With the growing phenomenon of an aging population, an increasing numberof older people are living alone for domestic and social reasons. Based on thisfact, falling accidents become one of the most important factors in threateningthe lives of the elderly. Therefore, it is necessary to set up an application to de-tect the daily activities of the elderly. However, falling detection is difficult to recognize because the "falling" motion is an instantaneous motion and easy to confuse with others.In this thesis, three data mining methods were employed on wearable sensors' value; first which contains the continuous data set concerning eleven activities of daily living, and then an analysis of the different results was performed. Not only could the fall be detected, but other activities could also be classified. In detail, three methods including Back Propagation Neural Network, Support Vector Machine and Hidden Markov Model are applied separately to train the data set.What highlights the project is that a new  idea is put forward, the aim of which is to design a methodology of accurate classification in the time-series data set. The proposed approach, which includes obtaining of classifier parts and the application parts allows the generalization of classification. The preliminary results indicate that the new method achieves the high accuracy of classification,and significantly performs better than other data mining methods in this experiment.

Adherence to medication is generally described as a huge problem in the health care system. The term adherence is generally preferred by many health care providers as the word 'compliance' describes a patient who is passively taking medication as advised/ordered by the doctor. This thesis goes in depth in identifying the problems faced to achieve maximum adherence to medication and the important factors contributing to it. The objective is to come up with an alternative approach to help improve medical adherence using a smart watch based application that reminds patients to consume their medicines in a timely fashion. It addresses precisely which medication to take and in what quantity. This form of reporting and alerting is believed to achieve higher levels of adherence based on grounded theory. Shedding light on the methodologies used while clearly identifies the assumptions and limitations such a system can have.

In the past decade, Human Activity Recognition (HAR) has been an important part of the regular day to day life of many people. Activity recognition has wide applications in the field of health care, remote monitoring of elders, sports, biometric authentication, e-commerce and more. Each HAR application needs a unique approach to provide solutions driven by the context of the problem. In this dissertation, we are primarily discussing two application of HAR in different contexts. First, we design a novel approach for in-home, fine-grained activity recognition using multimodal wearable sensors on multiple body positions, along with very small Bluetooth beacons deployed in the environment. State-of-the-art in-home activity recognition schemes with wearable devices are mostly capable of detecting coarse-grained activities (sitting, standing, walking, or lying down), but cannot distinguish complex activities (sitting on the floor versus on the sofa or bed). Such schemes are not effective for emerging critical healthcare applications – for example, in remote monitoring of patients with Alzheimer's disease, Bulimia, or Anorexia – because they require a more comprehensive, contextual, and fine-grained recognition of complex daily user activities. Second, we introduced Watch-Dog – a self-harm activity recognition engine, which attempts to infer self-harming activities from sensing accelerometer data using wearable sensors worn on a subject's wrist. In the United States, there are more than 35,000 reported suicides with approximately 1,800 of them being psychiatric inpatients every year. Staff perform intermittent or continuous observations in order to prevent such tragedies, but a study of 98 articles over time showed that 20% to 62% of suicides happened while inpatients were on an observation schedule. Reducing the instances of suicides of inpatients is a problem of critical importance to both patients and healthcare providers. Watch-dog uses supervised learning algorithm to model the system which can discriminate the harmful activities from non-harmful activities. The system is not only very accurate but also energy efficient. Apart from these two HAR systems, we also demonstrated the difference in activity pattern between elder and younger age group. For this experiment, we used 5 activities of daily living (ADL). Based on our findings we recommend that a context aware age-specific HAR model would be a better solution than all age-mixed models. Additionally, we find that personalized models for each individual elder person perform better classification than mixed models.

This paper addresses a design exploration focusing on interactive jewelry conducted with newcome LGBT+ immigrants in Sweden, leading to a necklace named PoWo that is “powered” by the spoken word through a mobile application that reacts to customizable keywords triggering LED-lights in the necklace. Interactive jewelry is in this paper viewed as a medium with a simultaneous relation to wearer and spectator thus affording use on the themes of symbolism, emotion, body and communication. These themes are demonstrated through specific use scenarios of the necklace relating to the participants of the design exploration e.g. addressing consent, societal issues, meeting situations and expressions of love and sexuality.  The potential of speech based interactive jewelry is investigated in this paper e.g. finding speech recognition in LED-jewelry to act as an amplifier of spoken words, actions and meaning; and as a visible extension of the smartphone and human body. In addition use qualities of visibility, ambiguity, continuity and fluency are discussed in relation to speech based LED-jewelry.

