular descriptors and supervised artificial neural network (ANN) was used to correlate mobile phase composition and selected descriptors with the experimentally derived retention times. Retention time values were used as the network's output and calculated molecular descriptors and mobile phase composition as the inputs. The best model with five input descriptors was chosen, and the significance of the selected descriptors for amino acid separation was examined. Results confirmed the dominant role of the organic modifier in such chromatographic systems in addition to lipophilicity (log P) and molecular size and shape (topological indices) of investigated solutes.
This paper investigates the use of a neural-network-based intelligent learning system for the prediction of drug release profiles. An experimental study in transdermal iontophoresis (TI) is employed to evaluate the applicability of a particular neural network (NN) model, i.e. the Gaussian mixture model (GMM), in modeling and predicting drug release profiles. A number of tests are systematically designed using the face-centered central composite design (CCD) approach to examine the effects of various process variables simultaneously during the iontophoresis process. The GMM is then applied to model and predict the drug release profiles based on the data samples collected from the experiments. The GMM results are compared with those from multiple regression models. In addition, the bootstrap method is used to assess the reliability of the network predictions by estimating confidence intervals associated with the results. The results demonstrate that the combination of the face-centered CCD and GMM can be employed as a useful intelligent tool for the prediction of time-series profiles in pharmaceutical and biomedical experiments.
Training methodology of the Back Propagation Network (BPN) is well documented. One aspect of BPN that requires investigation is whether or not the BPN would get trained for a given training data set and architecture. In this paper the behavior of the BPN is analyzed during its training phase considering convergent and divergent training data sets. Evolution of the weights during the training phase was monitored for the purpose of analysis. The evolution of weights was plotted as return map and was characterized by means of fractal dimension. This fractal dimensional analysis of the weight evolution trajectories is used to provide a new insight to understand the behavior of BPN and dynamics in the evolution of weights.
This paper presents two novel approaches to determine optimum growing multi-experts network (GMN) structure. The first method called direct method deals with expertise domain and levels in connection with local experts. The growing neural gas (GNG) algorithm is used to cluster the local experts. The concept of error distribution is used to apportion error among the local experts. After reaching the specified size of the network, redundant experts removal algorithm is invoked to prune the size of the network based on the ranking of the experts. However, GMN is not ergonomic due to too many network control parameters. Therefore, a self-regulating GMN (SGMN) algorithm is proposed. SGMN adopts self-adaptive learning rates for gradient-descent learning rules. In addition, SGMN adopts a more rigorous clustering method called fully self-organized simplified adaptive resonance theory in a modified form. Experimental results show SGMN obtains comparative or even better performance than GMN in four benchmark examples, with reduced sensitivity to learning parameters setting. Moreover, both GMN and SGMN outperform the other neural networks and statistical models. The efficacy of SGMN is further justified in three industrial applications and a control problem. It provides consistent results besides holding out a profound potential and promise for building a novel type of nonlinear model consisting of several local linear models.
Tacit knowledge of health-care experts is an important source of experiential know-how, yet due to various operational and technical reasons, such health-care knowledge is not entirely harnessed and put into professional practice. Emerging knowledge-management (KM) solutions suggest strategies to acquire the seemingly intractable and nonarticulated tacit knowledge of health-care experts. This paper presents a KM methodology, together with its computational implementation, to 1) acquire the tacit knowledge possessed by health-care experts; 2) represent the acquired tacit health-care knowledge in a computational formalism--i.e., clinical scenarios--that allows the reuse of stored knowledge to acquire tacit knowledge; and 3) crystallize the acquired tacit knowledge so that it is validated for health-care decision-support and medical education systems.
Legendre moments are continuous moments, hence, when applied to discrete-space images, numerical approximation is involved and error occurs. This paper proposes a method to compute the exact values of the moments by mathematically integrating the Legendre polynomials over the corresponding intervals of the image pixels. Experimental results show that the values obtained match those calculated theoretically, and the image reconstructed from these moments have lower error than that of the conventional methods for the same order. Although the same set of exact Legendre moments can be obtained indirectly from the set of geometric moments, the computation time taken is much longer than the proposed method.
