Search

Cardiac function is reduced after acute myocardial infarction due to myocardial injury and to changes in the viable non-ischemic myocardium, a process known as cardiac remodeling. Current treatment of patients with acute myocardial infarction (AMI) reduces infarct size, preserves left ventricular function, and improves survival. However, it does not prevent remodeling which leads to heart failure. The aim of the present study was to model the echocardiographically estimated data with respect to the surgically collected data using Neural Networks. In particular, we attempted to analyze the relationship between cardiac remodeling variables obtained from echo and the infarct variables obtained from surgical data using neural networks. Towards that purpose, 199 rats were separated in two groups. The first group was subjected to coronary artery ligation, while the second underwent a sham operation. Echocardiography was used for rat monitoring. Scar weight and area were estimated after surgical incision. It appeared that several factors could be modelled with neural networks. Such modeling approaches could be developed to enable the simulation of the pathophysiological process after an Acute Myocardial Infarction (AMI) and predict with accuracy the effects of novel or current treatments that act via modulation of tissue injury, Left Ventricular dilation, geometry and hypertrophy.

Background Through this paper, we present the initial steps for the creation of an integrated platform for the provision of a series of eHealth tools and services to both citizens and travelers in isolated areas of thesoutheast Mediterranean, and on board ships travelling across it. The platform was created through an INTERREG IIIB ARCHIMED project called INTERMED. Methods The support of primary healthcare, home care and the continuous education of physicians are the three major issues that the proposed platform is trying to facilitate. The proposed system is based on state-of-the-art telemedicine systems and is able to provide the following healthcare services: i) Telecollaboration and teleconsultation services between remotely located healthcare providers, ii) telemedicine services in emergencies, iii) home telecare services for "at risk" citizens such as the elderly and patients with chronic diseases, and iv) eLearning services for the continuous training through seminars of both healthcare personnel (physicians, nurses etc) and persons supporting "at risk" citizens. These systems support data transmission over simple phone lines, internet connections, integrated services digital network/digital subscriber lines, satellite links, mobile networks (GPRS/3G), and wireless local area networks. The data corresponds, among others, to voice, vital biosignals, still medical images, video, and data used by eLearning applications. The proposed platform comprises several systems, each supporting different services. These were integrated using a common data storage and exchange scheme in order to achieve system interoperability in terms of software, language and national characteristics. Results The platform has been installed and evaluated in different rural and urban sites in Greece, Cyprus and Italy. The evaluation was mainly related to technical issues and user satisfaction. The selected sites are, among others, rural health centers, ambulances, homes of "at-risk" citizens, and a ferry. Conclusions The results proved the functionality and utilization of the platform in various rural places in Greece, Cyprus and Italy. However, further actions are needed to enable the local healthcare systems and the different population groups to be familiarized with, and use in their everyday lives, mature technological solutions for the provision of healthcare services.

Background: In the development of tissue classification methods, classifiers rely on significant differences between texture features extracted from normal and abnormal regions. Yet, significant differences can arise due to variations in the image acquisition method. For endoscopic imaging of the endometrium, we propose a standardized image acquisition protocol to eliminate significant statistical differences due to variations in: (i) the distance from the tissue (panoramic vs close up), (ii) difference in viewing angles and (iii) color correction. Methods: We investigate texture feature variability for a variety of targets encountered in clinical endoscopy. All images were captured at clinically optimum illumination and focus using 720 × 576 pixels and 24 bits color for: (i) a variety of testing targets from a color palette with a known color distribution, (ii) different viewing angles, (iv) two different distances from a calf endometrial and from a chicken cavity. Also, human images from the endometrium were captured and analysed. For texture feature analysis, three