다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Data Mining Practical Machine Learning Tools and Techniques

An Introduction to MultiAgent Systems.

A recent paper posed the question: "Graph Matching: What are we really talking about?". Far from providing a definite answer to that question, in this paper we will try to characterize the role that graphs play within the Pattern Recognition field. To this aim two taxonomies are presented and discussed. The first includes almost all the graph matching algorithms proposed from the late seventies, and describes the different classes of algorithms. The second taxonomy considers the types of common applications of graph-based techniques in the Pattern Recognition and Machine Vision field.

/pdf/thirty-years-of-graph-matching-in-pattern-recognition-58lvpdkcaj.pdf

Thirty years of graph matching in pattern recognition

This chapter presents an abstract view of the regional health information networks (RHN). A presentation of the architecture and structures of the RHN is included. It also reveals the need for integration of information in the framework of a RHN and key issues for the applicability of health information standards to achieve interoperable health care organizations which are the stakeholders of the RHN are presented. Furthermore, a list of case studies for the implementation of RHN in health systems in European, as well as non-European countries, such as the U.S., Canada, New Zealand, and Australia are demonstrated. Finally, important areas to focus when evaluating RHN are described. The authors hope that the abstract view of RHN would assist in the understanding of the key areas when building regional health networks. deFInItIon oF rhn Since 1990s, the regional innovation systems gain the focus of researchers and policy makers as the framework for financial development in regional level and generally for innovation policy making. The territorial dimension of regional systems remains ambiguous, along with the role of actors in those systems. The regional health network offers health care services in a region through efficient administrative organization of health care providers in region and by means of regional innovation systems like the information systems, eHealth tools and services. The definition of regional systems should start with the definition of the region. The literature is ambiguous on the nature and characteristics of the regional innovation systems (Doloreux & Parto, 2005). The term region has sometime the same meaning with the administrative regions or divisions of DOI: 10.4018/978-1-60566-356-2.ch008

Health Information Standards: Towards Integrated Health Information Networks

The main objective of this work is the study of EEG signals in order to investigate sleepiness induced from drug administration for insomnia and sleep deprivation. Data used in this work were obtained from real experiments in FORENAP, France and in CERTH, Thessaloniki, Greece. The features under consideration are Power Spectrum in certain frequency areas, alpha slow-wave index (ASI) and Fractal Dimension (FD) for placebo and verum subjects. Studying these features in the above groups, we found that sleepiness due to hypnotic medication and due to sleep deprivation can cause different behaviour in brain activity at certain locations. These EEG characteristics could be used for the classification of the medication intake (verum or placebo) and its effect.

/pdf/investigation-of-sleepiness-induced-by-insomnia-medication-2rb05pq8ch.pdf

Investigation of Sleepiness Induced by Insomnia Medication Treatment and Sleep Deprivation

WELCOME project aims at the development of a technical solution that will leverage integrated care of COPD and comorbidities via wearable technologies, user applications and cloud computing. This work focuses on the latter part and proposes technological approaches that will enable standardized cloud based data management and analytics in a challenging multimorbid health scenario. WELCOME cloud approach encompasses the data model and semantics binding to HL7-FHIR, the persistent data storage for performance and scalability, and the orchestration in a loose coupling with the feature extraction services. While DSS is currently being built on existing clinical knowledge combining the calculated features, the richness of new features and their solid semantics will enable the generation of new knowledge towards a learning health system. This can leverage coordinated care for the management of COPD and comorbidities.

Clinical flows and decision support systems for co-ordinated and integrated care in COPD

Atrial fibrillation (AF) is a complex phenomenon, related with a multitude of factors, including the electrical properties of the atrial substrate. The purpose of this work is to present a method that highlights electrocardiographic differences between normal subjects and patients with paroxysmal AF episodes (PAF), potentially related with substrate differences. Vectorcardiography recordings are considered and, for each lead (X-Y-Z), on a beat by beat basis, a steady window before QRS, corresponding to the atrial activity, is analysed via continuous wavelet transform. Wavelet-based parameters are calculated and compared between the normal and AF group, with the beat to beat variation of wavelet energy as the most important feature showing a significantly higher variability in the AF group.

Beat to beat wavelet variability in atrial fibrillation

The study of the huge diversity of immune receptors, often referred to as immune repertoire profiling, is a prerequisite for diagnosis, prognostication and monitoring of hematological disorders. In the era of high-throughput sequencing (HTS), the abundance of immunogenetic data has revealed unprecedented opportunities for the thorough profiling of T-cell receptors (TR) and B-cell receptors (BcR). However, the volume of the data to be analyzed mandates for efficient and ease-to-use immune repertoire profiling software applications. This work introduces Immune Repertoire Profiler (IRProfiler), a novel software pipeline that delivers a number of core receptor repertoire quantification and comparison functionalities on high-throughput TR and BcR sequencing data. Adopting 5 alternative clonotype definitions, IRProfiler implements a series of algorithms for 1) data filtering, 2) calculation of clonotype diversity and expression, 3) calculation of gene usage for the V and J subgroups, 4) detection of shared and exclusive clonotypes among multiple repertoires, and 5) comparison of gene usage for V and J subgroups among multiple repertoires. IRProfiler has been implemented as a toolbox of the Galaxy bioinformatics platform, comprising 6 tools. Theoretical and experimental evaluation has shown that the tools of IRProfiler are able to scale well with respect to the size of input dataset(s). IRProfiler has been utilized by a number of recently published studies concerning hematological disorders. IRProfiler is made freely available via 3 distribution channels, including the Galaxy Tool Shed. Despite being a new entry in a crowded ecosystem of immune repertoire profiling software, IRProfiler founds its added value on its support for alternative clonotype definitions in conjunction with a combination of properties stemming from its user-centric design, namely ease-of-use, ease-of-access, exploitability of the output data, and analysis flexibility.

/pdf/irprofiler-a-software-toolbox-for-high-throughput-immune-3cnn6z7uzu.pdf

Nicos Maglaveras

Papers

Health Information Standards: Towards Integrated Health Information Networks

Investigation of Sleepiness Induced by Insomnia Medication Treatment and Sleep Deprivation

Clinical flows and decision support systems for co-ordinated and integrated care in COPD

Beat to beat wavelet variability in atrial fibrillation

IRProfiler - a software toolbox for high throughput immune receptor profiling.