A stochastic modelling framework for the reconstruction of cardiovascular signals.

This paper presents a common stochastic modelling framework for physiological signals which allows patient simulation following a synthesis-by-analysis approach. Within this framework, we propose a general model-based methodology able to reconstruct missing or artifacted signal intervals in cardiovascular monitoring applications. The proposed model consists of independent stages which provide high flexibility to incorporate signals of different nature in terms of shape, cross-correlation and variability. The reconstruction methodology is based on model sampling and selection based on a wide range of boundary conditions, which include prior information. Results on real data show how the proposed methodology fits the particular approaches presented so far for electrocardiogram (ECG) reconstruction and how a simple extension within the framework can significantly improve their performance.

Stochastic modeling of the PPG signal: a synthesis-by-analysis approach with applications.

In this paper, we propose a stochastic model of photoplethysmographic signals that is able to synthesize an arbitrary number of other statistically equivalent signals to the one under analysis. To that end, we first preprocess the pulse signal to normalize and time-align pulses. In a second stage, we design a single-pulse model, which consists of ten parameters. In the third stage, the time evolution of this ten-parameter vector is approximated by means of two autoregressive moving average models, one for the trend and one for the residue; this model is applied after a decorrelation step which let us to process each vector component in parallel. The experiments carried out show that the model we here propose is able to maintain the main features of the original signal; this is accomplished by means of both a linear spectral analysis and also by comparing two measures obtained from a nonlinear analysis. Finally, we explore the capability of the model to: 1) track physical activity; 2) obtain statistics of clinical parameters by model sampling; and 3) recover corrupted or missing signal epochs by synthesis.

Automatic diagnosis of ADHD based on multichannel nonlinear analysis of actimetry registries.

Attention-Deficit Hyperactivity Disorder (ADHD) is the most common mental health problem in childhood and adolescence. It is commonly diagnosed by means of subjective methods which tend to overestimate the severity of the pathology. A number of objective methods also exist, but they are either expensive or time-consuming. Some recent proposals based on nonlinear processing of activity registries have deserved special attention. Since they rely on actigraphy measurements, they are both inexpensive and non-invasive. Among these methods, those shown to have higher reliability are based on single-channel complexity assessment of the activity patterns. This way, potentially useful information related to the interaction between the different channels is discarded. In this paper we propose a new methodology for ADHD diagnosis based on joint complexity assessment of multichannel activity registries. Results on real data show that the proposed method constitute a useful diagnostic aid tool reaching 87:10% sensitivity and 84.38% specificity. The combination of ADHD indicators extracted with the proposed method with single-channel complexity-based indices previously proposed lead to sensitivity and specifity values above 90%.

Characterization of activity epochs in actimetric registries for infantile colic diagnosis: Identification and feature extraction based on wavelets and symbolic dynamics.

The diagnosis and therapy planning of high prevalence pathologies such as infantile colic can be substantially improved by statistical signal processing of activity/rest registries. Assuming that colic episodes are associated to activity episodes, diagnosis aid systems should be based on preprocessing techniques able to separate real activity from rest epochs, and feature extraction methods to identify meaningful indices with diagnostic capabilities. In this paper, we propose a two step diagnosis aid methodology for infantile colic in children below 3 months old. Identification of activity periods is performed by means of a wavelet based activity filter which does not depend on the acquisition device (as so far proposed methods do). In addition, symbolic dynamic analysis is used for extraction of discriminative indices from the activity time series. Results on real data yielded 100% sensitivity and 80% specificity in a study group composed of 46 cases and 10 control subjects.