Syed Anas Imtiaz

Compression in Wearable Sensor Nodes: Impacts of Node Topology

Wearable sensor nodes monitoring the human body must operate autonomously for very long periods of time. Online and low-power data compression embedded within the sensor node is therefore essential to minimize data storage/transmission overheads. This paper presents a low-power MSP430 compressive sensing implementation for providing such compression, focusing particularly on the impact of the sensor node architecture on the compression performance. Compression power performance is compared for four different sensor nodes incorporating different strategies for wireless transmission/on-sensor-node local storage of data. The results demonstrate that the compressive sensing used must be designed differently depending on the underlying node topology, and that the compression strategy should not be guided only by signal processing considerations. We also provide a practical overview of state-of-the-art sensor node topologies. Wireless transmission of data is often preferred as it offers increased flexibility during use, but in general at the cost of increased power consumption. We demonstrate that wireless sensor nodes can highly benefit from the use of compressive sensing and now can achieve power consumptions comparable to, or better than, the use of local memory.

Automatic detection of sleep spindles using Teager energy and spectral edge frequency

Sleep spindles are the hallmark of N2 stage of sleep. They are transient waveforms observed on sleep electroencephalogram and their identification is required for sleep staging. Due to the large number of sleep spindles appearing on an overnight sleep EEG, automating the detection of sleep spindles would be desirable, not only to save specialist time but also for fully automated sleep staging systems. A simple algorithm for automatic sleep spindle detection is presented in this paper using only one channel of EEG input. This algorithm uses Teager energy and spectral edge frequency to mark sleep spindles and results in a sensitivity of 80% and specificity of about 98%. It is also shown that more than 91% of spindles detected by the algorithm were in N2 and N3 stages combined.

Automatic sleep staging using state machine-controlled decision trees.

Automatic sleep staging from a reduced number of channels is desirable to save time, reduce costs and make sleep monitoring more accessible by providing home-based polysomnography. This paper introduces a novel algorithm for automatic scoring of sleep stages using a combination of small decision trees driven by a state machine. The algorithm uses two channels of EEG for feature extraction and has a state machine that selects a suitable decision tree for classification based on the prevailing sleep stage. Its performance has been evaluated using the complete dataset of 61 recordings from PhysioNet Sleep EDF Expanded database achieving an overall accuracy of 82% and 79% on training and test sets respectively. The algorithm has been developed with a very small number of decision tree nodes that are active at any given time making it suitable for use in resource-constrained wearable systems.

Evaluating the use of line length for automatic sleep spindle detection.

Sleep spindles are transient waveforms observed on the electroencephalogram (EEG) during the N2 stage of sleep. In this paper we evaluate the use of line length, an efficient and low-complexity time domain feature, for automatic detection of sleep spindles. We use this feature with a simple algorithm to detect spindles achieving sensitivity of 83.6% and specificity of 87.9%. We also present a comparison of these results with other spindle detection methods evaluated on the same dataset. Further, we implemented the algorithm on a MSP430 microcontroller achieving a power consumption of 56.7 μW. The overall detection performance, combined with the low power consumption show that line length could be a useful feature for detecting sleep spindles in wearable and resource-constrained systems.

A Cough-Based Algorithm for Automatic Diagnosis of Pertussis

Pertussis is a contagious respiratory disease which mainly affects young children and can be fatal if left untreated. The World Health Organization estimates 16 million pertussis cases annually worldwide resulting in over 200,000 deaths. It is prevalent mainly in developing countries where it is difficult to diagnose due to the lack of healthcare facilities and medical professionals. Hence, a low-cost, quick and easily accessible solution is needed to provide pertussis diagnosis in such areas to contain an outbreak. In this paper we present an algorithm for automated diagnosis of pertussis using audio signals by analyzing cough and whoop sounds. The algorithm consists of three main blocks to perform automatic cough detection, cough classification and whooping sound detection. Each of these extract relevant features from the audio signal and subsequently classify them using a logistic regression model. The output from these blocks is collated to provide a pertussis likelihood diagnosis. The performance of the proposed algorithm is evaluated using audio recordings from 38 patients. The algorithm is able to diagnose all pertussis successfully from all audio recordings without any false diagnosis. It can also automatically detect individual cough sounds with 92% accuracy and PPV of 97%. The low complexity of the proposed algorithm coupled with its high accuracy demonstrates that it can be readily deployed using smartphones and can be extremely useful for quick identification or early screening of pertussis and for infection outbreaks control.

Recommendations for performance assessment of automatic sleep staging algorithms.

