| 1. |
- Gharehbaghi, Arash, et al.
(författare)
-
A novel method for discrimination between innocent and pathological heart murmurs
- 2015
-
Ingår i: Medical Engineering and Physics. - : Elsevier. - 1350-4533 .- 1873-4030. ; 37:7, s. 674-682
-
Tidskriftsartikel (refereegranskat)abstract
- This paper presents a novel method for discrimination between innocent and pathological murmurs using the growing time support vector machine (GTSVM). The proposed method is tailored for characterizing innocent murmurs (IM) by putting more emphasis on the early parts of the signal as IMs are often heard in early systolic phase. Individuals with mild to severe aortic stenosis (AS) and IM are the two groups subjected to analysis, taking the normal individuals with no murmur (NM) as the control group. The AS is selected due to the similarity of its murmur to IM, particularly in mild cases. To investigate the effect of the growing time windows, the performance of the GTSVM is compared to that of a conventional support vector machine (SVM), using repeated random sub-sampling method. The mean value of the classification rate/sensitivity is found to be 88%/86% for the GTSVM and 84%/83% for the SVM. The statistical evaluations show that the GTSVM significantly improves performance of the classification as compared to the SVM.
|
|
| 2. |
- Rattfält, Linda, 1979-, et al.
(författare)
-
Robust heartbeat detector based on weighted correlation and multichannel input : Implementation on the ECG recorded with textile electrodes
- 2013
-
Ingår i: International Journal of E-Health and Medical Communications. - : IGI Global. - 1947-315X .- 1947-3168. ; 4:1, s. 61-71
-
Tidskriftsartikel (refereegranskat)abstract
- The aim of this study was to develop and evaluate a robust heartbeat detector for noisy electrocardiograms (ECGs) recorded with textile electrodes. The authors suggest a method based on weighted correlation in a multi-channel ECG to obtain a heartbeat detector. Signals were acquired during rest and at movements which simulate every day activities. From each recording a segment corresponding to a heartbeat was extracted and correlated with the whole signal. From the correlation data, heartbeat candidates were derived and weighted based on their variance similarity with the heartbeat model and previous heartbeats. Finally, the outputs of each channel were added to create the global output. The output was compared to the Pan Tompkins heartbeat detector. Results are promising for recordings at rest (sensitivity = 0.97, positive predictive value (PPV) = 0.97). For static muscle tension in the torso the results were much higher than the reference method (sensitivity = 0.77, PPV = 0.85). Corresponding values for the reference method were sensitivity = 0.96 and PPV = 0.95 at rest and sensitivity = 0.52 and PPV = 0.75 during muscle tension.
|
|
