<p>In recent years, there has been an increase in the popularity of wearable sensors such as electroencephalography (EEG) sensors, electromyography (EMG) sensors, gyroscopes, accelerometers, and photoplethysmography (PPG) sensors. This work is focused on PPG sensors, which are used to measure heart rate in real time. They are currently used in many commercial products such as Fitbit Watch and Muse Headband. Due to their low cost and relative implementation simplicity, they are easy to add to custom-built wearable devices.</p><p><br></p> <p>We built an Arduino-based wearable wrist sensor-kit that consists of a PPG sensor in addition to other low cost commercial biosensors to measure biosignals such as pulse rate, skin temperature, skin conductivity, and hand motion. The purpose of the sensor-kit is to analyze the effects of stress on students in a classroom based on changes in their biometric signals. We noticed some failures in the measured PPG signal, which could negatively affect the accuracy of our analysis. We conjectured that one of the causes of failure is movement. Therefore, in this thesis, we build automatic failure detection methods and use these methods to study the effect of movement on the signal.</p><p><br></p> <p>Using the sensor-kit, PPG signals were collected in two settings. In the first setting, the participants were in a still sitting position. These measured signals were manually labeled and used in signal analysis and method development. In the second setting, the signals were acquired in three different scenarios with increasing levels of activity. These measured signals were used to investigate the effect of movement on the reliability of the PPG sensor. </p><p><br></p> <p>Four types of failure detection methods were developed: Support Vector Machines (SVM), Deep Neural Networks (DNN), K-Nearest Neighbor (K-NN), and Decision Trees. The classification accuracy is evaluated by comparing the resulting Receiver Operating Characteristic (ROC) curves, Area Above the Curve (AAC), as well as the duration of failure and non-failure sequences. The DNN and Decision Tree results are found to be the most promising and seem to have the highest error detection accuracy. </p> <p> </p> <p>The proposed classifiers are also used to assess the reliability of the PPG sensor in the three activity scenarios. Our findings indicate that there is a significant presence of failures in the measured PPG signals at rest, which increases with movement. They also show that it is hard to obtain long sequences of pulses without failure. These findings should be taken into account when designing wearable systems that use heart rate values as input.</p>

La reconnaissance d’activités humaines est un sujet de recherche d’actualité comme en témoignent les nombreux travaux de recherche sur le sujet. Dans ce cadre, la reconnaissance des activités physiques humaines est un domaine émergent avec de nombreuses retombées attendues dans la gestion de l’état de santé des personnes et de certaines maladies, les systèmes de rééducation, etc.Cette thèse vise la proposition d’une approche pour la reconnaissance automatique et non-intrusive d’activités physiques quotidiennes, à travers des capteurs inertiels de type accéléromètres, placés au niveau de certains points clés du corps humain. Les approches de reconnaissance d’activités physiques étudiées dans cette thèse, sont catégorisées en deux parties : la première traite des approches supervisées et la seconde étudie les approches non-supervisées. L’accent est mis plus particulièrement sur les approches non-supervisées ne nécessitant aucune labellisation des données. Ainsi, nous proposons une approche probabiliste pour la modélisation des séries temporelles associées aux données accélérométriques, basée sur un modèle de régression dynamique régi par une chaine de Markov cachée. En considérant les séquences d’accélérations issues de plusieurs capteurs comme des séries temporelles multidimensionnelles, la reconnaissance d’activités humaines se ramène à un problème de segmentation jointe de séries temporelles multidimensionnelles où chaque segment est associé à une activité. L’approche proposée prend en compte l’aspect séquentiel et l’évolution temporelle des données. Les résultats obtenus montrent clairement la supériorité de l’approche proposée par rapport aux autres approches en termes de précision de classification aussi bien des activités statiques et dynamiques, que des transitions entre activités. / Human activity recognition is currently a challengeable research topic as it can be witnessed by the extensive research works that has been conducted recently on this subject. In this context, recognition of physical human activities is an emerging domain with expected impacts in the monitoring of some pathologies and people health status, rehabilitation procedures, etc. In this thesis, we propose a new approach for the automatic recognition of human activity from raw acceleration data measured using inertial wearable sensors placed at key points of the human body. Approaches studied in this thesis are categorized into two parts : the first one deals with supervised-based approaches while the second one treats the unsupervised-based ones. The proposed unsupervised approach is based upon joint segmentation of multidimensional time series using a Hidden Markov Model (HMM) in a multiple regression context where each segment is associated with an activity. The model is learned in an unsupervised framework where no activity labels are needed. The proposed approach takes into account the sequential appearance and temporal evolution of data. The results clearly show the satisfactory results of the proposed approach with respect to other approaches in terms of classification accuracy for static, dynamic and transitional human activities