A number of automatic sleep scoring algorithms have been published in the last few years. These can potentially help save time and reduce costs in sleep monitoring. However, the use of both R&K and AASM classification, different databases and varying performance metrics makes it extremely difficult to compare these algorithms. In this paper, we describe some readily available polysomnography databases and propose a set of recommendations and performance metrics to promote uniform testing and direct comparison of different algorithms. We use two different polysomnography databases with a simple sleep staging algorithm to demonstrate the usage of all recommendations and presentation of performance results. We also illustrate how seemingly similar results using two different databases can have contrasting accuracies in different sleep stages. Finally, we show how selection of different training and test subjects from the same database can alter the final performance results.

Discriminating between best performing features for seizure detection and data selection

Seizure detection algorithms have been developed to solve specific problems, such as seizure onset detection, occurrence detection, termination detection and data selection. It is thus inherent that each type of seizure detection algorithm would detect a different EEG characteristic (feature). However most feature comparison studies do not specify the seizure detection problem for which their respective features have been evaluated. This paper shows that the best features/algorithm bases are not the same for all types of algorithms but depend on the type of seizure detection algorithm wanted. To demonstrate this, 65 features previously evaluated for online seizure data selection are re-evaluated here for seizure occurrence detection, using performance metrics pertinent to each seizure detection type whilst keeping the testing methodology the same. The results show that the best performing features/algorithm bases for data selection and occurrence detection algorithms are different and that it is more challenging to achieve high detection accuracy for the former seizure detection type. This paper also provides a comprehensive evaluation of the performance of 65 features for seizure occurrence detection to aid future researchers in choosing the best performing feature(s) to improve seizure detection accuracy.

TaiNi: Maximizing research output whilst improving animals' welfare in neurophysiology experiments

Understanding brain function at the cell and circuit level requires representation of neuronal activity through multiple recording sites and at high sampling rates. Traditional tethered recording systems restrict movement and limit the environments suitable for testing, while existing wireless technology is still too heavy for extended recording in mice. Here we tested TaiNi, a novel ultra-lightweight (<2 g) low power wireless system allowing 72-hours of recording from 16 channels sampled at ~19.5 KHz (9.7 KHz bandwidth). We captured local field potentials and action-potentials while mice engaged in unrestricted behaviour in a variety of environments and while performing tasks. Data was synchronized to behaviour with sub-second precision. Comparisons with a state-of-the-art wireless system demonstrated a significant improvement in behaviour owing to reduced weight. Parallel recordings with a tethered system revealed similar spike detection and clustering. TaiNi represents a significant advance in both animal welfare in electrophysiological experiments, and the scope for continuously recording large amounts of data from small animals.

An open-source toolbox for standardized use of PhysioNet Sleep EDF Expanded Database

PhysioNet Sleep EDF database has been the most popular source of data used for developing and testing many automatic sleep staging algorithms. However, the recordings from this database has been used in an inconsistent fashion. For example, arbitrary selection of start and end times from long term recordings, data-hypnogram mismatches, different performance metrics and hypnogram conversion from R&K to AASM. All these differences result in different data sections and performance metrics being used by researchers thereby making any direct comparison between algorithms very difficult. Recently, a superset of this database has been made available on PhysioNet, known as the Sleep EDF Expanded Database which includes 61 recordings. This provides an opportunity to standardize the way in which signals from this database should be used. With this goal in mind, we present in this paper a toolbox for automatically downloading and extracting recordings from the Sleep EDF Expanded database and converting them to a suitable format for use in MATLAB. This toolbox contains functions for selecting appropriate data for sleep analysis (based on our previous recommendations for sleep staging), hypnogram conversion and computation of performance metrics. Its use makes it simpler to start using the new sleep database and also provides a foundation for much-needed standardization in this research field.

Performance-Power Consumption Tradeoff in Wearable Epilepsy Monitoring Systems

Automated seizure detection methods can be used to reduce time and costs associated with analyzing large volumes of ambulatory EEG recordings. These methods however have to rely on very complex, power hungry algorithms, implemented on the system backend, in order to achieve acceptable levels of accuracy. In size, and therefore power-constrained EEG systems, an alternative approach to the problem of data reduction is online data selection, in which simpler algorithms select potential epileptiform activity for discontinuous recording but accurate analysis is still left to a medical practitioner. Such a diagnostic decision support system would still provide doctors with information relevant for diagnosis while reducing the time taken to analyze the EEG. For wearable systems with limited power budgets, data selection algorithm must be of sufficiently low complexity in order to reduce the amount of data transmitted and the overall power consumption. In this paper, we present a low-power hardware implementation of an online epileptic seizure data selection algorithm with encryption and data transmission and demonstrate the tradeoffs between its accuracy and the overall system power consumption. We demonstrate that overall power savings by data selection can be achieved by transmitting less than 40% of the data. We also show a 29% power reduction when selecting and transmitting 94% of all seizure events and only 10% of background EEG.