Different approaches to gray and white matter measurements in magnetic resonance imaging (MRI) have been studied. For clinical use, the estimated values must be reliable and accurate when, unfortunately, many techniques fail on these criteria in an unrestricted clinical environment. A recent method for tissue clusterization in MRI analysis has the advantage of great simplicity, and it takes the account of partial volume effects. In this study, we will evaluate the intensity of MR sequences known as T1-weighted images in an axial sliced section. Intensity group clustering algorithms are proposed to achieve further diagnosis for brain MRI, which has been hardly studied. Subjective study has been suggested to evaluate the clustering group intensity in order to obtain the best diagnosis as well as better detection for the suspected cases. This technique makes use of image tissue biases of intensity value pixels to provide 2 regions of interest as techniques. Moreover, the original mathematic solution could still be used with a specific set of modern sequences. There are many advantages to generalize the solution, which give far more scope for application and greater accuracy.
We describe a semi-automated technique for the quantitative assessment of breast density from digitized mammograms in comparison with patterns suggested by Tabar. It was developed using the MATLAB-based graphical user interface applications. It is based on an interactive thresholding method, after a short automated method that shows the fibroglandular tissue area, breast area and breast density each time new thresholds are placed on the image. The breast density is taken as a percentage of the fibroglandular tissue to the breast tissue areas. It was tested in four different ways, namely by examining: (i) correlation of the quantitative assessment results with subjective classification, (ii) classification performance using the quantitative assessment technique, (iii) interobserver agreement and (iv) intraobserver agreement. The results of the quantitative assessment correlated well (r2 = 0.92) with the subjective Tabar patterns classified by the radiologist (correctly classified 83% of digitized mammograms). The average kappa coefficient for the agreement between the readers was 0.63. This indicated moderate agreement between the three observers in classifying breast density using the quantitative assessment technique. The kappa coefficient of 0.75 for intraobserver agreement reflected good agreement between two sets of readings. The technique may be useful as a supplement to the radiologist's assessment in classifying mammograms into Tabar's pattern associated with breast cancer risk.
Biometric analysis for identity verification is becoming a widespread reality. Such implementations necessitate large-scale capture and storage of biometric data, which raises serious issues in terms of data privacy and (if such data is compromised) identity theft. These problems stem from the essential permanence of biometric data, which (unlike secret passwords or physical tokens) cannot be refreshed or reissued if compromised. Our previously presented biometric-hash framework prescribes the integration of external (password or token-derived) randomness with user-specific biometrics, resulting in bitstring outputs with security characteristics (i.e., noninvertibility) comparable to cryptographic ciphers or hashes. The resultant BioHashes are hence cancellable, i.e., straightforwardly revoked and reissued (via refreshed password or reissued token) if compromised. BioHashing furthermore enhances recognition effectiveness, which is explained in this paper as arising from the Random Multispace Quantization (RMQ) of biometric and external random inputs.
Several incidents that occurred around the world involving power failure
caused by unscheduled line outages were identified as one of the main
contributors to power failure and cascading blackout in electric power
environment. With the advancement of computer technologies, artificial
intelligence (AI) has been widely accepted as one method that can be applied
to predict the occurrence of unscheduled disturbance. This paper presents
the development of automatic contingency analysis and ranking algorithm
for the application in the Artificial Neural Network (ANN). The ANN is
developed in order to predict the post-outage severity index from a set of preoutage
data set. Data were generated using the newly developed automatic
contingency analysis and ranking (ACAR) algorithm. Tests were conducted
on the 24-bus IEEE Reliability Test Systems. Results showed that the developed
technique is feasible to be implemented practically and an agreement was
achieved in the results obtained from the tests. The developed ACAR can be
utilised for further testing and implementation in other IEEE RTS test systems
particularly in the system, which required fast computation time. On the other
hand, the developed ANN can be used for predicting the post-outage severity
index and hence system stability can be evaluated.