A New Expert System for Pediatric Respiratory Diseases by ... - wseas.us

Recent Researches ιn Applied Informatics

A New Expert System for Pediatric Respiratory Diseases by Using Neural Networks AHMED A. RADWAN

HAZEM M. EL-BAKRY

HAGER M. EL HADAD

Faculty of Computer Science and Information Systems, Minia University, Minia, Egypt

Faculty of Computer Science and Information Systems, Mansoura University, Egypt Email:[email protected]

Faculty of Computer Science, Nahda University, BeniSuef, Egypt

Abstract- The successful application of data mining in highly visible fields like e-business and marketing have led to the popularity of its use in knowledge discovery in databases (KDD) in other industries and sectors. Among these sectors that are just discovering data mining are the fields of medicine and public health. The medical industries collect huge amounts of healthcare data which, unfortunately, are not “mined” to discover hidden information. We can describe this data as being ‘information rich’ yet ‘knowledge poor’. In this study, we briefly examine the use of the most important data mining techniques such as Artificial Neural Network to massive volume of data in medical field which is pediatric respiratory disease. Using medical symptoms such as dry cough, productive cough, fever, heamoptysis, tachypnea, dysnea and etc. Also using doctor sign such as bronchial breathing, chest pain, clubbing of finger, crepitation, ronchi, cyanosis, decrease brearthing sound on auscultation, dullness on percussion, hyper resonant on percussion, inability to swallow, mucopurelent sputum, pleural rub, respiratory distress, sputum (white), stridor, upper respiratory infection, wheezing, X-ray [showing lung consolidation], X-ray [shawing edematous epiglottic], X-ray [shawing subglottic narrowing and classic narrow trachea], X-ray [showing hypertanslucent lung], X-ray [showing lobar collapse and (increase) bronchovascular marking], Xray[ showing diffuse haziness], it can predict the likelihood of patients getting a respiratory disease. They enable significant knowledge, e.g. patterns, relationships between medical factors related to respiratory disease, to be established. Keywords - Data mining, artificial neural network, medical application respiratory diseases.

information found in databases [5] [6]. Its functions include clustering, classification, prediction, and associations. One of the most important data mining applications is that of mining association rules which, first introduced in 1993 [7], that used to identify relationships among a set of items in databases [8]. Respiratory disease databases are elements of our research. This database consists of 699 case and this cases have many common symptom such as dry cough, productive dry cough, productive cough, fever, heamoptysis, tachypnea, dysnea and etc. this symptom are common in eight disease as bronchiolitis, pneumonia, acute epiglottitis, pleurisy, emphysema, acute laryngotracheobronchitis, bronchial asthma and bronchiectasis.

1. Introduction Data mining and its application to medicine and public health is a relatively young field of study. In 2003, Wilson et al. began to scan cases where KDD and data mining techniques were applied in health databases. They found confusion regarding what constituted data mining. “Some authors refer to data mining as the process of acquiring information, whereas others refer to data mining as utilization of statistical techniques within the knowledge discovery process.” [1]. Despite the differences and clashes in approaches, the health sector has more need for data mining today. There are several arguments that could be advanced to support the use of data mining in the health sector, covering not just concerns of public health but also the private health sector (which, in fact, as can be shown later, are also stakeholders in public health) [2]. Because of misconceptions still going on in the medical community about what data mining comprises, let us first define what we mean by it. The generally accepted definition of data mining today is the set of procedures and techniques for discovering and describing patterns and trends in data [3].

The rest of the paper is organized as follows. Medical expert system used in this paper is described in Section 2. Architecture of a typical expert system is detailed introduced in Section 2.1. In Section 2.2 the design of medical expert systems is presented. In Section 3, review of medical expert system in respiratory system is examined. Design of expert system by using artificial neural network and conclusion are presented in Section 4 and Section 5, respectively.

In fact, some experts believe that medical breakthroughs have slowed down, attributing this to the prohibitive scale and complexity of present-day medical information. Computers and data mining are best-suited for this purpose [4]. Data Mining is the discovery of unknown

ISBN: 978-1-61804-034-3

2. Medical expert system Medical expert systems for diagnosis, treatment and management of diseases are gaining importance in the

296


practice of modern medicine. Expert system is a program which uses artificial intelligence to provide solutions to difficult questions requiring specialized training to answer. Decision making in Intensive Care Unit (ICU), analysis of practice of respiratory medicine which will benefit from an unbiased guidance. Artificial neural networks, object oriented framework, fuzzy logic, genetic algorithm are some of the recent techniques used for the development of faster, efficient and reliable medical expert systems [14]. Recent years have witnessed an exponential increase in the application of computers in the medical field. Application of computers in biomedical engineering provides unparalleled opportunities to explore newer strategies for prevention, early diagnosis and specific therapy for large number of medical needs. Analyzing the recent developments in medical computer science, it is clear that the trend is to develop new methods for computer aided decision making in medicine to evaluate critically the methods in clinical practice. Development of expert systems in medical field has already emerged as a major area in the application of computers in medicine [15]. Despite some promising examples of the use of expert system to enhance patient care, very few products are routinely used in clinical setting [16]. One of the important factors that have limited the acceptance of decision making support tools for clinicians is the lack of medical staff involvement in the system development [17]. Expert system is a program to provide solutions to difficult questions requiring specialized training to answer. In an expert system, comparison, classification and identification are essential requirements. Providing all possible symptoms and corresponding disease conditions in a computer through programs it can make a sufficiently sound solution to a particular problem, perhaps as good as the case an expert doctor would have made in that context, based on his intelligence, experience and knowledge. But here accuracy of diagnosis depends on the domain expert's doctor contribution and knowledge engineer's skill in the development of the system [18]. When a computer is independently involving in any type of decision making or computing activity, it is said that activity is automated. The biggest single use of Artificial Intelligence (AI) technique today is in expert system. Expert system is designed using AI technique to operate in a real-world domain. The goal is to develop a system which can make expert level judgments and recommendations in the field. In most of the cases it is a rule based. Artificial intelligence application program provides expert quality solutions to problems in a specific domain. Its knowledge is extracted from human experts in the domain and encoded in a formal language and it attempts to emulate their methodology and performance. They are open to inspection, modification and analysis [18].

