A. Taddei

A system for the detection of ischemic episodes in ambulatory ECG

Describes a system for the automated detection of ischemic changes in double channel ambulatory EGG. The high performance QRS detector, previously designed and tested by the authors, was applied. A subset of ECG records of the European ST-T Database was used for development, whereas the full set of 90 records and selected 24-hour ECG recordings were used for evaluation. Noisy ECG segments were rejected by the QRS detector or processed by the use of median filters. ST segment deviations, measured beat by beat on both channels, were used for detection of ischemic episodes. The time series of parameters were smoothed by nonlinear filters. Episode detection is based on the analysis of the 2-D path of the two ST deviations. Results of performance evaluation on the European ST-T Database are reported.

Development of an electronic medical record for patient care in cardiology

An electronic medical record system has been designed and developed with the aim of supporting patient care in the department of cardiology. A clinical information system was implemented to integrate the different heterogeneous sources of patient information. All the clinically relevant patient data were gathered in a timely way and stored in the clinical database. A viewer/editor of patient medical records was designed, addressing friendliness, flexibility and communication. The basic function for consultation is the time-oriented (weekly or daily) representation of patient parameters, care events and examinations. World Wide Web technology has been applied to implement the medical record. The system is currently under clinical evaluation.

Fuzzy modelling of the expert's knowledge in ECG-based ischaemia detection

Abstract We concentrate on computational implementations of the detection of ischaemic episodes in order to allow continuous monitoring of patients interned in Coronary Care Units (CCU). The consideration of several alternatives led us to choosing an implementation based on the experts knowledge as the most adequate. In order to solve the problem of transforming the imprecise knowledge provided by the expert into computable structures we have decided to make use of a fuzzy implementation. We illustrate and comment the operation of our solution over some records of the European ST-T Database, which is an electrocardiographic database specifically aimed at the validation of algorithms whose objective is the detection of ischaemic episodes. This work can be included in the more general problem of the design, physical implementation and validation of an electronic-computational system (SUTIL), which by means of the simultaneous, integrated and intelligent evaluation of a large amount of information extracted from real time processing of haemodynamic and electrocardiographic signals, performs an intensive and exhaustive follow up of patients with ischaemic cardiopathies interned in CCUs.

The prognostic impact of objective nutritional indices in elderly patients with ST-elevation myocardial infarction undergoing primary coronary intervention.

BACKGROUND The prognostic impact of nutritional status in ST-elevation myocardial infarction (STEMI) patients is poorly understood. METHODS We used the controlling nutritional status (CONUT) score and the prognostic nutritional index (PNI) score on outcomes of 945 patients with acute STEMI undergoing primary percutaneous coronary intervention with stent. RESULTS During a median follow-up of 2years (1-3.3years, interquartile range), 56 patients (5.9%) died for all-cause of death. In the dead group, the CONUT and PNI scores were more severe than in the alive group. Elderly patients (≥71years) had nutritional indices more serious than patients <71years. In the whole population of the study, both CONUT and PNI correlated with clinical markers of poor prognosis such as brain natriuretic peptide (BNP), creatinine and liver enzymes. Kaplan-Meier curves revealed that the patients with severe CONUT but not patients with severe PNI index had the highest event rate for all-cause death, with a log-rank of p<0.001. The Cox proportional hazard analyses showed that, contrary to PNI score, the CONUT score was associated with increased risk of all-cause death for both unadjusted model and age- and sex-adjusted model, while in a full-adjusted model the best predictors were age and BNP. CONCLUSIONS In STEMI patients, the nutritional status evaluated by the CONUT score, in addition to other comorbidities, can affect the prognosis in elderly patients. These results suggest a personalized nutritional treatment as well as an accurate assessment of the appropriateness of lipid-lowering treatment after coronary revascularization.

