Browsing by Author "Kianimajd, A."
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Comparison of different methods of measuring similarity in physiologic time seriesPublication . Kianimajd, A.; Ruano, M G; Carvalho, P.; Henriques, J.; Rocha, T.; Paredes, S.; Ruano, AntonioSearching for similarity between time series plays an important role when large amounts of information need to be clustered to integrate intelligent supported personal health care diagnosis systems. The performance of classification, clustering and disease prediction are influenced by the prior stage where similarity between time series is performed. Physiologic signals vary even within the same patient, so an analysis of their possible variation without affecting future clustering accuracy is hereby addressed. Commonly employed methods of measuring similarity between time series were tested on longer data segments than the typical cardiac cycle envisaging its use integrated on personalized health care cardiovascular diagnosis systems. Euclidean distance, Discrete Wavelet Transform, Discrete Fourier Transform, Correlation Coefficient, Mahalanobis distance, Minkowski Distance, and Dynamic Time Warping Distance were compared when 20 levels of small variations in amplitude scaling and shift, time scaling and shift, baseline variance and additive Gaussian noise are forced to the tested time series. Concentrating on the performance of the similarity methods in terms of their insensibility to small data variations results demonstrate that the time domain Correlation Coefficient is the most robust method while the Discrete Wavelet Transform is the elected one between the transform-based methods tested. Selection of a similarity method to be applied should also take into account implementation issues, namely need of data reduction to avoid computational burden, and in this case transform-based methods should be elected. (C) 2017, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.
- Validation of a similarity measurement method for clustering cardiac signalsPublication . Kianimajd, A.; Graca Ruano, Maria; Carvalho, P.; Henriques, J.; Rocha, T.; Paredes, S.; Morgado, M.; Bernardes, R.; Amador, M.; Afonso, P. M.Development of personalized cardiovascular management systems involves automatic identification of the current data as normal or pathological; considering cardiac data as time-series, the illness identification may be performed by seeking similarity between the current patient's time-series data and a reference signal and then proceeding to illness stratification (clustering). Seven of the most common methods of time-series similarity measurement were assessed by imposing 6 types of distortions to the reference signal, considering for each distortion 20 possible variations. This study employed 10 seconds length records of arterial blood pressure signals of healthy subjects, collected from a public database. Then clustering using Partitioning Around Medoids was performed among pathological and non-pathological data considering 3 different clusters. Clustering results confirm usage of the reduced basis Discrete Wavelet Transform resulting from the combination of Haar wavelet decomposition with the Karhunen-Loeve transforms, presenting an accuracy ranging from 76% to 85% when partitioning around Medoids clustering is used.