The expertise comes from education, experience, intelligence and intuition. The effective and efficient integration of the above qualities make the decision making process more meaningful figure (1).

Expert systems can be either rule based or frame based. The various functionary blocks of a typical expert system are user interface, explanation subsystem, knowledge base editor, case specific data, general knowledge base and inference engine [19, 20] figure (2).

User Interface: This is for the interaction between the user and expert system.

The explanation subsystem: This allows the program to explain its reasoning to the user. The explanation includes justification for the system's conclusions.

Knowledge base editor: It is to locate bugs in the program's performance, to assist in the addition of new knowledge, to help maintain correct rule syntax and perform consistency checks on the updated knowledge base.

Case-specific data: The program must track case specific, the facts, conclusions and other relevant information of the case under consideration.

The general Knowledge base: The heart of the expert system is the general knowledge base, which contains the problem solving knowledge of the particular application. It should be knowledge rich.

2.1 Architecture of a typical expert System

ISBN: 978-1-61804-034-3

297


Inference engine: This is a computer term given to that part of the program (software) which applies mainly to the solution of actual problems. It is the interpreter for the knowledge base.

Rule based representation: The use of if then rules (also called production rules or condition-action pairs) is a straight ward popular way to represent the expert know how. The type of knowledge is often said to be shallow. Given a set of facts, the reasoning process is then modeled as the successive activation of rules, which in turn produce new facts to be considered, until no applicable rule can be found.

Frame based representation: They are based on symbolic models of disease entities and their relationships to patient factors and clinical manifestations. Diagnostic expert systems help the doctors to determine the range of alternative diagnosis or a definitive diagnosis systems used for treatment helps in deciding on an action plan where multiple options are available in individual cases. Quick decision making in differential diagnosis and selection of proper treatment in a short time are the features of the medical expert system [2325].

Frame-based representations have been introduced as a way to organize prototypical knowledge about the world. A frame defines the prototypical description of concepts sharing similar properties and behavior. It is defined as a set of slots describing the concept attributes and their rules. The personnel involved in the development of an expert system are the domain expert, the knowledge engineer, and the end user.

2.2 Design of medical expert systems The vast amount of existing medical knowledge and the rapid growth of that knowledge have resulted in a situation where most physicians find it increasingly difficult to assimilate all the information which would be useful in making optimal clinical judgments, even after specialization. It is not surprising that empirical studies have demonstrated that physicians do not always make optimal decisions [21]. Now it has been approved by all concerned that computer assisted medical decision making system, medical expert system, might provide a solution to the problems in diagnosis and patient management. Such systems are generally intended to support and not replace physicians, complementing their natural abilities to make judgments with the computer's memory, reliability and processing capabilities [22]. Medical expert system has evolved from computer science through artificial intelligence figure (3).

ISBN: 978-1-61804-034-3

3. Review of medical expert system in respiratory system In respiratory system, as well as in other field of medicine, the differentiation between normal and abnormal is difficult. Reason is that there are no two subjects exactly alike because every biological quantity is a variable influenced by heredity, environment, occupation, nutrition, age, sex, culture etc. Hence even in perfectly normal subjects most biological quantities are scattered over certain range of values [26]. These facts should be well taken care of during the development of medical expert systems for respiratory system diagnosis. It reveals the requirement of the close association of the domain expert and the knowledge engineer. The scope of developing a wide variety of medical Expert Systems in Respiratory medicine is unlimited. The complex jargon of respiratory physiology and mechanics have made pulmonary function tests very unpopular among general practitioners. The decision making in intensive care units for mechanically ventilated becomes challenging even to specialists because of the complex ventilator mechanics. The evaluation of unexplained breathlessness or assessment of respiratory function in sportsmen is not an easy task. Thus creating a stable, popular, power full and user friendly program can help the physician in many ways. Chronic diseases like asthma, chronic obstructive pulmonary diseases, interstitial lung diseases and tuber cluosis can be managed better if suitable algorithms are created and a set of treatment choices are outlined. This then becomes useful in managing a large group of patients in busy hospitals. Decision making in individual cases also

298


will be made easy then. There is only a limited number of medical expert systems available in this area. Even though the development of medical expert systems started right from the mid 1970s with the development MYCIN [27], the number of articles in computer aided medical diagnosis up to 1979 was only two [28]. Mycin was developed to assist in the treatment of infectious diseases in particular bacterial infections in the blood. In this, in addition to the performance program, there are three adjunct programs that increase system utility and flexibility. After this period, development of some medical expert systems in the area of respiratory system has taken place [29-36] (Table 1). On close observation, it can be seen that each of them are more or less concentrating on very narrow specific areas of respiratory system. Note of the expert systems developed can analyze respiratory system as a whole. Artificial neural network is another artificial intelligence (AI) tool now in use for the development of expert systems. It consists of numerous, simple processing units or neurons that we can globally program for computation. We can program or train neural networks to store, recognize and associatively retrieve patterns or data base entries and to solve different types of optimization problems, i.e., to estimate sampled functions when we do not know the form of the functions. The overall network behaves as an adaptive function estimator. It has been used successfully in many different applications [37-122]

TABLE I. Expert System APACHE11 Ref(29)

CAPNEX Ref(30)

CENTAUR Ref (32)

GUIDON Ref(31)

PUFF Ref(32)

Expert System

VM Ref(32)

WHEESE Ref(32)

WEANPRO Ref(35)

Purpose

Developed for the task of pulmonary function tests.

RESPAID Ref(33)

To support interpretation of respiratory data for ventilated patient.

VENTEX Ref(34)

To support decision making in the management of

ISBN: 978-1-61804-034-3

Comments

failure.

For the management of ventilatory therapy.

