| 1. |
- Abtahi, Farhad, et al.
(författare)
-
Software tool for analysis of breathing-related errors in transthoracic electrical bioimpedance spectroscopy measurements
- 2012
-
Ingår i: Journal of Physics, Conference Series. - : IOP Publishing. - 1742-6588 .- 1742-6596. ; 407:1, s. 012028-
-
Tidskriftsartikel (refereegranskat)abstract
- During the last decades, Electrical Bioimpedance Spectroscopy (EBIS) has been applied in a range of different applications and mainly using the frequency sweep-technique. Traditionally the tissue under study is considered to be timeinvariant and dynamic changes of tissue activity are ignored and instead treated as a noise source. This assumption has not been adequately tested and could have a negative impact and limit the accuracy for impedance monitoring systems. In order to successfully use frequency-sweeping EBIS for monitoring time-variant systems, it is paramount to study the effect of frequency-sweep delay on Cole Model-based analysis. In this work, we present a software tool that can be used to simulate the influence of respiration activity in frequency-sweep EBIS measurements of the human thorax and analyse the effects of the different error sources. Preliminary results indicate that the deviation on the EBIS measurement might be significant at any frequency, and especially in the impedance plane. Therefore the impact on Cole-model analysis might be different depending on method applied for Cole parameter estimation.
|
|
| 2. |
- Cuba-Gyllensten, Illapha, et al.
(författare)
-
Removing respiratory artefacts from transthoracic bioimpedance spectroscopy measurements
- 2013
-
Ingår i: XV International Conference on Electrical Bio-Impedance (ICEBI) & XIV Conference on Electrical Impedance Tomography (EIT). - : Institute of Physics Publishing (IOPP).
-
Konferensbidrag (refereegranskat)abstract
- Transthoracic impedance spectroscopy (TIS) measurements from wearable textile electrodes provide a tool to remotely and non-invasively monitor patient health. However, breathing and cardiac processes inevitably affect TIS measurements, since they are sensitive to changes in geometry and air or fluid volumes in the thorax. This study aimed at investigating the effect of respiration on Cole parameters extracted from TIS measurements and developing a method to suppress artifacts. TIS data were collected from 10 participants at 16 frequencies (range: 10 kHz - 1 MHz) using a textile electrode system (Philips Technologie Gmbh). Simultaneously, breathing volumes and frequency were logged using an electronic spirometer augmented with data from a breathing belt. The effect of respiration on TIS measurements was studied at paced (10 and 16 bpm) deep and shallow breathing. These measurements were repeated for each subject in three different postures (lying down, reclining and sitting). Cole parameter estimation was improved by assessing the tidal expiration point thus removing breathing artifacts. This leads to lower intra-subject variability between sessions and a need for less measurements points to accurately assess the spectra. Future work should explore algorithmic artifacts compensation models using breathing and posture or patient contextual information to improve ambulatory transthoracic impedance measurements.
|
|
| 3. |
- Gyllensten, Illapha Cuba, et al.
(författare)
-
Identifying Types of Physical Activity With a Single Accelerometer : Evaluating Laboratory-trained Algorithms in Daily Life
- 2011
-
Ingår i: IEEE Transactions on Biomedical Engineering. - 0018-9294 .- 1558-2531. ; 58:9, s. 2656-2663
-
Tidskriftsartikel (refereegranskat)abstract
- Accurate identification of physical activity types has been achieved in laboratory conditions using single-site accelerometers and classification algorithms. This methodology is then applied to free-living subjects to determine activity behavior. This study is aimed at analyzing the reproducibility of the accuracy of laboratory-trained classification algorithms in free-living subjects during daily life. A support vector machine (SVM), a feed-forward neural network (NN), and a decision tree (DT) were trained with data collected by a waist-mounted accelerometer during a laboratory trial. The reproducibility of the classification performance was tested on data collected in daily life using a multiple-site accelerometer augmented with an activity diary for 20 healthy subjects (age: 30 +/- 9; BMI: 23.0 +/- 2.6 kg/m(2)). Leave-one-subject-out cross validation of the training data showed accuracies of 95.1 +/- 4.3%, 91.4 +/- 6.7%, and 92.2 +/- 6.6% for the SVM, NN, and DT, respectively. All algorithms showed a significantly decreased accuracy in daily life as compared to the reference truth represented by the IDEEA and diary classifications (75.6 +/- 10.4%, 74.8 +/- 9.7%, and 72.2 +/- 10.3%; p<0.05). In conclusion, cross validation of training data overestimates the accuracy of the classification algorithms in daily life.
|
|