different sets were considered: (i) Statistical Features (SF), (ii) Spatial Gray Level Dependence Matrices (SGLDM), and (iii) Gray Level Difference Statistics (GLDS). All images were gamma corrected and the extracted texture feature values were compared against the texture feature values extracted from the uncorrected images. Statistical tests were applied to compare images from different viewing conditions so as to determine any significant differences. Results: For the proposed acquisition procedure, results indicate that there is no significant difference in texture features between the panoramic and close up views and between angles. For a calibrated target image, gamma correction provided an acquired image that was a significantly better approximation to the original target image. In turn, this implies that the texture features extracted from the corrected images provided for better approximations to the original images. Within the proposed protocol, for human ROIs, we have found that there is a large number of texture features that showed significant differences between normal and abnormal endometrium. Conclusion: This study provides a standardized protocol for avoiding any significant texture feature differences that may arise due to variability in the acquisition procedure or the lack of color correction. After applying the protocol, we have found that significant differences in texture features will only be due to the fact that the features were extracted from different types of tissue (normal vs abnormal). © 2007 Neofytou et al licensee BioMed Central Ltd. 6 Tradenames: IP4.1 RGB video camera, Circon Manufacturers: Circon Cited By :10

The objective of this study was to classify hysteroscopy images of the endometrium based on texture analysis for the early detection of gynaecological cancer. A total of 418 Regions of Interest (ROIs) were extracted (209 normal and 209 abnormal) from 40 subjects. Images were gamma corrected and were converted to gray scale. The following texture features were extracted: (i) Statistical Features, (ii) Spatial Gray Level Dependence Matrices (SGLDM), and (iii) Gray level difference statistics (GLDS). The PNN and SVM neural network classifiers were also investigated for classifying normal and abnormal ROIs. Results show that there is significant difference (using Wilcoxon Rank Sum Test at a=0.05) between the texture features of normal and abnormal ROIs for both the gamma corrected and uncorrected images. Abnormal ROIs had lower gray scale median and homogeneity values, and higher entropy and contrast values when compared to the normal ROIs. The highest percentage of correct classifications score was 77% and was achieved for the SVM models trained with the SF and GLDS features. Concluding, texture features provide useful information differentiating between normal and abnormal ROIs of the endometrium. © 2006 IEEE. 3005 3008 Conference code: 69200 Cited By :6

The objective of this study was to develop a standardized protocol for the capturing and analysis of endoscopy digital images for subsequent use in a Computer Aided Diagnosis (CAD) system in gynaecological cancer. Images were captured at optimum Illumination and focus at 720×576 pixels using 24 bits color in the following cases: (i) for a variety of testing targets from a color palette with known color distribution, (ii) different viewing angles and distances from calf endometrium, and (iii) images from the human endometrium. Images were then gamma corrected and their classification performance was compared against that of non-gamma corrected images. No significant difference in texture features was found between the close up and panoramic views, and between angles, either before or after gamma correction. There was significant difference in certain texture features between normal and abnormal endometrium, both before and after gamma correction. Our findings suggest that proper color correction can significantly impact CAD system performance, and we recommend its application prior to quantitative texture analysis in gynaecological endoscopy. © 2005 IEEE. 7 VOLS 3336 3339 Conference code: 69123 Cited By :6

The objective of this study is to investigate the usefulness of texture analysis in the endometrium during hysteroscopy in endoscopic imaging of the uterine cavity. Endoscopy images from the endometrium from three subjects, at optimum illumination and focus, were frozen and digitized at 720×576 pixels using 24bits color. Regions of Interest (ROI) of normal (N=61) and abnormal (N=69) regions were manually selected by the physician. ROI images were converted into gray scale and Statistical Features (SF) and Spatial Gray Level Dependence Matrix features (SGLDM) were computed. The non-parametric Wilcoxon rank sum test at a = 0.05 was carried out for comparing the differences between normal and abnormal tissue. There was significant difference between normal and abnormal endometrium for the SF features variance, energy and entropy and for the SGLDM feature of angular second moment. There was no significant difference for the SF features mean, median, and SGLDM features of contrast, correlation and homogeneity. 26 II 1483 1486 Sponsors: Institute of Electrical and Electronics Engineers, IEEE IEEE Engineering in Medicine and Biology Society Conference code: 64132 Cited By :5