It is also a rule based management system in ICU

To interpret measurements of pulmonary function tests to diagnose the severity of lung disease. To assist respiratory therapists and nurses in weaning post-operative cardio vascular surgery patients in ICU

It is also a rule based pulmonary function test analyzer

It suggests ventilator settings which progress the patient towards independent breathing

4. Design of expert system by using artificial neural network Imagine the way the human mind works when presented with a problem. At first, the problem’s facts are analyzed and weighted at some sensorial level. Next, these facts are passed through neural paths, which act as filters and are based on previously known patterns. This leads to conclusions, which may be possible solutions to the problem or may serve as additional facts for a new iteration over the neural paths [13]. Artificial Neural Networks (ANNs) have been extensively used in many research areas from marketing to medicine [9]. They first received much attention from computer scientists, neurophysiologists, psychologists, and engineers, interested in biological nervous system organization and artificial intelligence. Their two main applications in medicine are pattern recognition (classification) and prediction: during these last years (from 1990s and increasing in the 2000s), the applications for prognostic and diagnostic classification in medicine have attracted growing interest in the medical literature [10]. Building Neural Networks (ANNs) have been applied in a wide range of problems and have given, in many cases, superior results to standard statistical models [11]. In particular, the predictive reliability of ANN models has been demonstrated in medical diagnosis [12].

MEDICAL EXPERT SYSTEM IN RESPIRATORY SYSTEM

To help to estimate patient mortality, length of stay and artificial ventilation. To help to perform patient medical diagnosis and instrument fault detection To overcome some of the problems PUFF could not handle well To teach facts and problem solving strategies related to pulmonary function analysis.

Purpose

ventilatory therapy.

Comments

It is for a larger patient management system for use in the ICU. It is based on clinical capnography data and experience. It is a rule based pulmonary function analysis system.

Neural networks are pattern- learning instruments that are more sophisticated than decision trees and Naïve Bayes. Figure (4) shows a couple of examples. A neural network contains a set of nodes (neurons) and edges that form a network. There are three types of nodes: input, hidden, and output. Each edge links two nodes with an associated weight. The direction of an edge represents the data flow during the prediction process. Each node is a unit of processing.

It uses the knowledge base for MYCIN. It is only a tutor package. It used standard pulmonary function tests and laboratory measurements for diagnosis. It has been developed in order to acquire automatic collection of data from standard monitors and/ respirators. The system is intended to be used in the case of patients with respiratory

299


Input nodes form the first layer of the network. In most neural networks, each input node is mapped to one input attribute (such as productive cough, stridor, or fever). Hidden nodes are the nodes in the intermediate layers. A hidden node receives input from nodes in the input layers or precedent hidden layer. It combines all the input based on the weight of associated edges, processes some calculations, and emits a result value of the processing to the subsequent layer. Output nodes usually represent the predictable attributes. A neural network may have multiple output attributes, combining several output nodes in the same network reduces the processing time because such networks can share the common cost of scanning the source data. The result of the output node is often a floating number between 0 and 1 [13]. In our medical application input nodes represent the symptoms of the eight cases of respiratory disease. Such as dry cough, productive cough, fever, heamoptysis, tachypnea, chest pain, dysnea and etc. also using medical sign such as bronchial breathing, cyanosis, dullness on percussion, hyper resonant on percussion, pleural rub, respiratory distress, wheezing and et al. In addition to some medical investigation as X-ray [showing lung consolidation], X-ray [shawing edematous epiglottic], Xray [shawing subglottic narrowing and classic narrow trachea], X-ray [showing diffuse haziness]. The output nodes represent eight cases of respiratory disease. As we see in our study that the output value is a floating number between 0 and 1. This means that our result is as the previous concept. And the original value of an input attribute must be massaged to a floating number in the same scale (often between -1 and 1) before processing [13].

Figure (5): sample of the inputs used for Training the Neural Network.

Figure (6): sample of target data (output)

We use MATLAB source and train the network with the input vector and the target vector as described in figure (7).

Use the input vector data set. This data set consists of 699 Twenty-seven-element input vectors and eightelement target vectors. Input vector contain 699 cases with some common symptoms and different in some symptoms. The results of examination for these cases are described in figure (6).

ISBN: 978-1-61804-034-3

300


Figure (9): training curve true and false position rate

Figure (10): validation curve true and false position rate

Figure (7): Neural Network Training screen

We can see that the mean Square Error is correct of the trained data, validation data, and test data, all take the same curve. This means that the network learns from cases and this depends on the amount of the input cases. If the input cases are large then the out put cases in test cases is extremely correct. Figure (11): test curve true and false position rate

Figure (12): all curves true and false position rate Figure (8): mean Squared Error for (Train, Validation, Test) data

After we train network with input and target vectors now we test the network by enter twenty cases to the network

ISBN: 978-1-61804-034-3

301


with out the target vector. And let the network to give us the right examination that is one of eight cases.

some different cases with out the target output vector and the network gives us the correct diagnosis. So the artificial neural network is very useful data mining technique for recognition of pediatric respiratory disease.