A multichannel template based data compression algorithm

Data compression methods for electrocardiogram (EGG) signals play an important role in computer processing and analysis of ECG. Ambulatory monitoring of ECG (AECG) is the area of application which mostly benefits from data compression techniques. AECG is usually done using the conventional Holter monitor, which consists of a 24 hour, 2-channel, analog, cassette recording of the EGG. Modern Holter monitors with digital memory cards are now expected to improve fidelity of recording and decrease system size. The cost of high capacity integrated memory cards makes it convenient to store ECG samples in compact form. A powerful template matching algorithm (ABS) has been recently introduced. The authors present a method which improves ABS algorithm, in order to achieve a better compression ratio as well as faster processing.<<ETX>>

Use of an electronic medical record in a department of cardiology

A computer-network infrastructure was realized to integrate the different remote cardiovascular diagnostic laboratories with the data derived from CCU and the administrative information. A variety of heterogeneous data, texts, signals, images is gathered from each peripheral unit, stored into a relational database (ARCA), processed and presented to health-care personnel by network-connected clinical workstations. From August 1999 to August 2000 the Electronic Medical Records (EMR) of 754 patients hospitalized in a Cardiological department were collected; 7808 procedures were digitally integrated and for each patient was possible to calculate the number of tests performed during hospitalization and the quantity of drugs assumed. The use of EMR allowed to obtain rapidly a clinical data integration, to access to patient data from any clinical lab; to collect information to obtain patient cost definition.

QRS morphological classification using artificial neural networks

Artificial neural networks (ANNs) were applied to electrocardiographic (ECG) signals to classify QRS complexes. Several ANN paradigms were considered, and two were selected for the ECG analysis: backpropagation (BP) and the Kohonen feature map (KFM). ANNs were trained on 8 groups of 20 QRS complexes each, extracted from the VALE database (DB); each group was related to a QRS morphology as obtained by the DB annotations. The ANN performances were evaluated using both the learning set and the whole case as a recall set. The BP network showed a good specificity and was found able to separate morphologies with ambiguous DB annotations. The KFM network was able to create a clustering of QRS morphologies with a high agreement with the original annotations.<<ETX>>

An informative system for structured data management to build a cardiological multidimensional database

To collect clinical data generated during patient care in a multidisciplinary environment an informative system should manage each clinical aspect of involved disciplines. In this project we developed an application of what we call a Dynamic Dataset Definition (D/sup 3/) that allows rapid system deployment due to its capability of building a graphical interface directly related to a definition of collectable data. It is sufficient to initialize the system with a minimal data-set, not yet completely established, and let to successive steps to refine detailed information with current clinical use, where annotations are followed by an immediate program changes. Moreover, in a multidisciplinary clinical environment, information is shared providing a partial view on the common defined dataset, in order to inherit data definitions of shared parameters and let them to be displayed and/or entered by several physicians. At last, the physician will retrieve information based on his personal choice or by a common scheme or calculated indexes, with a final generation of a WEB based report. This system has been implemented in a network environment to guarantee an effortless data sharing between physicians, and tested in a Cardiology and Cardiac Surgery departments at our institute. Presently, 12 clinical subsystems are based on this model and more than 270,000 parameters have been collected on 1200 patients since August 2001 to August 2002.

Laboratory data integration into medical record

Laboratory Information System, integrated with the Hospital Information System, has been developed at the G.Pasquinucci Hospital, section of Institute of Clinical Physiology of National Research Council (CNR), specialized in adult and paediatric cardiac surgery. The aim was to automate the testing process from clinical departments to laboratory and back into medical record. Laboratory workflow consists of three parts: (a) test ordering by clinical staff, printing bar-coded ID labels and transmitting orders by network to laboratory; (b) processing test requests and controlling identified specimens by laboratory staff, providing work orders to analytical instruments and validation of results authorizing delivery into the hospital clinical repository; (c) consulting test results in clinical departments by referring physicians through the electronic medical record. This year the system has been used on adult patients processing 135000 laboratory tests concerning chemistry, haematology, coagulation and immunology

A workstation for data consultation and clinical decision-support in the department of cardiology

A project for the realization of a Local Area Network in the department of Cardiology (LAN-C) is in progress at our institute with the objective of the integration of heterogeneous data acquired from diagnostic equipment to help clinical decision making. A special workstation of LAN-C, presented in this paper, has been designed for supporting data consultation and decision making. It integrates a set of applications: data retrieval and presentation, diagnostic and treatment planning protocols, in-depth analysis clinical decision support. Initially, the protocol for diagnosis of coronary artery disease and a module for the analysis of follow-up data were developed. Relevant diagnostic information is retrieved from the LAN-C database according to a protocol logic, defined by the cardiological staff. Regional myocardial wall motion and perfusion are represented at rest and during stress by 2-D synthetic diagrams. The module for statistical analysis allows one to select patients, followed up after discharge, for the identification of prognostic variables by the use of Cox regression model (BMDP 2L).<<ETX>>