The provision of effective emergency telemedicine and home monitoring solutions are the major fields of interest discussed in this study. Ambulances, Rural Health Centers (RHC) or other remote health location such as Ships navigating in wide seas are common examples of possible emergency sites, while critical care telemetry and telemedicine home follow-ups are important issues of telemonitoring. In order to support the above different growing application fields we created a combined real-time and store and forward facility that consists of a base unit and a telemedicine (mobile) unit. This integrated system: can be used when handling emergency cases in ambulances, RHC or ships by using a mobile telemedicine unit at the emergency site and a base unit at the hospital-expert's site, enhances intensive health care provision by giving a mobile base unit to the ICU doctor while the telemedicine unit remains at the ICU patient site and enables home telemonitoring, by installing the telemedicine unit at the patient's home while the base unit remains at the physician's office or hospital. The system allows the transmission of vital biosignals (3-12 lead ECG, SPO2, NIBP, IBP, Temp) and still images of the patient. The transmission is performed through GSM mobile telecommunication network, through satellite links (where GSM is not available) or through Plain Old Telephony Systems (POTS) where available. Using this device a specialist doctor can telematically "move" to the patient's site and instruct unspecialized personnel when handling an emergency or telemonitoring case. Due to the need of storing and archiving of all data interchanged during the telemedicine sessions, we have equipped the consultation site with a multimedia database able to store and manage the data collected by the system. The performance of the system has been technically tested over several telecommunication means in addition the system has been clinically validated in three different countries using a standardized medical protocol. © 2003 Kyriacou et al licensee BioMed Central Ltd. 2 Tradenames: Borland Delphi 5 Connectix CRITIKON DINAMAP PLUS Monitor Model 8200, Critikon CRITIKON DINAMAP PLUS Monitor Model 9700, Critikon Logitech Micronet Option FirstFone Pentium PROTOCOL-Welch Allyn Propaq 1xx Vital Signs Monitor, Welch Allyn PROTOCOL-Welch Allyn Propaq Encore 2xx Vital Signs Monitor, Welch Allyn Toshiba libretto 100 ct portable PC, Toshiba TT-3064A CAPSAT Inmarsat Maritime Phone, Thrane and Thrane US-Robotics Windows 2000, Microsoft Windows 95, Microsoft Windows 98, Microsoft Windows NT, Microsoft Windows, Microsoft Manufacturers: Critikon Microsoft Nokia Thrane and Thrane Toshiba Welch Allyn Cited By :104

In this study we present a multipurpose health care telemedicine system, which can be used for emergency or patient monitoring cases. Ambulances, Rural Health Centers (RHC) or other remote health location, Ships navigating in wide seas and Airplanes in flight are common examples of possible emergency sites, while critical care telemetry and telemedicine home follow-ups are important issues of patient monitoring. The telemedicine system is a combined real-time and store and forward facility that consists of a base unit and a telemedicine (mobile) unit. The telemedicine unit (patient site) allows the transmission of vital biosignals (3-12 lead ECG, SPO2, NIBP, IBP, Temp) and still images of the patient from the incident place to the base unit (consultation site). The transmission is performed through GSM, Satellite links or POTS. Using this device a specialist doctor can telematically "move" to the patient site and instruct medical personnel when handling a patient. The consultation site is equipped with a multimedia database able to store and manage the data collected by the system. The system was validated in four different countries using a standardized medical protocol. 4 3544 3547 Sponsors: IEEE Conference code: 60340 Cited By :16