References [1] Wilson, A.M; Thabane, L.and Holbrook, A. (2003): "Application of data mining techniques in pharmacovigilance". British Journal of Clinical Pharmacology. (57) 2, 127-134. [2] Ruben, D.; Canlas, Jr. and Msit, M.B.A. (2009): DATA MINING IN HEALTHCARE: CURRENT APPLICATIONS AND ISSUES. Australia: Carnegie Mellon University. [3] Witten, I. H. and Frank, E. (2005): Data mining: practical machine learning tools and techniques. Morgan Kaufmann series in data management systems. Morgan Kaufman. [4] Shillabeer, A. and Roddick, J. (2007): Establishing a Lineage for Medical Knowledge Discovery. ACM International Conference Proceeding Series. (311) 70, 29-37. [5] Ming-Syan, C.; Jiawei, H. and Philip, S.Yu. (1996): Data Mining: An Overview from a Database Perspective, IEEE ransactions on Knowledge and Data Engineering, Vol. 8, No. 6, pp. 866-883. [6] Usama, M.; Fayyad, Gregory P. and Padhraic S. (1996): From Data Mining to knowledge Discovery: An Overview, Advances in Knowledge Discovery and Data Mining, AAAI Press, pp 134. [7] Agrawal, R.; Imielinski, T. and Swami, A. (1993): Mining Associations between Sets of Items in Massive Databases, Proc. of the ACM-SIGMOD Int'l Conference on Management of Data,Washington D.C. [8] Ullah, I. (2010): Data Mining Algorithms and Medical Sciences. International Journal of Computer Science & Information Technology (IJCSIT), Vol. 2, No 6. [9] Lisboa, P. J. G. (2002): “A review of evidence of health benefit from artificial neural networks in medical intervention,” Neural Networks, vol. 15, no. 1, pp. 11–39. [10] Bourdès, V.; Bonnevay, S.; Lisboa, P.; Defrance, R.; Pérol, D.; Chabaud, S.; Bachelot, T.; Gargi, T.and Négrier, S. (2010): Comparison of Artificial Neural Network with Logistic Regression as Classification Models for Variable Selection for Prediction of Breast Cancer Patient Outcomes. Volume 2010, Article ID 309841, 10 pages. doi:10.1155/2010/309841. [11] White, H. (1989): “Learning in artificial neural networks: a statistical approach,” Neural Computation, vol. 1, pp. 425–464. [12] Baxt, W. G. (1995): “Application of artificial neural networks to clinical medicine,” Lancet, vol. 346, no. 8983, pp. 1135–1138. [13] MacLennan, J.; Tang, Z. and Crivat, B. (2008): Data Mining with Microsoft SQL Server. Copyright© 2009 by Wiley Publishing, Inc., Indianapolis, Indiana, pp. 371-387. [14] Sukesh, K. A.; jayakrishnan, B. and Kandaswamy, A. (1998): medical expert systems in respiratory system. Lung India, XVL, No. 4 (P 162-197). [15] Wyatt, J. (1991): computerbased Knowledge systems. Lancet 338: 1431-1435. [16] Morell, R.A.; Bronzino, J.D. and Goethe, J.W. (1987): Expert systems in psychiatry: A review. Proceeding of 20th Hawaii international conference on Systems Science 3:84. [17] Sukesh, K. A. and Kandaswamy, A.(1997) : Role of medical expert systems in health care in the Indian context. Defence Science Journal, DRDO, India 1997; 47:499-504. [18] Mathew, M.A. (1994): lecture notes for a course on artificial intelligence and expert systems. 1st ed. India: Tech publishers 10-107.

Figure (13): data used in testing the network

As we see in figure (14) the result of test data. The first column indicate that the first value in column is 0.8130 ≈ 1 and all values follow it ≈ to zero this mean that the patient examination is pneumonia and so on.

Figure (14): examination result of the test data

5.

DISCUSSION

In the presented study we have trained the network with large amount of data and we see that as we train the network, the network can give correct decision about the correct examination to the entered symptoms. As we know that our study is about pediatric respiratory disease. We also see that the network learning way simulation of human learn manner. Human learn depends on the amount of cases that we face. Artificial neural networks also learns after it faces 699 cases with different symptoms. The computer program was performed under MATLAB software using the neural network toolbox. In the training, the number of neurons on the hidden layer is 20. A dataset including 699 data samples obtained from experimental studies were used for ANNs. From these, 699 data patterns were used for training the network, and the remaining 20% patterns were randomly selected and used as the test dataset. The results are shown in Figs.5~14. After training the network we test it with

ISBN: 978-1-61804-034-3

302


[37] Hazem M. El-Bakry, "Fast Virus Detection by using High Speed Time Delay Neural Networks," Journal of Computer Virology, vol.6, no.2, 2010, pp.115-122. [38] Hazem M. El-Bakry, "A Novel High Speed Neural Model for Fast Pattern Recognition," Soft Computing Journal, vol. 14, no. 6, 2010, pp. 647-666. [39] Hazem M. El-Bakry, "An Efficient Algorithm for Pattern Detection using Combined Classifiers and Data Fusion," Information Fusion Journal, vol. 11, 2010, pp. 133-148. [40] Hazem M. El-bakry, Ahmed Atwan, and Nikos Mastorakis “Aِ New Technique for Realization of Boolean Functions, " Proc. of Recent Advances in Artificial Intelligence, Koweledge Engineering and Databases, Cambridge, UK, February 20-22, 2010, pp. 260-270. [41] Hazem M. El-Bakry, "New Fast Principal Component Analysis For Real-Time Face Detection," MG&V Journal, vol. 18, no.4, 2009, pp. 405-426. [42] Hazem M. El-Bakry, and Nikos Mastorakis, "Fast Word Detection in a Speech Using New High Speed Time Delay Neural Networks," WSEAS Transactions on Information Science and Applications, issue 7, vol. 5, July 2009, pp. 261270. [43] Hazem M. El-Bakry, and Nikos Mastorakis, "Fast Information Retrieval from Web Pages," WSEAS Transactions on Information Science and Applications, issue 6, vol. 6, June 2009, pp. 1018-1036. [44] Hazem M. El-Bakry, and Nikos Mastorakis, "Fast Image Matching on Web Pages," WSEAS Transactions on Signal Processing, issue 4, vol. 5, June 2009, pp. 157-166. [45] Hazem M. El-Bakry, and Nikos Mastorakis, "Fast Detection of Specific Information in Voice Signal over Internet Protocol," WSEAS Transactions on Communications, issue 5, vol. 8, May 2009, pp. 483-494. [46] Hazem M. El-Bakry, "A New Neural Design for Faster Pattern Detection Using Cross Correlation and Matrix Decomposition," Neural World journal, 2009, vol. 19, no. 2, pp. 131-164. [47] Hazem M. El-Bakry, and Nikos Mastorakis, "A New Technique for Detecting Dental Diseases by using High Speed Artificial Neural Network," WSEAS Transactions on Computers, Issue 12, vol. 7, December 2008, pp. 1977-1987. [48] Hazem M. El-Bakry, and Nikos Mastorakis, " An Effective Method for Detecting Dental Diseases by using Fast Neural Networks," WSEAS Transactions on Biology and Biomedicine, issue 11, vol. 5, November 2008, pp. 293-301. [49] Hazem M. El-Bakry, and Nikos Mastorakis, "A Novel Fast Kolmogorov’s Spline Complex Network for Pattern Detection," WSEAS Transactions on Systems, Issue 11, vol. 7, November 2008, pp. 1310-1328. [50] Hazem M. El-Bakry, and Nikos Mastorakis, "A New Fast Forecasting Technique using High Speed Neural Networks ," WSEAS Transactions on Signal Processing, Issue 10, vol. 4, October 2008, pp. 573-595. [51] Hazem M. El-Bakry and Mohamed Hamada, "A New Implementation for High Speed Neural Networks in Frequency Space," Lecture Notes in Artificial Intelligence, Springer, KES 2008, Part I, LNAI 5177, pp. 33-40. [52] Hazem M. El-Bakry, "New Faster Normalized Neural Networks for Sub-Matrix Detection using Cross Correlation in the Frequency Domain and Matrix Decomposition," Applied Soft Computing journal, vol. 8, issue 2, March 2008, pp. 11311149. [53] Hazem M. El-Bakry, "New Fast Principal Component Analysis for Face Detection," Journal of Advanced