A Wearable Mobility Monitoring System (WMMS) can be an advantageous device for rehabilitation decision-making. This thesis presents the design and evaluation of a proof-of-concept WMMS that uses the BlackBerry Smartphone platform. A Java program was developed for the BlackBerry 9550, using the integrated tri-axial accelerometer, Global Positioning System sensor (GPS), CMOS digital video camera, and timer to identify change-of-state (CoS) among static states, dynamic states, small activity of daily living (ADL) movements, and car riding. Static states included sitting, lying, standing, and taking an elevator. Dynamic states included walking on level ground, walking on stairs, and walking on a ramp. Small activity of daily living movements included bathroom activities, working in the kitchen, and meal preparation. Following feature extraction from the sensor data, two decision trees were used to distinguish CoS and mobility activities. CoS identification subsequently triggered video recording for improved mobility context analysis during post-processing.

Cada vez mais o rastreamento do olhar tem sido usado para interação humano-computador em diversos cenários, como forma de interação (usualmente substituindo o mouse, principalmente para pessoas com deficiências físicas) ou estudo dos padrões de atenção de uma pessoa (em situações como fazendo compras no mercado, olhando uma página na internet ou dirigindo um carro). Ao mesmo tempo, dispositivos vestíveis tais quais pequenas telas montadas na cabeça e sensores para medir dados relativos à saúde e exercício físico realizado por um usuário, também têm avançado muito nos últimos anos, finalmente chegando a se tornarem acessíveis aos consumidores. Essa forma de tecnologia se caracteriza por dispositivos que o usuário usa junto de seu corpo, como uma peça de roupa ou acessório. O dispositivo e o usuário estão em constante interação e tais sistemas são feitos para melhorar a execução de uma ação pelo usuário (por exemplo dando informações sobre a ação em questão) ou facilitar a execução de várias tarefas concorrentemente. O uso de rastreadores de olhar em computação vestível permite uma nova forma de interação para tais dispositivos, possibilitando que o usuário interaja com eles enquanto usa as mãos para realizar outra ação. Em dispositivos vestíveis, o consumo de energia é um fator importante do sistema que afeta sua utilidade e deve ser considerado em seu design. Infelizmente, rastreadores oculares atuais ignoram seu consumo e focam-se principalmente em precisão e acurácia, seguindo a ideia de que trabalhar com imagens de alta resolução e frequência maior implica em melhor desempenho. Porém tratar mais quadros por segundo ou imagens com resolução maior demandam mais poder de processamento do computador, consequentemente aumentando o gasto energético. Um dispositivo que seja mais econômico tem vários benefícios, por exemplo menor geração de calor e maior vida útil de seus componentes eletrônicos. Contudo, o maior impacto é o aumento da duração da bateria para dispositivos vestíveis. Pode-se economizar energia diminuindo resolução e frequência da câmera usada, mas os efeitos desses parâmetros na precisão e acurácia da estimação do olhar não foram investigados até o presente. Neste trabalho propomos criar uma plataforma de testes, que permita a integração de alguns algoritmos de rastreamento de olhar disponíveis, tais como Starburst, ITU Gaze Tracker e Pupil, para estudar e comparar o impacto da variação de resolução e frequência na acurácia e precisão dos algoritmos. Por meio de um experimento com usuários analisamos o desempenho e consumo desses algoritmos sob diversos valores de resolução e frequência. Nossos resultados indicam que apenas a diminuição da resolução de 480 para 240 linhas (mantendo a proporção da imagem) já acarreta em ao menos 66% de economia de energia em alguns rastreadores sem perda significativa de acurácia. / Eye tracking has been used more and more in human-computer interaction in several scenarios, as a form of interaction (mainly replacing the mouse for the physically handicapped) or as a means to study attention patterns of a person (performing activities such as grocery shopping, reading web pages or driving a car). At the same time, wearable devices such as small head-mounted screens and health-related sensors, have improved considerably in these years, finally becoming accessible to mainstream consumers. This form of technology is defined by devices that an user uses alongside his body, like a piece of clothing or accessory. The device and the user are in constant interaction and such systems are usually made to improve the user\'s ability to execute a task (for example, by giving contextualized information about the task in question) or to facilitate the parallel execution of several tasks. The use of eye trackers in wearable computing allows a new form of interaction in these devices, allowing the user to interact with them while performing another action with his hands. In wearable devices, the energy consumption is an important factor of the system which affects its utility and must be considered in its design. Unfortunately, current eye trackers ignore energy consumption and instead mainly focus on precision and accuracy, following the idea that working with higher resolution and higher frequency images will improve performance. However, processing more frames, or larger frames, per second require more computing power, consequentially increasing energy expense. A device that is more economical has several benefits, such as less heat generation and a greater life-span of its components. Yet the greatest impact is the increased battery duration for the wearable devices. Energy can be saved by lowering the frequency and resolution of the camera used by the tracker, but the effect of these parameters in the precision and accuracy of eye tracking have not been researched until now. In this work we propose an eye tracking testing platform, that allows integration with existing eye tracking algorithms such as Starburst, ITU Gaze Tracker and Pupil, to study and compare the impact of varying the resolution and frequency of the camera on accuracy and precision of the algorithms. Through a user experiment we analyzed the performance and consumption of these algorithms under various resolution and frequency values. Our result indicate that only lowering the resolution from 480 to 240 lines (keeping the image aspect ratio) already amounts to a 66% energy economy in some trackers without any meaningful loss of accuracy.

