Sparse Decomposition of Heart Rate Using a Bernoulli-Gaussian Model: Application to Sleep Apnoea Detection.

In this paper, we propose a sparse decomposition of the heart rate during sleep with an application to apnoea-RERA detection. We observed that the tachycardia following an apnoea event has a quasi-deterministic shape with a random amplitude. Accordingly, we model the apnoea-perturbed heart rate as a Bernoulli-Gaussian (BG) process convolved with a deterministic reference signal that allows the identification of tachycardia and bradycardia events. The problem of determining the BG series indicating the presence or absence of an event and estimating its amplitude is a deconvolution problem for which sparsity is imposed. This allows an almost syntactic representation of the heart rate on which simple detection algorithms are applied.

Cross-Gram matrices and their use in transfer learning: Application to automatic REM detection using heart rate.

BACKGROUND AND OBJECTIVES: while traditional sleep staging is achieved through the visual - expert-based - annotation of a polysomnography, it has the disadvantages of being unpractical and expensive. Alternatives have been developed over the years to relieve sleep staging from its heavy requirements, through the collection of more easily assessable signals and its automation using machine learning. However, these alternatives have their limitations, some due to variabilities among and between subjects, other inherent to their use of sub-discriminative signals. Many new solutions rely on the evaluation of the Autonomic Nervous System (ANS) activation through the assessment of the heart-rate (HR); the latter is modulated by the aforementioned variabilities, which may result in data and concept shifts between what was learned and what we want to classify. Such adversary effects are usually tackled by Transfer Learning, dealing with problems where there are differences between what is known (source) and what we want to classify (target). In this paper, we propose two new kernel-based methods of transfer learning and assess their performances in Rapid-Eye-Movement (REM) sleep stage detection, using solely the heart rate. METHODS: our first contribution is the introduction of Kernel-Cross Alignment (KCA), a measure of similarity between a source and a target, which is a direct extension of Kernel-Target Alignment (KTA). To our knowledge, KCA has currently never been studied in the literature. Our second contribution is two alignment-based methods of transfer learning: Kernel-Target Alignment Transfer Learning (KTATL) and Kernel-Cross Alignment Transfer Learning (KCATL). Both methods differ from KTA, whose traditional use is kernel-tuning: in our methods, the kernel has been fixed beforehand, and our objective is the improvement of the estimation of unknown target labels by taking into account how observations relate to each other, which, as it will be explained, allows to transfer knowledge (transfer learning). RESULTS: we compare performances with transfer learning (KCATL, KTATL) to performances without transfer using a fixed classifier (a Support Vector Classifier - SVC). In most cases, both transfer learning methods result in an improvement of performances (higher detection rates for a fixed false-alarm rate). Our methods do not require iterative computations. CONCLUSION: we observe improved performances using our transfer methods, which are computationally efficient, as they only require the computation of a kernel matrix and are non-iterative. However, some optimisation aspects are still under investigation.

A multiclass multivariate group comparison test: Application to drug safety.

Hypothesis tests are used to compare and show the efficiency of drugs. However, usual tests do not perform properly whenever the number of variables is greater than, or of the same order of magnitude as, the number of observations. In this paper, we propose an alternative to usual multiclass multivariate group comparison tests such as MANOVA or Wilcoxon tests. We present a pattern recognition approach to compare drugs in high dimensional spaces. Our test is based on the classification probability of error of a classifier. The decision statistics is obtained using the leave one out procedure. The statistics power density function has been experimentally shown independent from the data distribution under the null hypothesis, that allows to determine the threshold, or the p-values, of our test. This test has been applied on clinical data registered to ensure the safety side and tolerability of drugs tested.