[19] Charnlak, E. and McDermott, D. (1985): introduction to artificial intelligence. Addison-Wesley publishing company, 140-180. [20] Levine, R. L.; Drange, D.E. and Edelson, B. (1986): A comprehensive guide to AI and Expert Systems. Mc Graw-Hill International editions, New York 40-85. [21] Koran, L.M. (1975): The reliability of Clinical methods, data and Judgements. N Engl J med 293:695-701. [22] Reggia, J.A. and Tuhrim, S. E. (1985): Computer-Assisted Medical Decision making, Vol. 1. New York, Springer- Verlag 5-25. [23] Redier, H.; Daures, J.P.; Michel, C.; proundhon, H.; Vervloet, D.; Charpin, D., Marsac, J.; Dusser, D.; Brambilla, C.; Wallaert, B.; Kopferschmitt, M.C.; Pauli, G.; Taytard, A.; Cogis, O.; Michel, F.B. and Godard, P.(1995): Assessment of the severity of asthma by an Expert System: description and evaluation. Am J Respir Crit Care med 151:345-52. [24] Hickam, D.H.; Shortiffe, E.H.; Bischoff, M.B.; Scott, C.A. and Jacobs, C.D. (1985): The treatment advice of a computerBased cancer Chemotherapy protocol Advisor. Ann intern Med 103:928-936. [25] Shahsaver, N.; Ludwigs, V.; Blomquist, H.; Gill, H.; Wigertz, O. and Matell, G. (1995): Evaluation of a knowledge- based decision- support system for ventilator therapy management. Artificial intelligence in med 7:37-52. [26] Webster, J.G. (1988): Encyclopedia of medical devices and instrumentation. 10th edition, USA, John Wiley & sons Inc, 4:2379-95. [27] Shortiffe, E.H. (1976): Computer- based medical consultation, MYCIN. New York. Elsevier 25-105. [28] Rogers, W. (1979): computer-aided medical diagnosis: Literature review: Int J Biomed Comp 10:267-289. [29] Timothy, B.; Stevenson; maryhelen, S.; Monique, F. and Fred, S. (1994): Estimation of ventilation, length of stay and mortality using artificial neural networks. Proceeding of the 1994 canadian conference on electrical and computer engineering, Halifax, Canada 2:726-729. [30] Ventzas, D.E. and CAPNEX (1994): An expert system for Capnography (co2 respiration analysis). Transactions of the institute of measurement and Control 16:233-244. [31] Clancy, W.J. (1979): Intelligent computer-aided instruction for medical diagnosis. Proceedings of the 3rd symposium on computer applications in medical care 173-183. [32] Lecky, J.H.; Matsiras, P.V.; Garfinkel, D.; Aukburg, S.J.; Carson, E.R. and PONI (1988): A prototype respiratory and circulatory monitoring system for operating rooms. IEEE Engineering in medicine & biology society. 10th annual international conference 1406-1407. [33] Miller, R.K. and Walker, T.C. (1988): Artificial intelligence Applications in Sensors and Instrumentation, 10th ed, USA, SEAI Technical Publications, 96-111. [34] Chambrin, M.C.; Chopin, C.; Ravaux, P.; Mangalaboyi, J.; Lestavel, O.; Fourrier, F. and RESPAID (1989): computer aided decision support for respiratory data in ICU. IEEE Engineering in medicine & biology society. 11th annual international conference 1776-77. [35] Shahsaver, N.; Gill, H.; Ludwigs, U.; Carstensen, A.; Larsson, H.; Wigertz, O.; matell, G. and Ventex (1994): an on line knowledge based system to support ventilator management. Technology and health care 1:233-243. [36] Tong, D.A.; Golden, E.B.; Napper, S.A. and WEANPRO (1989): A weaning protocol Expert system. IEEE Engineering in medical & biology society. 11th annual international conference, 1857-58.

