Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2019
Motion Artifact Removal of Photoplethysmogram (PPG) Signal.
Photoplethysmogram (PPG) has been used with great effect for predicting human vitals such as heart rate variability or blood oxygen saturation (SpO2), etc. The quality of PPG signal is affected mainly by noise, drift and motion artifacts. Although noise and drift are relatively easy to remove, motion artifacts pose a challenge. ⋯ However, this affects the signal as the PPG signal and motion artifacts distortion lie in the same frequency band, making it difficult to remove the motion artifacts without affecting the signal. In this work, we propose a motion artifact removal technique in time domain, which is based on correcting individual pulses in the PPG signal, considering a global pulse average and a windowed local pulse average. We show the effectiveness of our approach both qualitatively as well as quantitatively.
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2019
EEG Feature Analysis for Detecting the Fluctuation of Consciousness Level during Propofol Anesthesia.
Various EEG features have been proposed for differentiating the consciousness and unconsciousness states during general anesthesia. However, their performance for detecting the fluctuation of consciousness level remains unclear. In this work, we recorded 60-channels EEG data during propofol anesthesia, and extracted 110 EEG features that were shown to be sensitive to the change of consciousness level. ⋯ Using EEG features selected specifically for detecting consciousness fluctuation, approximately 10% improvement in accuracy was obtained. Our results suggested that the EEG features that were sensitive to the stable change of consciousness level and fluctuation of consciousness level were largely different. EEG features including theta band power and functional connectivity are more relevant to the fluctuation of consciousness level.
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2019
Machine Learning based SpO2 Computation Using Reflectance Pulse Oximetry.
Continuous monitoring of blood oxygen saturation level (SpO2) is crucial for patients with cardiac and pulmonary disorders and those undergoing surgeries. SpO2 monitoring is widely used in a clinical setting to evaluate the effectiveness of lung medication and ventilator support. Owing to its high levels of accuracy and stability, transmittance pulse oximeters are widely used in the clinical community to compute SpO2. ⋯ The proposed model overcomes the limitations imposed by the traditional R-value based calibration method through the use of a machine learning model using various time and frequency domain features. The model was trained and tested using the clinical data collected from 95 subjects with SpO2 levels varying from 81-100% using the custom SpO2 data acquisition platform along with reference measures. The proposed model has an absolute mean error of 0.5% with an accuracy of 96 ± 2% error band for SpO2 values ranging from 81-100%.
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2018
Early Prediction of Sepsis in EMR Records Using Traditional ML Techniques and Deep Learning LSTM Networks.
Sepsis is a life-threatening condition caused by infection and subsequent overreaction by the immune system. Physicians effectively treat sepsis with early administration of antibiotics. However, excessive use of antibiotics on false positive cases cultivates antibiotic resistant bacterial strains and can waste resources while false negative cases result in unacceptable mortality rates. ⋯ We used the Medical Information Mart for Intensive Care III (MIMIC3) dataset to test machine learning (ML) techniques including traditional methods (i.e., random forest (RF) and logistic regression (LR)) as well as deep learning techniques (i.e., long short-term memory (LSTM) neural networks). We successfully created a data pipeline to process and clean data, identified important predictive features using RF and LR techniques, and trained LSTM networks. We found that our best performing traditional classifier, RF, had an Area Under the Curve (AUC-ROC) score of 0.696, and our LSTM networks did not outperform RF.
-
Annu Int Conf IEEE Eng Med Biol Soc · Jul 2018
Exploring different impaired speed of genetic-related brain function and structures in schizophrenic progress using multimodal analysis.
Schizophrenia (SZ) is a highly heritable disease exhibiting substantial structural and functional brain impairments. The duration of illness and medication use may cause different presentations of impairments in patients. To understand the progressive variations of the disease, most recent studies have reported brain functional or structural abnormalities associated with illness duration, but a comprehensive study of pathology underlying brain structure, function and illness duration is still limited. ⋯ Results demonstrated significant group differences on GM and FC in hippocampus, temporal gyrus and cerebellum between SZ and HC, which are also significantly correlated with SNPs residing in genes like GABBR2, SATB2, CACNA1C, PDE4B, involved in pathways of cell junction, synapse and neuron projection. Moreover, two-sample t-tests showed that GM volume and FC strength presented similar trends of progressive decrease with the increase of the illness duration (HC>FESZ>CSZ). Besides that genetic-related GM and FC components both showed significant associations with illness duration, FC indicates the higher impairment speed than GM, suggesting that functional connectivity may serve as a more sensitive measure to detect the disruptions in SZ at the very early stage.