Features extraction from vital signs to characterize acute respiratory distress syndrome
| Autor: | Aline Taoum, Farah Mourad-Chehade, Marwan Sabbagh, Hassan Amoud |
|---|---|
| Přispěvatelé: | Université Libanaise, Lou Ruvo Center for Brain Health [Las Vegas], Cleveland Clinic, Laboratoire Modélisation et Sûreté des Systèmes (LM2S), Institut Charles Delaunay (ICD), Université de Technologie de Troyes (UTT)-Centre National de la Recherche Scientifique (CNRS)-Université de Technologie de Troyes (UTT)-Centre National de la Recherche Scientifique (CNRS) |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
medicine.medical_specialty
ARDS Respiratory rate business.industry Vital signs 020207 software engineering 02 engineering and technology medicine.disease Standard deviation Sample entropy Internal medicine 0202 electrical engineering electronic engineering information engineering Detrended fluctuation analysis Cardiology Kurtosis Poincaré plot Medicine 020201 artificial intelligence & image processing [SDV.IB]Life Sciences [q-bio]/Bioengineering business |
| Zdroj: | 2018 IEEE 4th Middle East Conference on Biomedical Engineering (MECBME) 2018 IEEE 4th Middle East Conference on Biomedical Engineering (MECBME), Mar 2018, Tunis, Tunisia. pp.62-66, ⟨10.1109/MECBME.2018.8402407⟩ |
| DOI: | 10.1109/MECBME.2018.8402407⟩ |
| Popis: | International audience; Acute Respiratory Distress Syndrome (ARDS) is a crucial pathology affecting 40% of patients, and may cause mortality. However, there is a lack in the literature in characterizing it using vital signs. Thus, the aim of this study is to characterize the ARDS using different time series data, and to find out linear and non-linear parameters that can differentiate statistically between subjects who are going to develop ARDS and others who do not. For this purpose, three vital signs are considered that are the heart rate, the respiratory rate and the oxygen saturation. These signals are used for the simplicity of the procedure of acquisition from patients without disturbing them. From these signals, windows having different lengths (12, 18, 24 and 30 hours) are taken from the end of the signals or before the onset of ARDS. Then, Linear and non-linear parameters were extracted, as mean, standard deviation, skewness, sample entropy, detrended fluctuation analysis, poincare plot, and others. Statistical analysis was performed using the kolmogorov-smirnov test. Results show that sample entropy and detrended fluctuation analysis presented significant difference between ARDS and non-ARDS groups over almost all the windows for heart rate and respiratory rate, while oxygen saturation presented different parameters for each window as a, kurtosis, SD2 and both factors from the detrended fluctuation analysis. Therefore, ARDS can be modeled using parameters extracted from time series data in order to implement prediction algorithms. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|