Energy-Aware Embedded Classifier Design for Real-Time Emotion Analysis

Detection and classification of human emotions from multiple bio-signals has a wide variety of applications. Though electronic devices are available in the market today that acquire multiple body signals, the classification of human emotions in real-time, adapted to the tight energy budgets of wearable embedded systems is a big challenge. In this paper we present an embedded classifier for real-time emotion classification. We propose a system that operates at different energy budgeted modes, depending on the available energy, where each mode is constrained by an operating energy bound. The classifier has an offline training phase where feature selection is performed for each operating mode, with an energy-budget aware algorithm that we propose. Across the different operating modes, the classification accuracy ranges from 95% - 75% and 89% - 70% for arousal and valence respectively. The accuracy is traded off for less power consumption, which results in an increased battery life of up to 7.7 times (from 146.1 to 1126.9 hours).

An Energy Efficient Power Converter for Zero Power Wearable Devices

Milad Ataei Ashtiani

Early diagnosis of Alzheimer's and epilepsy requires monitoring a subject's development of symptoms through electroencephalography (EEG) signals over long periods. Wearable devices enable convenient monitoring of biosignals, unlike complex and costly hospital equipment. The key to achieving a fit and forgettable wearable device is to increase its operating cycle and decrease its size and weight. Instead of batteries, which limit the life cycle of electronic devices and set their form factor, body heat and environmental light can power wearable devices through energy-scavenging technologies. The harvester transducers should be tailored according to on the application and the sensor placement. This leaves a wide variety of transducers with an extensive range of impedances and voltages. To realize an autonomous wearable device, the power converter energy harvester, has to be very efficient and maintain its efficiency despite potential transducer replacement or variations in environmental conditions. This thesis presents a detailed design of an efficient integrated power converter for use in an autonomous wearable device. The design is based on the examination of both power losses and power transfer in the power converter. The efficiency bound of the converter is derived from the specifications of its transducer. The tuning ranges for the reconfigurable parameters are extracted to keep the converter efficient with variations in the transducer specifications. With the efficient design and the manual tuning of the reconfigurable parameters, the converter can work optimally with different types of transducers, and keeps its efficiency in the conversion of low voltages from the harvesters. Measurements of the designed converter demonstrate an efficiency of higher than 50% and 70% with two different transducers having an open-circuit voltage as low as 20 mV and 100 mV, respectively. The power converter should be able to reconfigure itself without manual tunings to keep its efficiency despite changes in the harvesters' specifications. The second portion of this dissertation addresses this issue with a proposed design methodology to implement a control section. The control section adjusts the converter's reconfigurable parameters by examining the power transfer and loss and through concurrent closed loops. The concurrent loops working together raise a serious concern regarding stability. The system is designed and analyzed in the time domain with the state-space averaging (SSA) model to address the stability issue. The ultra-low-power control section obtained from the SSA model estimates the power and loss with a reasonable accuracy, and adjusts the timings in a stable manner. The entire control section consumes only 30 nW dynamic power at 10 kHz. The control section tunes the converter's speed or its working frequency depending on the available power. The frequency clocks the entire architecture, which is designed asynchronously; therefore, the power consumption of the system depends on the power available from the transducer. The system is implemented using 0.18 µm CMOS technology. For an input as low as 7 mV, the converter is not only functional but also has an efficiency of more than 40%. The efficiency can reach 70% with an input voltage of 50 mV. The system operates in a range of just a few of millivolts to half a volt with ample efficiencies. It can work at an optimal point with different transducers and environmental conditions.