ISBN: 978-1-61804-034-3

303


Computational Intelligence and Intelligent Informatics, vol.11, no.2, 2007, pp. 195-201. [54] Hazem M. El-Bakry, and Nikos Mastorakis "New Fast Normalized Neural Networks for Pattern Detection," Image and Vision Computing Journal, vol. 25, issue 11, 2007, pp. 17671784. [55] Hazem M. El-Bakry and Nikos Mastorakis, "Fast Code Detection Using High Speed Time Delay Neural Networks," Lecture Notes in Computer Science, Springer, vol. 4493, Part III, May 2007, pp. 764-773. [56] Hazem M. El-Bakry, "New Fast Time Delay Neural Networks Using Cross Correlation Performed in the Frequency Domain," Neurocomputing Journal, vol. 69, October 2006, pp. 2360-2363. [57] Hazem M. El-Bakry, and Nikos Mastorakis, "A Novel Model of Neural Networks for Fast Data Detection," WSEAS Transactions on Computers, Issue 8, vol. 5, November 2006, pp. 1773-1780. [58] Hazem M. El-Bakry, and Nikos Mastorakis, "A New Approach for Fast Face Detection," WSEAS Transactions on Information Science and Applications, issue 9, vol. 3, September 2006, pp. 1725-1730. [59] Hazem M. El-Bakry, "A New Implementation of PCA for Fast Face Detection," International Journal of Intelligent Technology, Vol. 1, No.2, 2006, pp. 145-153. [60] Hazem M. El-Bakry, and Qiangfu Zhao, "Fast Normalized Neural Processors For Pattern Detection Based on Cross Correlation Implemented in the Frequency Domain," Journal of Research and Practice in Information Technology, Vol. 38, No.2, May 2006, pp. 151-170. [61] Hazem M. El-Bakry, "Fast Painting with Different Colors Using Cross Correlation in the Frequency Domain," International Journal of Computer Science, vol.1, no.2, 2006, pp. 145-156. [62] Hazem M. El-Bakry, "Faster PCA for Face Detection Using Cross Correlation in the Frequency Domain," International Journal of Computer Science and Network Security, vol.6, no. 2A, February 2006, pp.69-74. [63] Hazem M. El-Bakry, "New High Speed Normalized Neural Networks for Fast Pattern Discovery on Web Pages," International Journal of Computer Science and Network Security, vol.6, No. 2A, February 2006, pp.142-152. [64] Hazem M. El-Bakry, and Qiangfu Zhao, "Fast Time Delay Neural Networks," International Journal of Neural Systems, vol. 15, no.6, December 2005, pp.445-455. [65] Hazem M. El-Bakry, and Qiangfu Zhao, "Speeding-up Normalized Neural Networks For Face/Object Detection," Machine Graphics & Vision Journal (MG&V), vol. 14, No.1, 2005, pp. 29-59. [66] Hazem M. El-Bakry, and Qiangfu Zhao, "A New Technique for Fast Pattern Recognition Using Normalized Neural Networks," WSEAS Transactions on Information Science and Applications, issue 11, vol. 2, November 2005, pp. 1816-1835. [67] Hazem M. El-Bakry, "Pattern Detection Using Fast Normalized Neural Networks," Lecture Notes in Computer Science, Springer, vol. 3696, September 2005, pp. 447-454. [68] Hazem M. El-Bakry, "Human Face Detection Using New High Speed Modular Neural Networks," Lecture Notes in Computer Science, Springer, vol. 3696, September 2005, pp. 543-550. [69] Hazem M. El-Bakry, and Qiangfu Zhao, "Fast Pattern Detection Using Normalized Neural Networks and Cross Correlation in the Frequency Domain," EURASIP Journal on Applied Signal Processing, Special Issue on Advances in

ISBN: 978-1-61804-034-3

Intelligent Vision Systems: Methods and Applications—Part I, vol. 2005, no. 13, 1 August 2005, pp. 2054-2060. [70] Hazem M. El-Bakry, "A New High Speed Neural Model For Character Recognition Using Cross Correlation and Matrix Decomposition," International Journal of Signal Processing, vol.2, no.3, 2005, pp. 183-202. [71] Hazem M. El-Bakry, and Qiangfu Zhao, "A Fast Neural Algorithm for Serial Code Detection in a Stream of Sequential Data," International Journal of Information Technology, vol.2, no.1, pp. 71-90, 2005. [72] Hazem M. El-Bakry, and Qiangfu Zhao, "Fast Complex Valued Time Delay Neural Networks," International Journal of Computational Intelligence, vol.2, no.1, pp. 16-26, 2005. [73] Hazem M. El-Bakry, "Face detection using fast neural networks and image decomposition," Neurocomputing Journal, vol. 48, October 2002, pp. 1039-1046. [74] Hazem M. El-Bakry, "Human Iris Detection Using Fast Cooperative Modular Neural Nets and Image Decomposition," Machine Graphics & Vision Journal (MG&V), vol. 11, no. 4, 2002, pp. 498-512. [75] Hazem M. El-Bakry, "Fast Iris Detection for Personal Verification Using Modular Neural Networks," Lecture Notes in Computer Science, Springer, vol. 2206, October 2001, pp. 269283. [76] Hazem M. El-bakry, and Nikos Mastorakis, “An Intelligent Approach for Fast Detection of Biological Viruses in DNA Sequence,” Proc. of Int. Conf., Spain, March 29-31, 2011, pp. 237-244. [77] Hazem M. El-bakry, and Nikos Mastorakis, “A New Approach for Prediction by using Integrated Neural Networks,” Proc. of Int. Conf., Mexico, Jan. 29-31, 2011, pp. 17-28. [78] Hazem M. El-bakry, and Mohamed Hamada, “A Developed Watermark Technique for Distributed Database,” Proc. of 3rd International Conference on Computational Intelligence in Information Security (CISIS'10), Spain, November 11-12, 2010, pp. 173-180. [79] Hazem M. El-bakry, and Mohamed Hamada, “A Novel Watermark Technique for Relational Databases,” Proc. of Artificial Intelligence and Computational Intelligence International Conference, AICI 2010, Sanya, China, October 2324, 2010, pp. 226-232. [80] Hazem M. El-bakry, “Fast Karnough Map for Simplification of Complex Boolean Functions,” Proc. of 10th WSEAS International Conference on Applied Computer Science (ACS'10), Japan, October 4-6, 2010, pp. 478-483. [81] Hazem M. El-bakry, and Nikos Mastorakis, “Prediction of Market Price by using Fast Time Delay Neural Networks,” Proc. of 10th WSEAS Int. Conf. on NEURAL NETWORKS (NN'10), Romania, June 13-15, 2010, pp. 230-237. [82] Hazem M. El-bakry, and Nikos Mastorakis “Studying the Efficiency of XML Web Services for Real-Time Applications" Proc. of WSEAS International Conference on Sensors, Signals, Visualization, Imaging, Simulation and Materials, USA, November 7-9, 2009, pp.209-219. [83] Hazem M. El-bakry, and Mohamed Hamada “Fast Time Delay Neural Networks for Detecting DNA Coding Regions,” Proc. of Kes 2009, Part I, LNAI 5711, ٍSpringer, September 2830, 2009, pp. 334-342. [84] Hazem M. El-bakry, and Nikos Mastorakis “Ear Recognition by Using Neural Networks " Proc. of APPLIED COMPUTING CONFERENCE 2009 (ACC '09), Vouliagmeni Beach, Athens, Greece, September 28-30, 2009, pp.770 -804. [85] Hazem M. El-bakry, and Nikos Mastorakis “Fast Human Motion Tracking by using High Speed Neural Networks " Proc.

