Biomedical subjects

J P de Sá

Publications and source records attributed to J P de Sá.

6 recordsLinked to original sources

PubMed2001

[Computerized electrocardiography system of Porto. I -- General description and experience with its use].

After a brief review of computer assisted ECG interpretation techniques, a microcomputer-based system for off-line ECG (Frank) analysis and interpretation developed at the University of Porto is presented. The program is menu-structured and includes report correction and editing facilities. Clinical data can be stored along with the ECG in the individual patient file. More than 500 characteristics are extracted from the ECG signal by the measurement section of the program. The diagnostic section uses Boolean-tree logic. The diagnostic threshold values and interpretation statements are kept in an independent file and can be changed at any moment by the user. Taking the cardiologist using the same logic and criteria as reference standard, an evaluation of the diagnostic efficiency of the system was performed in 509 reports from pediatric and adult patients: 7% of the reports needed corrections and 113 (3.7%) out of 3070 diagnostic statements produced had to be changed. Output options include the clinical data, more or less extensive listings of the wave measurements, X, Y and Z and polar plots and the 12-lead (derived) ECG, besides the interpretation report itself.

Diagnosis, Computer-Assisted↗

PubMed2000

Classification of breast tissue by electrical impedance spectroscopy.

Electrical impedance spectroscopy is a minimally invasive technique that has clear advantages for living tissue characterisation owing to its low cost and ease of use. The present paper describes how this technique can be applied to breast tissue classification and breast cancer detection. Statistical analysis is used to derive a set of rules based on features extracted from the graphical representation of electrical impedance spectra. These rules are used hierarchically to discriminate several classes of breast tissue. Results of statistical classification obtained from a data set of 106 cases representing six classes of excised breast tissue show an overall classification efficiency of approximately 92% with carcinoma discrimination > 86%.

Breast Neoplasms↗

PubMed1998

Classification of foetal heart rate sequences based on fractal features.

Visual inspection of foetal heart rate (FHR) sequences is an important means of foetal well-being evaluation. The application of fractal features for classifying physiologically relevant FHR sequence patterns is reported. The use of fractal features is motivated by the difficulties exhibited by traditional classification schemes to discriminate some classes of FHR sequence and by the recognition that this type of signal exhibits features on different scales of observation, just as fractal signals do. To characterise the signals by fractal features, two approaches are taken. The first modes the FHR sequences as temporal fractals. The second uses techniques from the chaos-theory field and aims to model the attractor based on FHR sequences. The fractal features determined by both approaches are used to design a Bayesian classification scheme. Classification results for three classes are presented; they are quite satisfactory and illustrate the importance of this type of methodology.

Bayes Theorem↗

PubMed1998

Automatic classifiers for the interpretation of electrocardiograms.

The morphological diagnosis of ECGs is a pattern recognition procedure. The way the clinician does this is not clearly elucidated. Nevertheless, several models aimed at achieving identical results by automatic means are empleyed. While in the doctor's case this is not exactly so, the computer task for ECG interpretation comprises two distinct and sequential phases: feature extraction and classification. A set of signal measurements containing information for the characterization of the waveform is first obtained. These waveform descriptors are then used to allocate the ECG to one or more diagnostic classes in the classification phase. The classifier can embody rules-of-thumb used by the clinician to decide between conflicting ECG diagnosis and formal or fuzzy logic as a reasoning tool (heuristic classifiers). On the other hand, it can use complex and even abstract signal features as waveform descriptors and different discriminant function models for class allocation (statistical classifiers). More recently, artificial neural network techniques have also been used for signal classification. The authors review feature selection techniques and classification strategies, problems and methods of performance evaluation and results obtained by different classification approaches. A brief discussion of the relative merits of the two main types of ECG classifiers, logical and statistical, is included.

Diagnosis, Computer-Assisted↗

PubMed1992

[Automated analysis of cardiotocographic signals].

Cardiotocographic signals, used as a perinatal diagnostic tool, comprise fetal heart beat signals (FHR), uterine contractions (UC) and fetal movements (FM). Visual inspection of these three signals is limited, subjective, time consuming and with low reproducibility. The present paper illustrates an application of signal processing techniques to the automatic analysis of cardiotocographic signals, allowing to overcome visual inspection limitations. Porto system of cardiotocograms automatic analysis acquires the signals from a conventional cardiotocograph. The signals are stored and processed by a personal computer. System software allows the user to perform several operations such as signal acquisition, signal storage and retrieval from files, signal analysis and display. Signal analysis is preceded by various signal conditioning operations (filtering, spike removal, etc.) and consists in the estimation of several parameters with diagnostic value: FHR baseline; FHR accelerations and decelerations; uterine contractions; long and short term FHR variability. The system prototype is working on a routine basis at the Obstetrics Department of S. João Hospital (main Oporto Hospital) and is being evaluated using a large data set. A preliminary evaluation performed by 3 experts on a 50 cases set, yielded an agreement rate with computer measurements of 98% for the baseline, 72% to 82% for accelerations and decelerations and 76% for the uterine contractions.

Cardiotocography↗

PubMed1990

Interpretation of short ECGs with a personal computer: the Porto program.

A computer program for ECG analysis and interpretation developed at the University of Porto, Portugal is presented. The program runs on a microcomputer and employs the three-lead Frank VCG. The signals are sampled at 250 Hz with 8-bit precision during 5.5 s. Details on signal conditioning and wave recognition and measurement techniques are given. The diagnostic part of the program uses decision-tree logic. The decision rules are mainly derived from the Washington Code. The diagnostic accuracy of four classes (normal, left and right ventricular hypertrophies and myocardial infarction), evaluated in a sample of 1,075 pediatric and adult patients and classified by ECG-independent means was 76%. The program is currently being evaluated on the CSE database.

Adolescent↗