Feature selection techniques are very useful approaches for dimensionality reduction in data analysis. They provide interpretable results by reducing the dimensions of the data to a subset of the original set of features. When the data lack annotations, unsupervised feature selectors are required for their analysis. Several algorithms for this aim exist in the literature, but despite their large applicability, they can be very inaccessible or cumbersome to use, mainly due to the need for tuning non-intuitive parameters and the high computational demands. In this work, a publicly available ready-to-use unsupervised feature selector is proposed, with comparable results to the state-of-the-art at a much lower computational cost. The suggested approach belongs to the methods known as spectral feature selectors. These methods generally consist of two stages: manifold learning and subset selection. In the first stage, the underlying structures in the high-dimensional data are extracted, while in the second stage a subset of the features is selected to replicate these structures. This paper suggests two contributions to this field, related to each of the stages involved. In the manifold learning stage, the effect of non-linearities in the data is explored, making use of a radial basis function (RBF) kernel, for which an alternative solution for the estimation of the kernel parameter is presented for cases with high-dimensional data. Additionally, the use of a backwards greedy approach based on the least-squares utility metric for the subset selection stage is proposed. The combination of these new ingredients results in the utility metric for unsupervised feature selection U2FS algorithm. The proposed U2FS algorithm succeeds in selecting the correct features in a simulation environment. In addition, the performance of the method on benchmark datasets is comparable to the state-of-the-art, while requiring less computational time. Moreover, unlike the state-of-the-art, U2FS does not require any tuning of parameters.

Electrical storm (ES) in ICD patients, defined as 3 or more appropriate ICD interventions within a time span of 24 hours, is a medical emergency associated with adverse outcome. However, it is debated if ES is only a marker of progressive near end-stage cardiac disease or an ar-rhythmogenic entity on its own. Better understanding and prediction are necessary to manage the burden of ES. The goal of this study is to explore the relation between the presence of fragmented QRS (fQRS) and the manifestation of electrical storm in patients with an ICD for ischemic heart disease. A balanced dataset of 100 patients was considered for this study, where 50 patients with ischemic heart disease and dilated cardiomyopathy present ES. 12-lead ECG signals were analyzed from 3 years before until the moment of ES, divided in 4 visits. The fQRS level in the 12-lead ECG data recorded in each visit was automatically quantified with a score between 0 and 1 for each lead. A Friedman test between the first and last visit for each of the groups showed a significant increase in the average level of fragmentation for the patients presenting ES, absent in the control group. This suggests that there is a trend towards deterioration in fQRS for patients manifesting ES with an ICD for ischemic heart disease.

Tetralogy of Fallot (ToF) is a congenital structural heart disease. While early diagnosis and corrective surgery allow most patients to live normal lives, some patients slowly deteriorate. The current inability to quantify the deterioration and predict these events prompts a data driven approach. Laplacian Eigenmaps (LEs) are a dimensionality reduction technique that can be used to project multi-lead ECGs onto a lower dimensional space. This pilot study aims to evaluate the ability of LEs to characterize deterioration of ToF patients. A general LE model is constructed, based on the 12-lead ECG recordings of 20 healthy controls. A set of distance metrics are developed to quantify the overall changes between different ECG recordings within this LE model. Statistically significant differences between control and ToF subjects were observed for most of the distance metrics. The analysis of changes over time in ToF patients indicates a general trend of increased distance over time in all the metrics, which can be related to a worsening condition. This indicates the relevance of LEs in multi-lead ECG processing, particularly for deterioration analysis.