Anal yzing electrocardiographic signals (ECG) includes not only inspection of P, QRS and T waves, but also the causal relations they have and the temporal sequences they build within long observation periods. In fact, the spatio-temporal patterns must be described through shape features as well as complete time occurrence distributions in order to match the observed data with the underlying processes. It has been demonstrated recently that in depth knowledge can be qualitatively modeled by means of anatomofunctional decompositions [1] and can be used to simulate complex phenomena (i.e., reentry, block, etc.) and to provide a real understanding of their behavior at different abstraction levels. However, the on-line examination of ECG recordings can only be carried out if the relations between data events and the model are available. In other words, signal processing techniques must feed the model with accurately estimated features to discard the non relevant interpretations. This task is difficult to achieve because of the composite nature of the ECG (i.e., a combination of signal and noise) and its non stationary behavior. These characteristics motivated the approach we de-scribe, which makes use of wavelet transforms (WT) or time scale representations. These new tools have already been applied in ECG analysis for enhancing late potentials [2-3], reducing noise [4], and QRS detection [5]. In this article, we limit our study to recognizing normal and abnormal beats, and assume that a prior segmentation has been performed. It differs from previous work by addressing four main questions: 1) what is the most appropriate WT to use? 2) what are the most relevant features for efficient encoding of cardiac patterns? 3) what decomposition levels must be retained? 4) does WT improve the recognition process? The first issue is critical in all application areas. There is no theoretical answer at the moment, and the only technique at our disposal is to compare the results provided by several wavelet families. Questions 2 and 3 have been examined by considering two stages: (a) a characterization phase based on a principal component analysis (PCA) [6], which allows us to jointly represent and interpret the objects (i.e., the cardiac complexes) and the descriptive variables; and (b) a discrimination step by means of a linear discriminant analysis (LDA) [7]. This last stage leads us to identify the variables capable of separating the patterns, the objective being to derive the most discriminant decomposition levels. A supervised procedure is first applied on a learning set. The resulting performance is further tested on an additional set of patterns. The fourth issue has been considered by comparing the best solution provided by the WT with classical signal descriptions.

Telemedicine is producing a great impact in the monitoring of patients located in remote nonclinical environments such as homes, elder communities, gymnasiums, schools, remote military bases, ships, and the like. A number of applications, ranging from data collection, to chronic patient surveillance, and even to the control of therapeutic procedures, are being implemented in many parts of the world. As part of this growing trend, this paper discusses the problems in electrocardiogram (ECG) real-time data acquisition, transmission, and visualization over the Internet. ECG signals are transmitted in real time from a patient in a remote nonclinical environment to the specialist in a hospital or clinic using the current capabilities and availability of the Internet. A prototype system is composed of a portable data acquisition and preprocessing module connected to the computer in the remote site via its RS-232 port, a Java-based client-server platform, and software modules to handle communication protocols between data acquisition module and the patient's personal computer, and to handle client-server communication. The purpose of the system is the provision of extended monitoring for patients under drug therapy after infarction, data collection in some particular cases, remote consultation, and low-cost ECG monitoring for the elderly.

No abstract