abstract: Current exosuit technologies utilizing soft inflatable actuators for gait assistance have drawbacks of having slow dynamics and limited portability. The first part of this thesis focuses on addressing the aforementioned issues by using inflatable actuator composites (IAC) and a portable pneumatic source. Design, fabrication and finite element modeling of the IAC are presented. Volume optimization of the IAC is done by varying its internal volume using finite element methods. A portable air source for use in pneumatically actuated wearable devices is also presented. Evaluation of the system is carried out by analyzing its maximum pressure and flow output. Electro-pneumatic setup, design and fabrication of the developed air source are also shown. To provide assistance to the user using the exosuit in appropriate gait phases, a gait detection system is needed. In the second part of this thesis, a gait sensing system utilizing soft fabric based inflatable sensors embedded in a silicone based shoe insole is developed. Design, fabrication and mechanical characterization of the soft gait detection sensors are given. In addition, integration of the sensors, each capable of measuring loads of 700N in a silicone based shoe insole is also shown along with its possible application in detection of various gait phases. Finally, a possible integration of the actuators, air source and gait detection shoes in making of a portable soft exosuit for knee assistance is given. / Dissertation/Thesis / Masters Thesis Mechanical Engineering 2020

This project aimed to explore how payments and identification might be moved from traditional wallets and implemented into a smart wearable.The thesis was that this would be doable through the use of NFC (Near Field Communication), IoT (Internet of Things) and Cloud-based services.Research was made by collecting data through questionnaires, local workshops, bodystorming and looking into online literature and articles on the subject of future consumer trends, wearable technology and biometric systems, security concerning wireless payments and identification as well as trend analysis articles concerning the Digital Self - also known as the Exoself - which came to have a major impact on the direction the project took. Through ideation five concepts concerning services and products using a person’s Digital Self were formed and evaluated with workshops. This eventually led to a concept called Shopal being decided as the endgame of the project. Shopal was a service which filtered the user’s information within the Digital Self to make shopping decisions easier, by learning what the user wants to know about certain types of products, then aiming to deliver said information. This service was called AIA (Artificial Intelligence Assistant). The second outcome of the Shopal concept was the AHW (At-Hand-Wearable) collection named SIGIL, which works as a means to extend AIA:s perception to what the user touches and interacts with, allowing it to gather information about the product, then delivering it through the physical AHW to the user.