304


of 9th WSEAS International Conference on SIGNAL, SPEECH AND IMAGE PROCESSING (SSIP '09), Budapest, Hungry, September 3-5, 2009, pp. 221-240. [86] Hazem M. El-bakry, and Nikos Mastorakis “A Fast Computerized Method For Automatic Simplification of Boolean Functions,” Proc. of 9th WSEAS International Conference on SYSTEMS THEORY AND SCIENTIFIC COMPUTATION (ISTASC '09), Moscow, Russia, August 26-28, 2009, pp. 99107. [87] Hazem M. El-bakry, and Nikos Mastorakis “User Interface for Internet Applications,” Proc. of 9th WSEAS International Conference on Applied Informatics and Communications (AIC'09), Moscow, Russia, August 26-28, 2009, pp. 383-392. [88] Hazem M. El-bakry, and Nikos Mastorakis “Fast Information Processing over Business Networks,” Proc. of 9th WSEAS International Conference on Applied Informatics and Communications (AIC'09), Moscow, Russia, August 26-28, 2009, pp.305-324. [89] Hazem M. El-bakry, “Fast Record Detection in Large Databases Using New High Speed Time Delay Neural Networks,” Proc. of IEEE IJCNN’09, Atlanta, USA, June 14-19, 2009, pp. 757-763. [90] Hazem M. El-bakry, and Mohamed Hamada “Fast Principal Component Analysis for Face Detection Using CrossCorrelation and Image Decomposition,” Proc. of IEEE IJCNN’09, Atlanta, USA, June 14-19, 2009, pp. 751-756. [91] Hazem M. El-bakry, and Nikos Mastorakis “A Simple Design and Implementation of Reconfigurable Neural Networks Detection,” Proc. of IEEE IJCNN’09, Atlanta, USA, June 14-19, 2009, pp. 744-750. [92] Hazem M. El-bakry, and Nikos Mastorakis “Fast Virus Detection by using High Speed Time Delay Neural Networks,” Proc. of 10th WSEAS Int. Conf. on NEURAL NETWORKS (NN'09), Prague, Czech Repulic, March 22-25, 2008, pp. 169183. [93] Hazem M. El-bakry, and Nikos Mastorakis “Fast Detection of Specific Information in Voice Signal over Internet Protocol,” Proc. of 7th WSEAS Int. Conf. on COMPUTATIONAL INTELLIGENCE, MAN-MACHINE SYSTEMS and CYBERNETICS (CIMMACS '08), Cairo, EGYPT, Dec. 29-31, 2008, pp. 125-136. [94] Hazem M. El-bakry, and Nikos Mastorakis, “Information Retrieval Based on Image Detection on Web Pages,” Proc. of 7th WSEAS Int. Conf. on COMPUTATIONAL INTELLIGENCE, MAN-MACHINE SYSTEMS and CYBERNETICS (CIMMACS '08), Cairo, EGYPT, Dec. 29-31, 2008, pp. 221228. [95] Hazem M. El-bakry, and Nikos Mastorakis, “Fast Information Retrieval from Web Pages,” Proc. of 7th WSEAS Int. Conf. on COMPUTATIONAL INTELLIGENCE, MANMACHINE SYSTEMS and CYBERNETICS (CIMMACS '08), Cairo, EGYPT, Dec. 29-31, 2008, pp. 229-247. [96] Hazem M. El-Bakry and Mohamed Hamada, "New Fast Decision Tree Classifier for Identifying Protein Coding Regions," Proc. of ISICA 2008 Conf., China, Dec. 3-5, 2008, pp. 489-500. [97] Hazem M. El-Bakry, and Nikos Mastorakis, " An Effective Method for Detecting Dental Diseases by using Fast Neural Networks, " 8th WSEAS International Conference on SIGNAL, SPEECH AND IMAGE PROCESSING (SSIP '08), Santander, Cantabria, Spain, September 23-25, 2008, pp. 144-152. [98] Hazem M. El-Bakry, and Nikos Mastorakis, " A New Fast Forecasting Technique using High Speed Neural Networks, " 8th WSEAS International Conference on SIGNAL, SPEECH AND