Includes bibliographical references (p. 116-125). Number of sources in the bibliography: 131. Thesis (Ph. D.) -- University of Cyprus, Faculty of Pure and Applied Sciences, Department of Computer Science, 2020. The University of Cyprus Library holds the printed form of the thesis. Με την εισαγωγή της επιστήμης της πληροφορικής στον τομέα της υγείας, πολλές οντότητες που συσχετίζονται με την υγεία, έχουν στην κατοχή τους τεράστιο όγκο από δεδομένα ασθενών. Παρά το γεγονός ότι η κοινοποίηση αυτών των δεδομένων σε ερευνητές μπορεί να αυξήσει την πιθανότητα ανακάλυψης καινοτόμων ευρημάτων, αυτό δεν είναι δυνατό λόγο νομικών και ηθικών ζητημάτων. Η μηχανική μάθηση μπορεί να χρησιμοποιηθεί σε αυτά τα δεδομένα για τον εντοπισμό των παραγόντων που αυξάνουν ή μειώνουν το ρίσκο κάποιος να αποκτήσει μια ασθένεια, αλλά οποιοσδήποτε αλγόριθμος που θα αναλύσει αυτά τα δεδομένα θα πρέπει να λαμβάνει υπόψη ότι οι περισσότερες κοινές ασθένειες επηρεάζονται από πολλαπλές γονιδιακές αλληλεπιδράσεις και αλληλεπιδράσεις με το περιβάλλον. Επομένως, θα πρέπει να χρησιμοποιηθούν αλγόριθμοι που επιτρέπουν την εύρεση τέτοιων αλληλεπιδράσεων. Στη διατριβή, αρχικά παρουσιάζεται ένας νέος αλγόριθμος για την ανωνυμοποίηση δεδομένων τα οποία έχουν διακριτές τιμές, μέσω κ-ανωνυμίας και αλγόριθμου επιλογής μεταβλητών, για προβλήματα ταξινόμησης. Ο αλγόριθμος αξιολογήθηκε σε διάφορα είδη ιατρικών δεδομένων και στην πλειονότητα των πειραμάτων τα ανώνυμα δεδομένα που παράχθηκαν, είχαν παρόμοια ή μεγαλύτερη ακρίβεια στην ταξινόμηση σε σχέση με τη χρήση των αρχικών μη-ανώνυμων δεδομένων. Στη συνέχεια παρουσιάζεται ένας καινούριος αλγόριθμος για τη μετατροπή συνεχών μεταβλητών σε διακριτές, με βάση την πυκνότητα των τιμών των μεταβλητών. Η μέθοδος αυτή έχει παρόμοιες επιδόσεις με τους αλγόριθμους που είναι ευρέως χρησιμοποιημένοι στον τομέα, και έχει το πλεονέκτημα ότι είναι υπολογιστικά αποδοτικός και μπορεί να χρησιμοποιηθεί σε μεγάλα δεδομένα. Για την αναγνώριση προτύπων σημειακών νουκλεοτιδικών πολυμορφισμών που συσχετίζονται με εμφάνιση ή όχι μιας ασθένειας, παρουσιάζεται ένας Χάρτης Αυτο-οργάνωσης για διακριτά δεδομένα. Η μέθοδος αυτή εφαρμόστηκε σε γενετικά δεδομένα και η κατηγοριοποίηση των δεδομένων που δημιούργησε ήταν στατιστικά σημαντική και αποκάλυψε ενδιαφέροντα πρότυπα που ήταν διαφορετικά μεταξύ των κατηγοριών που αντιπροσώπευαν ασθενείς και υγιή άτομα. Επίσης, προτείνεται ένα πλαίσιο για την αποτελεσματική ανακάλυψη ν-αλληλεπιδράσεων. Το πλαίσιο αυτό χρησιμοποιεί αλγόριθμους μηχανικής μάθησης για την μείωση του αριθμού των μεταβλητών ενός προβλήματος και για τη μείωση της διάστασης των μεταβλητών μετατρέποντάς τις σε δυαδική μορφή. Αυτό επιτρέπει την μείωση του προβλήματος των πολλαπλών συγκρίσεων και επίσης μειώνει τους βαθμούς ελευθερίας των στατιστικών υποθέσεων, το οποίο αυξάνει την στατιστική δύναμη για την αναγνώριση των ν-αλληλεπιδράσεων που συσχετίζονται με εμφάνιση ή όχι μιας ασθένειας. Τα αποτελέσματα δείχνουν ότι με τη νέα κωδικοποίηση, το προτεινόμενο πλαίσιο ήταν σε θέση να αναγνωρίσει περισσότερες στατιστικά σημαντικές ν-αλληλεπιδράσεις σε σύγκριση με τη χρήση της αρχικής κωδικοποίησης των μεταβλητών. In recent years, with the infiltration of information technology in healthcare, many healthcare related entities, have vast amounts of patients' data. Although sharing such data can increase the likelihood of identifying novel findings or even replicating existing research results, this is not happening due to legal and ethical issues. Machine learning can be used on such datasets to identify risk factors that can be used to improve our lives, but any algorithms that will analyze such data should take into consideration that most common diseases are influenced by multiple gene interactions and interactions with the environment. Hence they should use models that allow the finding of such multivariate associations. This thesis initially presents a novel algorithm for anonymizing categorical data with k-anonymity and performing feature selection for classification tasks. The algorithm was evaluated on various medical datasets and in the majority of the evaluated test cases the produced anonymized data had similar or better accuracies than using the full datasets. Additionally, a novel density based discretization algorithm is presented that has similar performance with state of the art algorithms while being computationally efficient and suitable for big data. For pattern recognition of n-SNP associations in case/control data, a Self Organizing Map for nominal categorical data is presented, which was able to produce statistically significant clustering revealing some interesting patterns between the clusters of cases and controls. Finally, a framework for efficient n-Way interaction testing is presented that uses machine learning to reduce the dimensionality of the data and to produce a targeted binary encoding of the features. This enables the reduction of the multiple testing problem and the degrees of freedom of the statistical tests applied for interaction testing, and hence increases the statistical power of the performed analysis. Results indicate that with the new encoding, the proposed framework was able to identify more statistically significant interactions compared to using the initial encoding of the features.