abstract: Wearable technology has brought in a rapid shift in the areas of healthcare and lifestyle management. The recent development and usage of wearable devices like smart watches has created significant impact in areas like fitness management, exercise tracking, sleep quality assessment and early diagnosis of diseases like asthma, sleep apnea etc. This thesis is dedicated to the development of wearable systems and algorithms to fulfill unmet needs in the area of cardiorespiratory monitoring. First, a pneumotach based flow sensing technique has been developed and integrated into a face mask for respiratory profile tracking. Algorithms have been developed to convert the pressure profile into respiratory flow rate profile. Gyroscope-based correction is used to remove motion artifacts that arise from daily activities. By using Principal Component Analysis, the follow-up work established a unique respiratory signature for each subject based on the flow profile and lung parameters computed using the wearable mask system. Next, wristwatch devices to track transcutaneous gases like oxygen (TcO2) and carbon dioxide (TcCO2), and oximetry (SpO2) have been developed. Two chemical sensing approaches have been explored. In the first approach, miniaturized low-cost commercial sensors have been integrated into the wristwatch for transcutaneous gas sensing. In the second approach, CMOS camera-based colorimetric sensors are integrated into the wristwatch, where a part of camera frame is used for photoplethysmography while the remaining part tracks the optical signal from colorimetric sensors. Finally, the wireless connectivity using Bluetooth Low Energy (BLE) in wearable systems has been explored and a data transmission protocol between wearables and host for reliable transfer has been developed. To improve the transmission reliability, the host is designed to use queue-based re-request routine to notify the wearable device of the missing packets that should be re-transmitted. This approach avoids the issue of host dependent packet losses and ensures that all the necessary information is received. The works in this thesis have provided technical solutions to address challenges in wearable technologies, ranging from chemical sensing, flow sensing, data analysis, to wireless data transmission. These works have demonstrated transformation of traditional bench-top medical equipment into non-invasive, unobtrusive, ergonomic & stand-alone healthcare devices. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2020

abstract: Global industrialization and urbanization have led to increased levels of air pollution. The costs to society have come in the form of environmental damage, healthcare expenses, lost productivity, and premature mortality. Measuring pollutants is an important task for identifying its sources, warning individuals about dangerous exposure levels, and providing epidemiologists with data to link pollutants with diseases. Current methods for monitoring air pollution are inadequate though. They rely on expensive, complex instrumentation at limited fixed monitoring sites that do not capture the true spatial and temporal variation. Furthermore, the fixed outdoor monitoring sites cannot warn individuals about indoor air quality or exposure to chemicals at worksites. Recent advances in manufacturing and computing technology have allowed new classes of low-cost miniature gas sensor to emerge as possible alternatives. For these to be successful however, there must be innovations in the sensors themselves that improve reliability, operation, and their stability and selectivity in real environments. Three novel gas sensor solutions are presented. The first is the development of a wearable personal exposure monitor using all commercially available components, including two metal oxide semiconductor gas sensors. The device monitors known asthma triggers: ozone, total volatile organic compounds, temperature, humidity, and activity level. Primary focus is placed on the ozone sensor, which requires special circuits, heating algorithm, and calibration to remove temperature and humidity interferences. Eight devices are tested in multiple field tests. The second is the creation of a new compact optoelectronic gas sensing platform using colorimetric microdroplets printed on the surface of a complementary-metal-oxide-semiconductor (CMOS) imager. The nonvolatile liquid microdroplets provide a homogeneous, uniform environment that is ideal for colorimetric reactions and lensless optical measurements. To demonstrate one type of possible indicating system gaseous ammonia is detected by complexation with Cu(II). The third project continues work on the CMOS imager optoelectronic platform and develops a more robust sensing system utilizing hydrophobic aerogel particles. Ammonia is detected colorimetrically by its reaction with a molecular dye, with additives and surface treatments enhancing uniformity of the printed films. Future work presented at the end describes a new biological particle sensing system using the CMOS imager. / Dissertation/Thesis / Doctoral Dissertation Materials Science and Engineering 2020