This paper deals with the reliability of algorithms predicting the location of the culprit lesion in a totally occluded coronary artery from a 12 lead ECG recording. Four commercial ECG analysis programs and two specific algorithms were evaluated. Five hundred well documented cases were used for testing. Evaluation of the results reveals that all algorithms perform suboptimal.

In the past years, organizations worldwide have put many efforts in establishing guidelines for the medical practice. However, it has become clear that in order to improve patient care it is necessary to check and ensure the adherence to the established guidelines. Therefore, in 2007, the Dutch Society of Cardiology (NVVC) has implemented a structure for national databases that will serve as benchmark tool for the quality assurance of the clinical outcome of cardiology procedures in all hospitals in the Netherlands. For this, the NVVC has instituted an independent foundation: the National Cardiovascular Data Registry (NCDR). The goal is implementation and management of various national databases. These databases will contain data on incidence and prevalence of various cardiovascular diseases, in combination with registration of numbers and results of procedures and interventions.

This paper describes the AMOC project which aims at the improvement of patient throughput in the outpatient clinics. Indicators describing the workflow were extracted and serve as a base for a so called continuous improvement model. As pro actively monitoring the workflow turned out to be the best tool to optimize the throughput, a computer system was developed to present in a ldquodashboardrdquo like manner the state of the indicators.

This paper describes the revival of a project for exchange of electrocardiograms (ECG’s) of the same patient between different centers. A national index-server was set up containing information on patients who’s ECG is available in the participating centers. Through the use of virtual private networks and standard browsers each ECG stored elsewhere can be retrieved.

In the Netherlands the Central Pacemaker Patient Registry (CPPR) collects information of pacemaker and ICD (implantable cardio defibrillator) patients from all 109 Dutch hospitals. Many pacemaker clinics use a computer to store their implant and follow-up data in a database. Because the devices are getting more and more complex more clinical data is needed for optimal use of the device. Since 1989 databases have been developed by several pacemaker industries and some clinics use databases developed themselves. When using these databases you depend on individual persons for support and update of the database. In order to improve the accuracy of received data and to ensure continuity a uniform pacemaker and ICD information system is developed where data is checked, after which it is sent to the central registry by e-mail and where support is guaranteed by the NPRF.

Patients with heart failure (HF) are admitted to the hospital in a non-elective way. Receiving intravenous medication, they are very limited in their activities of daily living (ADL), resulting in intensive nursing care. To predict the needed capacity, a distinction is made between currently-admitted patients and patients that will be admitted in the near future. Computer simulation is used to predict when new HF patients can be expected and when they will be discharged. The simulation model also provides information about the ADL status of the patients. Former HF patients are used to predict when current HF patients that have already been admitted into the cardiology department will be discharged. The results of tests of both methods are presented and discussed

To understand the etiology of multigenic diseases like atherosclerosis, a polymerase chain reaction (PCR) based gene array containing 65 single nucleotide polymorphisms (SNPs) was analyzed. To asses the possibilities of pattern recognition techniques in detecting unfavorable genetic combinations, two approaches were analysed. A selection of these 65 SNPs formed the input both to binary logistic regression models and to self-learning artificial neural networks (ANNs). Repeated analyses showed that both methods performed equally well. Further research to improve the differentiating power of both methods should focus first on decreasing the number of otherwise indeterminable polymorphisms.

Currently the ICT departments working together in the Interuniversity Institute of Cardiology of the Netherlands are conducting a research project to conceive and build an Electronic Patient Record for CARdiology (EPD-CAR) In this paper one part of this project, MMM the module designed to archive all (cardiac) medication prescribed to a patient is discussed. This module not only forms a electronic patient medication file but also can be used to create new prescriptions, edit old ones and to generate all kind of overviews, like given medication, stopped medication, etc. to support optimal medical treatment. Apart from the features of the actual medication module, the integration of the module in Hospital or Departmental Information Systems is illustrated. Finally new options for research enabled by the use of such a uniform module to document prescribed medication are discussed

In the Netherlands many pacemaker clinics are using a computer to store their implant and follow-up data in a database. Still there are quite a number of clinics that are only using their computer for word processing.In order to improve the accuracy of the received data we developed a data entry package where the data is checked after which it is sent to the central registry by E-mail.

Analysis of Heart Rate Variability is a non-invasive quantitative tool to study the influence of the autonomic nervous system on the heart. Rapid variations in heart rate, related to breathing are primarily mediated by the vagal limb of the autonomic nervous system. The resulting variations in heart rate are usually referred to as respiratory sinus arrhythmia. Metronome breathing (MB) is often advocated to assess more accurately vagal control. However, the additional value of MB over SB has never been established. The authors studied the effect of MB (0.25 Hz) on HRV variables in 12 healthy male subjects under stable conditions using pharmacological autonomic blockade. During MB several variables showed a lower absolute value, however a strong correlation existed between variables computed during spontaneous breathing (SB) and MB. MB offers some, but no important advantages over spontaneous breathing.

This paper compares various classifiers for the discrimination between ventricular and supraventricular tachycardias. These classifiers are built by means of a neural network building tool and an induction algorithm. The performance of the classifiers is compared with the performance of expert clinicians. Also the knowledge represented by the neural network and two classification trees is compared.

Recently an electrocardiographic sign has been described enabling the recognition of 3-vessel or left main stem disease. In this study, using two self-learning techniques, the neural network and the induction algorithm approach, this sign was validated and further refined. Based on 113 ECGs, (63 training and 50 for testing), the influence of the number of parameters and the effect of additional weight factors to direct the classification process, was evaluated.<>

A flexible coding method was developed, called CACOLA (cardiology coding language), to be used for the registration of medical information within the fields of the cardiology subspecialties. CACOLA uses the principle of positive multi-item variables. This means that one database variable contains extensive information concerning topics like electrocardiogram description, medication, or diagnosis, using the coding language. The most important advantage of using this method is the standardization of the variables. An example is shown of how information from the coronary care unit in combination with the last outpatient clinic medication is stored in a database using CACOLA as the coding language

The development of a software package for cardiac echo data management, called CAESARS, is described. Using an IBM-compatible personal computer, data comprising the results of 13000 echocardiographic investigations were stored with the help of a simple database program. For each echocardiographic study a number of standard measurements and basic variables were entered. A descriptive code was used to diminish the amount of memory required to store both the diagnosis and left ventricular wall motion-data. The introduction of a local area network, providing the data to other users, made the development of a translation program inevitable, to encrypt the coded diagnosis back into natural language. Besides this `translator' several tools were integrated in the package, making this large amount of data more easily accessible for research