Includes bibliographical references (p. 157-176). Number of sources in the bibliography:185 Thesis (Ph. D.) -- University of Cyprus, Faculty of Pure and Applied Sciences, Department of Computer Science, 2017. The University of Cyprus Library holds the printed form of the thesis. Ο κύριος άξονας της έρευνας που διεξήχθη στα πλαίσια αυτής της διδακτορικής διατριβής είναι η εξερεύνηση διαφόρων λύσεων σε θέματα που αφορούν τον σχεδιασμό, την κατασκευή και την λειτουργικότητα των Ασαφών Γνωστικών Χαρτών (ΑΓΧ). Η συνεισφορά αυτού του έργου μπορεί να αναλυθεί σε τρία επίπεδα. Το πρώτο επίπεδο αφορά στην πρόταση μιας μεθοδολογίας κατασκευής ΑΓΧ για προβλήματα τα οποία δεν περιγράφονται από σύνολα δεδομένων και η μόνη πηγή πληροφόρησης για την κατασκευή τους προέρχεται από μια ομάδα ειδικών στον τομέα του προβλήματος. Τέτοια προβλήματα βρίσκονται στον κοινωνικό, πολιτικό κι οικονομικό τομέα. Στα πλαίσια αυτής της δουλειάς, προτάθηκε επίσης μια συνάρτηση ενεργοποίησης με σκοπό την καλύτερη αναπαράσταση του τρόπου που διαχέεται η αιτιώδης πληροφορία δυναμικά στο σύστημα. Το δεύτερο επίπεδο αφορά στην κατασκευή ΑΓΧ όταν υπάρχουν σύνολα δεδομένων που περιγράφουν το σύστημα όσο αφορά τις παραμέτρους του και τις διασυνδέσεις μεταξύ τους. Σε αυτόν τον τομέα υπάρχει δύο κατηγορίες. Η πρώτη κατηγορία ασχολείται με προβλήματα για τα οποία υπάρχουν σύνολα δεδομένων σε αριθμητική μορφή τα οποία όμως δεν μπορούν να αξιοποιηθούν για την κατασκευή ασαφών συστημάτων. Ως εκ τούτου, μια μεθοδολογία μεταμόρφωσης των αριθμητικών συνόλων δεδομένων σε ασαφή προτάθηκε και χρησιμοποιήθηκε σε συνδυασμό με τις Εξελικτικές Στρατηγικές για την δημιουργία ενός ΑΓΧ. Η συμμετοχή των ειδικών στην όλη διαδικασία είναι απαραίτητη καθώς δίνουν την δική τους οπτική για το πώς ερμηνεύουν διάφορες αριθμητικές τιμές των παραμέτρων. Η δεύτερη κατηγορία επικεντρώνεται στις περιπτώσεις για τις οποίες υπάρχει μεν διαθέσιμο σύνολο δεδομένων προς αξιοποίηση, δεν καθίσταται παρόλα αυτά δυνατή η εύρεση ή η επικοινωνία με ειδικούς που να γνωρίζουν το πρόβλημα. Παράλληλα, είναι διαθέσιμη μια πλούσια βιβλιογραφία μέσα στην οποία μπορεί κανείς να βρει πληροφορίες για διάφορα θέματα που αφορούν στην δομή και στην αναπαράσταση του προβλήματος. Ορμώμενοι από αυτή την παρατήρηση, προτείνουμε μια μεθοδολογία αναγνώρισης σημαντικών παραγόντων μέσα από την μελέτη σχετικής βιβλιογραφίας με το πρόβλημα προς μοντελοποίηση. Έπειτα, οι παράγοντες ορίζονται και αρχικοποιούνται βάσει πιθανοτικών μοντέλων που εξάγονται από την ανάλυση του διαθέσιμου συνόλου δεδομένων. Το τρίτο επίπεδο αυτής της δουλειάς εισάγει μια ιεραρχική αρχιτεκτονική των ΑΓΧ η οποία εκφράζεται μέσα από την χρήση δυναμικών βαρών. Διαφορετικοί συνδυασμοί αρχικών καταστάσεων κόμβων που θεωρούνται «πιο σημαντικής αξίας» οδηγούν σε διαφορετικές τιμές βαρών που έχει ως αποτέλεσμα την ενδυνάμωση ή την αποδυνάμωση μιας σχέσης. Αυτό το σχήμα προτάθηκε ως ένας μηχανισμός ελέγχου διαφόρων δυνατών συνεργιών που μπορεί να εμφανιστούν στο σύστημα κάτω από διαφορετικές συνθήκες. Οι προαναφερθείσες μεθοδολογίες και μηχανισμοί υλοποιήθηκαν και δοκιμάστηκαν σε πραγματικά προβλήματα. Το πρώτο πρόβλημα προέρχεται από την κυπριακή πραγματικότητα των τελευταίων ετών και σχετίζεται με την οικονομική κρίση στην Κύπρο κι Ελλάδα. Το δεύτερο πρόβλημα προέρχεται από τον ιατρικό χώρο και πιο συγκεκριμένα τον διαγνωστικό και αφορά στην διάγνωση της Τρισωμίας 21 που είναι πιο γνωστή ως σύνδρομο Down. Το πρώτο μέρος είναι εισαγωγικό