EPFL2018

Monitoring of Cardiovascular and Neurological Diseases Using Wearable Devices

Dionisije Sopic

In this thesis, I focus on monitoring of patients suffering from cardiovascular and neurological diseases through the use of wearable devices. The main diseases considered in this thesis are atrial fibrillation (AF), myocardial infarction (MI), and epilepsy. The proposed methods for the detection of the considered cardiovascular diseases use ECG as the main biosignal, whereas the main biosignal used for detecting epileptic seizures is EEG. Firstly, I propose a hierarchical heart-rhythm classification method for AF detection from a single lead ECG recording. The proposed method is based on the features that capture the morphology of important ECG signal segments, along with heart-rate oscillations and time and frequency domain features of the ECG signal. Furthermore, the classification scheme used in this method incorporates two different classifiers: a multiclass classifier and a random forest classifier, resulting in an F1 score of 78.95% for AF detection. Secondly, I address the problem of early detection and prevention of MI by introducing a real-time event-driven classification technique that uses a hierarchical classifier with multiple levels. The main goal of this technique is to maintain a high classification accuracy while reducing the complexity of the classification algorithm. The reduction in computational complexity, in turn, results in a longer battery life, which is an important factor for wearable devices. The proposed technique is validated on a public database and ported on a real-life wearable device. The experimental evaluation of the proposed technique on MI data shows that this scheme reduces the energy consumption by a factor of 2.60, with no significant loss in classification performance. The third chapter of this thesis is split into two parts and it focuses on epilepsy. In the first part, I present a real-time method for personalized epileptic seizure detection that uses EEG signals acquired from two electrode pairs: F7T3, and F8T4. The proposed method reaches a sensitivity of 90.98% and specificity of 92.10% on the database used in this study. Furthermore, this method is ported on a pair of eyeglasses in which the used electrodes are embedded and hidden in the temples, allowing for 2.71 days of operation on a single battery charge. This method overcomes the lack of portability and the effect of social stigma of EEG caps, which are used as a gold standard technique for epilepsy detection. Since the main pitfall of epilepsy detection algorithms is the unacceptably high number of false alarms, in the last part of my thesis, I propose an interpretable patient-specific approach to false alarm reduction for epilepsy detection. This approach is based on similarly occurring morphological EEG signal patterns (seizure signature) that occur frequently during seizures. The proposed approach has been experimentally validated on more than 5500 hours of long-term EEG recordings, resulting in a high classification performance with no false positive alarms. The high degree of interpretability of this method can help physicians discover seizures faster, as well as it can be used to improve data labeling quality in publicly available databases, which confirms its applicability for long-term seizure monitoring.

EPFL2020

CAFS: Cost-Aware Features Selection Method for Multimodal Stress Monitoring on Wearable Devices

Adriana Arza Valdes, David Atienza Alonso, João Pedro de Matos Rodrigues, Niloofar Momeni, Maria del Carmen Sandi Perez

Objective: Today, stress monitoring on wearable devices is challenged by the tension between high-detection accuracy and battery lifetime driven by multimodal data acquisition and processing. Limited research has addressed the classification cost on multimodal wearable sensors, particularly when the features are cost-dependent. Thus, we design a Cost-Aware Feature Selection (CAFS) methodology that trades-off between prediction-power and energy-cost for multimodal stress monitoring. Methods: CAFS selects the most important features under different energy-constraints, which allows us to obtain energy-scalable stress monitoring models. We further propose a self-aware stress monitoring method that intelligently switches among the energy-scalable models, reducing energy consumption. Results: Using CAFS methodology on experimental data and simulation, we reduce the energy-cost of the stress model designed without energy constraints up to 94.37%. We obtain 90.98% and 95.74% as the best accuracy and confidence values, respectively , on unseen data, outperforming state-of-the-art studies. Analyzing our interpretable and energy-scalable models, we showed that simple models using only heart rate (HR) or skin conductance level (SCL), confidently predict acute stress for HR > 93.30BP M and non-stress for SCL < 6.42µS, but, outside these values, a multimodal model using respiration and pulse wave's features is needed for confident classification. Our self-aware acute stress monitoring proposal saves 10x energy and provides 88.72% of accuracy on unseen data. Conclusion: We propose a comprehensive solution for the cost-aware acute stress monitoring design addressing the problem of selecting an optimized feature subset considering their cost-dependency and cost-constraints. Significant: Our design framework enables long-term and confident acute stress monitoring on wearable devices.

2021