ISBN: 978-1-61804-034-3

IMAGE PROCESSING (SSIP '08), Santander, Cantabria, Spain, September 23-25, 2008, pp. 116-138. [99] Hazem M. El-Bakry, and Nikos Mastorakis, " A Novel Fast Kolmogorov's Spline Complex Network for Pattern Detection," 8th WSEAS International Conference on SIMULATION, MODELLING and OPTIMIZATION (SMO '08), Santander, Cantabria, Spain, September 23-25, 2008, pp. 261-279. [100] Hazem M. El-Bakry, and Nikos Mastorakis, " A New Technique for Detecting Dental Diseases by using High Speed Neuro-Computers," European computing Conf. (ECC '08), Malta, September 11-13, 2008, pp. 432-440. [101] Hazem M. El-Bakry, and Nikos Mastorakis, " A Real-Time Intrusion Detection Algorithm for Network Security, " 8st WSEAS International Conference on Applied Informatics and Communications (AIC '08), Rhodes, Greece, August 20-22, 2008, pp. 533-545. [102] Hazem M. El-bakry, and Qiangfu Zhao, “Fast Neural Implementation of PCA for Face Detection,” Proc. of IEEE World Congress on Computational Intelligence, IJCNN’06, Vancouver, BC, Canada, July 16-21, 2006, pp. 1785-1790. [103] Hazem M. El-bakry, “A Simple Design for High Speed Normalized Neural Networks Implemented in the Frequency Domain for Pattern Detection,” Proc. of IEEE World Congress on Computational Intelligence, IJCNN’06, Vancouver, BC, Canada, July 16-21, 2006, pp. 2296-2303. [104] Hazem M. El-bakry, “Fast Co-operative Modular Neural Processors for Human Face Detection,” Proc. of IEEE World Congress on Computational Intelligence, IJCNN’06, Vancouver, BC, Canada, July 16-21, 2006, pp. 2304-2311. [105] Hazem M. El-bakry, “New Fast Time Delay Neural Networks Using Cross Correlation Performed in the Frequency Domain,” Proc. of IEEE World Congress on Computational Intelligence, IJCNN’06, Vancouver, BC, Canada, July 16-21, 2006, pp. 4990-4997. [106] Hazem M. El-Bakry, and Nikos Mastorakis, “A Novel Model of Neural Networks for Fast Data Detection,” Proc. of the 7th WSEAS International Conference on Neural Networks, Cavtat, Croatia, June 12-14, 2006, pp. 144-151. [107] Hazem M. El-Bakry, and Nikos Mastorakis, “A New Approach for Fast Face Detection,” Proc. of the 7th WSEAS International Conference on Neural Networks, Cavtat, Croatia, June 12-14, 2006, pp. 152-157. [108] Hazem M. El-Bakry, “Pattern Detection Using Fast Normalized Neural Networks,” Proc. of 15th International Conf. on Artificial Neural Nets "ICANN 2005", Warsaw, Poland, September 11-15, 2005, pp. 447-454. [109] Hazem M. El-Bakry, “Human Face Detection Using New High Speed Modular Neural Networks,” Proc. of 15th International Conf. on Artificial Neural Nets "ICANN 2005", Warsaw, Poland, September 11-15, 2005, pp. 543-550. [110] Hazem M. El-bakry, and Qiangfu Zhao, “Modified Time Delay Neural Networks For Fast Data Processing,” Proc. of IEEE Eighth International Symposium on Signal Processing and its Applications, Sydney, Australia, August 28-31, 2005, pp.583586. [111] Hazem M. El-bakry, and Qiangfu Zhao, “Fast Co-Operative Modular Neural Networks for Fast Human Face Detection,” Proc. of IEEE Eighth International Symposium on Signal Processing and its Applications, Sydney, Australia, August 2831, 2005, pp. 679-682. [112] Hazem M. El-bakry, and Qiangfu Zhao, “Fast Pattern Detection Using Neural Networks and Cross Correlation in the Frequency Domain,” Proc. of IEEE IJCNN’05, Quebec, Montreal, Canada, July 31- August 4, 2005, pp. 1900-1905.

305


[113] Hazem M. El-bakry, and Qiangfu Zhao, “Normalized Neural Networks for Fast Pattern Detection,” Proc. of IEEE IJCNN’05, Quebec, Montreal, Canada, July 31- August 4, 2005, pp. 1889-1894. [114] Hazem M. El-bakry, and Qiangfu Zhao, “A Modified Frequency Domain Cross Correlation Implemented in MATLAB for Fast Sub-Image Detection Using Neural Networks,” Proc. of IEEE IJCNN’05, Quebec, Montreal, Canada, July 31- August 4, 2005, pp. 1794-1799. [115] Hazem M. El-Bakry, and Herbert Stoyan "Fast Neural Networks for Sub-Matrix (Object/Face) Detection," Proc. of IEEE International Symposium on Circuits and Systems, Vancouver, Canada, 23-26 May, 2004. [116] Hazem M. El-bakry, “Complexity Reduction Using Modular Neural Networks,” Proc. of IEEE IJCNN’03, Portland, Oregon, pp. 2202-2207, July, 20-24, 2003. [117] Hazem El-Bakry: “Comments on Using MLP and FFT for Fast Object/Face Detection,” Proc. of IEEE IJCNN’03, Portland, Oregon, pp. 1284-1288, July, 20-24, 2003. [118] Hazem El-Bakry, “Face Detection Using Neural Networks and Image Decomposition,” Proc. of INNS-IEEE International Joint Conference on Neural Networks, 12-17 May, 2002, Honolulu, Hawaii, USA. [119] Hazem M. El-bakry, “Human Face Detection Using Fast Neural Networks and Image Decomposition,” Proc. the fifth International Conference on Knowledge-Based Intelligent Information & Engineering Systems 6-8 September 2001, Osaka-kyoiku University, Kashiwara City, Japan, pp. 13301334. [120] Hazem M. El-bakry, “Human Iris Detection Using Fast Cooperative Modular Neural Nets,” Proc. of INNS-IEEE International Joint Conference on Neural Networks, pp. 577-582, 14-19 July, 2001, Washington, DC, USA. [121] Hazem M. El-bakry, M. A. Abo-elsoud, and M. S. Kamel, “ Fast Modular Neural Networks for Human Face Detection,” Proc. of IEEE-INNS-ENNS International Joint Conference on Neural Networks, Como, Italy, 24-27 July, 2000, Vol. III, pp. 320-324. [122] Hazem M. El-Bakry, "Automatic Human Face Recognition Using Modular Neural Networks," Machine Graphics & Vision Journal (MG&V), 2001, vol. 10, no. 1, pp. 47-73.

ISBN: 978-1-61804-034-3

306