Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference
-
Conf Proc IEEE Eng Med Biol Soc · Jan 2015
An improved artifact removal algorithm for continuous cardiac output and blood pressure recordings.
Measurement artifacts are common in hemodynamic recordings such as cardiac output and blood pressure. Manual artifact removal is cumbersome for large datasets, and automatic processing using algorithms may reduce workload and provide more reproducible outcomes. This paper presents an artifact removal algorithm which is more aggressive compared to a previously described method. ⋯ Precision, recall and F-score was determined by agreement with manual inspection by an expert. Based on the total of all measurements from CO and MAP by LiDCO and CO and MAP by Nexfin, precision was 86%, 79%, 79% and 68% respectively (87%, 62%, 76% and 58% for the reference method), recall was 97%, 94%, 89% and 97% (31%, 6%, 28% and 6% for reference), F-score was 91%, 85%, 84% and 80% (46%, 10%, 41% and 10% for reference). The proposed algorithm offers an improved performance in removing true artifacts, in some cases a reduced ability to preserve true measurements, but an improved overall accuracy.
-
Conf Proc IEEE Eng Med Biol Soc · Jan 2015
Quantitative EEG markers in severe post-resuscitation brain injury with therapeutic hypothermia.
Therapeutic hypothermia has been regarded as one of the most effective post-cardiac arrest (CA) treatments to improve survival and functional recovery. However, many clinical prognostic markers after resuscitation have become less reliable under hypothermia. In this study, we applied and compared two developed quantitative measures - information quantity (IQ) and sub-band IQ (SIQ) - to evaluate the accuracy of EEG markers on predicting cortical recovery under therapeutic hypothermia. ⋯ Contrary to IQ recovery but similarly to NDS scores, the SIQ recovery was not significantly different between the hypothermia (0.66±0.04) and normothermia (0.64±0.04) groups (p>0.05). IQ could identify the presence of high-frequency oscillations during the recovery from severe brain injury. We demonstrated that while SIQ was able to provide additional sub-band EEG information related to the recovery of different brain functions, both early IQ and SIQ markers are able to accurately predict neurologic outcome after CA.
-
Conf Proc IEEE Eng Med Biol Soc · Jan 2015
The impact of data preprocessing in traumatic brain injury detection using functional magnetic resonance imaging.
Traumatic brain injury (TBI) can adversely affect a person's thinking, memory, personality and behavior. For this reason new and better biomarkers are being investigated. Resting state functional network connectivity (rsFNC) derived from functional magnetic resonance (fMRI) imaging is emerging as a possible biomarker. ⋯ Results suggest that correction for motion variance before spatial smoothing is the best alternative. Following this preprocessing option a significant group difference was found between cerebellum and supplementary motor area/paracentral lobule. In this case the mTBI group exhibits an increase in rsFNC.
-
Conf Proc IEEE Eng Med Biol Soc · Jan 2015
Sleep stage classification by body movement index and respiratory interval indices using multiple radar sensors.
Disturbed sleep has become more common in recent years. To increase the quality of sleep, undergoing sleep observation has gained interest as an attempt to resolve possible problems. In this paper, we evaluate a non-restrictive and non-contact method for classifying real-time sleep stages and report on its potential applications. ⋯ The accuracy was 79.3% for classification and 71.9% for estimation. This is a novel system for measuring body movements and body-surface movements that are induced by respiration and for measuring high sensitivity pulse waves using multiple radar signals. This method simplifies measurement of sleep stages and may be employed at nursing care facilities or by the general public to increase sleep quality.
-
Conf Proc IEEE Eng Med Biol Soc · Jan 2015
Seizure detection using regression tree based feature selection and polynomial SVM classification.
This paper presents a novel patient-specific algorithm for detection of seizures in epileptic patients with low hardware complexity and low power consumption. In the proposed approach, we first compute the spectrogram of the input fragmented EEG signals from three or four electrodes. Each fragmented data clip is one second in duration. ⋯ The algorithm is tested using the intra-cranial EEG (iEEG) from the UPenn and Mayo Clinic's Seizure Detection Challenge database. It is shown that the proposed algorithm can achieve a sensitivity of 100.0%, an average area under curve (AUC) of 0.9818, a mean detection horizon of 5.8 seconds, and a specificity of 99.9% on using half of the training data for classification. The proposed approach also achieved a mean AUC of seizure detection and early seizure detection of 0.9136 on the testing data.