The concept of immersion in the context of VR represents how deeply a user is involved in the virtual environment. Developers of novel methods of sensory feedback in VR systems need to evaluate the contribution of these developments in improving the user experience. This creates a need for measuring immersion of a VR system. Questionnaires and behavioural analysis are used to evaluate immersion traditionally. A measure of immersion is required to be objective, real time and quantifiable. Since the existing methods fail to achieve these, this study proposes to investigate the suitability of biosignals to objectively quantify immersion in real time. The proposed approach employs experiments where EEG, ECG and EDA signals are acquired while subjects experience custom designed VR scenario with varying levels of immersion.

Detecting low quality records in ECG acquired with point of care devices

ECG obtained from personal devices by untrained users need to be assessed for quality before they are sent to physicians through telemedicine services. We developed a machine learning algorithm that identifies low quality ECG recordings to be used in these devices. The proposed algorithm uses a combination of rule based classification and decision trees. A set of 7 features describing physiological relationships between 12 ECG leads were used for machine learning. The algorithm was trained and tested using Physionet Computing in Cardiology Challenge 2011 test database using 5 fold cross validation. The technique of oversampling was used to reduce the effect of class imbalance in the database. The algorithm achieved a sensitivity of 91.2% and a specificity of 91.5% to differentiate low and high quality ECG recordings.

ECGImages: Covolutional Neural Networks for Cardiac Arrhythmia Detection

In this project I explored the limitations of a consumer EEG headset in applications of Motor Imagery detection. By designing and conducting experiments we evaluated the performance of the commercial device in comparison to that of a medical grade data. The consumer device showed poor performance in terms of, higher noise, the limited number of electrodes, the sub-optimal electrode locations. I developed an improved motor imagery detection algorithm that attempts to address this problem using machine learning .

Machine learning for the detection of Atrial Fibrillation using short single lead ECG

Point of care ECG devices can improve the early detection of atrial fibrillation (AF). The efficiency of such devices depends on the capability of automatic AF detection against normal sinus rhythm and other arrhythmias from a short single lead ECG record in the presence of noise and artifacts. The objective of this study was to develop an algorithm that classifies a short single lead ECG record into 'Normal', 'AF', 'Other' and 'Noisy' classes, and identify the challenges in developing such algorithms and potential mitigation steps. Rule-based identification was used to detect lead inversion and records too noisy to be of immediate use. A set of statistical and morphological features describing the rhythm was then extracted, and support vector machine classifiers were used to classify records into three classes: 'Normal', 'AF' or 'Other'. The algorithm was trained and tested using 12 186 short single lead ECGs recorded on a point of care device made available via the Computing in Cardiology Challenge 2017. The algorithm achieved a sensitivity of 77.5%, a specificity of 97.9% and an accuracy of 96.1% in the detection of AF from a non-AF rhythm in a five-fold cross validation. It achieved F1 measures of 89%, 78% and 67% for 'Normal', 'AF' and 'Other' classes, respectively, when evaluated with a hidden test set. The overall challenge score was 78%. Most existing algorithms can distinguish the AF rhythm from the normal sinus rhythm when ECG recordings are clean and are obtained with multi-lead systems, while their ability to discriminate against other arrhythmias and noise remains largely unknown. This study proposes an algorithm that classifies a short single lead ECG record from point of care devices into 'Normal', 'AF', 'Other' and 'Noisy' classes and discusses computational approaches to mitigate any unique challenges such as lead inversion, low amplitude signals, noise and artifacts.

Lumen Boundary Detection in Intra Venous Ultrasound Images