στο θέμα της διατριβής. Ακολουθεί μια βιβλιογραφική ανασκόπηση. Το τρίτο μέρος αποτελείται από τρία κεφάλαια και περιγράφει τα προτεινόμενα σχήματα των ΑΓΧ. Παρουσιάζονται τα αποτελέσματα που επιτεύχθηκαν από την υλοποίηση τους. Στο τέλος, παρατίθενται τα συμπεράσματα της εργασίας αυτής και κάποιες μελλοντικές κατευθύνσεις εργασίας. The main goal of this research is to explore new methodologies regarding the design, building and functionality of Fuzzy Cognitive Maps (FCM). The contributions derived from this thesis can be analyzed into three main dimensions. The first is about presenting a specific methodology of building FCM models for problems for which datasets are absent and the only source of information for building a FCM model is a group of experts in the problem’s domain. Such systems lie in the social, political and economic fields. In the context of this work, a new activation function is also proposed as a try to represent in a more realistic way the way causal dynamics are expressed in such problems. The second dimension of this work is about constructing FCMs when datasets do exist and useful information about the problem’s factors and their interrelations can be extracted. The work done in this field can be further analyzed into two categories. The first deals with the cases a dataset describing several problems’ factors exist in crisp form making it useless for developing a fuzzy system. To solve this problem, a methodology for transforming the crisp dataset into fuzzy is proposed and then actually used in combination with Evolutionary Strategies to build a FCM. Experts participate in this process giving their insight on how they interpret the crisp values of each parameter. The second category is more concerned about the cases were datasets do exist but experts in the problem’s domain are absent or there is a lack of communication with them making it difficult to exploit their knowledge and experience to build a FCM. At the same time, a rich bibliography describing different problem’s aspects is published and available. Hence, a methodology of identifying the concepts through published bibliography is proposed. Then, the concepts are defined and initialized by probabilistic models which are derived by proper analysis of the existing dataset. The third and last dimension of this work introduces a hierarchical architecture of FCMs which is essentially expressed in the use of dynamic weights of the network’s relations. Different combinations of the “higher in significance” concept states can lead to different weight values by strengthening of weakening the relation. This schema was proposed as a mechanism of handling potential synergies that might appear in a system under different conditions. All of the proposed FCM methodologies and mechanisms were implemented and tested on real problems. The first chosen problem was drawn from Cypriot reality of the last years related to economic crisis and the second problem was a medical diagnostic one related to the diagnosis of Trisomy 21. The structure of the thesis is divided into four main parts. The first is an introductory to the thesis’ subject. The second is a literature review in aspects related to the thesis main subjects. The third part presents in detail the proposed schemas about FCMs and the results derived from their implementation followed by the fourth part which includes the conclusions and future directions of this work.