INTRODUÇÃO: A gagueira do desenvolvimento é uma patologia caracterizada por rupturas involuntárias no fluxo de fala, causadas por déficits no processamento neuromotor para a fala. Estes sintomas têm como consequência uma variedade de reações fisiológicas, comportamentais, cognitivas e emocionais para a pessoa que gagueja. O objetivo geral deste trabalho foi investigar os efeitos do tratamento fonoaudiológico para gagueira baseado em dispositivos de alteração de feedback auditivo, e compará-los aos efeitos do tratamento tradicional, baseado em estratégias comportamentais. MÉTODOS: Esta tese foi desenvolvida no formato de compilação de artigos científicos. Foram compilados três artigos, dois já publicados em periódicos com indexação na base de dados Web of Science e um terceiro artigo já submetido para publicação, em processo de revisão por pares. Foi incluída ainda uma análise crítica das contribuições dos artigos compilados para a área da Fonoaudiologia. RESULTADOS: Os estudos analisados não apresentaram diferenças significativas entre os resultados das duas abordagens terapêuticas investigadas; ambos os protocolos terapêuticos alcançaram melhora pós-tratamento, medida por meio da performance da fluência de fala e da qualidade de vida auto referida. CONCLUSÕES: Os dispositivos de alteração do feedback auditivo podem ser utilizados como recurso no tratamento da gagueira em adultos / INTRODUCTION: Developmental stuttering is a pathology characterized by involuntary disruptions in speech flow and is caused by deficits in neuromotor processing involved in speech production. These symptoms result in a variety of physiological, behavioral, cognitive and emotional consequences to the person who stutters. This study aimed to investigate the effects of a therapeutic approach for stuttering treatment based on altered auditory feedback devices, and to compare them with a traditional behavioral therapeutic approach. METHODS: This thesis is a compilation of two papers published in peerreviewed scientific journals indexed in the Web of Science database, and one unpublished manuscript, submitted to a scientific journal and currently awaiting review by peers. A critical analysis of the contributions of the compiled papers to the Speech-Language Pathology field was performed. RESULTS: The results presented in the analyzed papers did not show differences between the outcomes of the two therapeutic approaches; both protocols achieved improvements, as measured by speech fluency performance and selfreported quality of life. CONCLUSIONS: Altered auditory feedback devices may be used as a resource for the treatment of adults who stutter

L’Artériopathie Oblitérante des Membres Inférieurs (AOMI) est une pathologie chronique grave, induite par le processus physiopathologique d’athérosclérose. Diagnostiquée par la mesure de l’Index de Pression Systolique (IPS) de repos cette pathologie se traduit chez la plupart des patients par une ischémie (apports sanguins insuffisants) à l’exercice pouvant causer l’apparition de douleurs au niveau des membres inférieurs lors de la marche. Le premier axe de travail de cette thèse visait à évaluer la pertinence pédagogique des méthodes actuelles d’enseignement de l’IPS dans les facultés de médecine françaises. Les résultats des premiers travaux de recherche qui ont été menés montrent que les méthodes actuelles ne permettent pas aux étudiants de maîtriser les fondamentaux de cette mesure diagnostique. D’autres travaux ont alors été menés afin de définir la place et la forme que devrait prendre l’enseignement de l’IPS dans le cursus médical afin qu’il puisse être acquis par les futurs médecins de manière fiable et durable. Le second axe de travail de cette thèse visait à proposer une nouvelle approche méthodologique basée sur l’utilisation conjointe de deux moniteurs portables (accéléromètre et montre marqueur d’événements) permettant la quantification des douleurs ischémiques à l’exercice des patients atteints d’AOMI en condition de vie réelle. Les résultats issus de ces travaux ouvrent des perspectives nouvelles pour le chercheur ou le clinicien pour la compréhension des limitations fonctionnelles des patients atteints d’AOMI dans leur contexte de vie. / Lower-extremity Peripheral Artery Disease (PAD) is a severe non-communicable disease that is associated with atherosclerosis. Diagnosed by Ankle- Brachial Index (ABI), PAD leads, in most patients, to ischemia (mismatch between blood demand and blood supply) during exercise that may lead to the occurrence of pain in the lower extremities during walking. The first axis of the present thesis was to assess the level of knowledge on this diagnostic tool among French medical students. Our results suggest that the way of teaching or learning the ABI procedure is deficient. According to these results we conducted another study to determine the best teaching method to improve students’ ABI proficiency. The second axis of the present thesis was to proposed a new method using wearable monitors (accelerometer and Micro Motion logger watch) to objectively identify and quantify lower limb walking pain manifestation as well as stop induced by walking pain during daily life in PAD patients. Our results can provide researchers and clinicians with a more realistic and holistic view of the functional limitations of PAD patients in free-living conditions.

L’électrophysiologie est l’étude des signaux électriques et électrochimiques générés par certaines cellules spécifiques tout comme par des organes entiers. Elle donne aux médecins l’opportunité de suivre le fonctionnement d’un seul neurone mais aussi de l’intégralité du cerveau. L’enregistrement de ces activités est essentiel pour le diagnostic de pathologies aussi diverses que les arythmies cardiaques, l’épilepsie ou la dégénération musculaire. Dans cette thèse, nous étudions différents types d’électrodes cutanées à base de matériaux organiques, de leur conception à leur évaluation préclinique. Notre approche est basée sur l’utilisation du polymère conducteur PEDOT :PSS et de gels ioniques, qui réduisent l’impédance de l’interface électrode-peau. De plus, nos électrodes sont conçues avec différents substrats fins et souples, plastiques ou textiles. Ceci appelle de nouvelles techniques de fabrications adaptées à ces substrats et aux matériaux organiques. Les électrodes sont caractérisées puis testées sur des volontaires afin de démontrer leurs excellentes performances par rapport aux électrodes médicales usuelles. L’évaluation de leur capacité à réduire le bruit et de leur stabilité sur plusieurs jours est effectuée sur des signaux venant des activités musculaires, cardiaques et cérébrales. Nous présentons également une électrode microscopique dite « active », basée sur le transistor organique électrochimique. Celui-ci permet d’amplifier et de filtrer in situ le signal. Parce que nos électrodes organiques cutanées possèdent un important potentiel industriel et clinique, nous étudions maintenant leur intégration dans des dispositifs médicaux de pointe. / Electrophysiology is the study of electrical and electrochemical signals generated by specific cells or whole organs. It gives doctors the opportunity to track the physiological behavior of a single neuron, as well as the integral brain. The recording of these activities is essential to diagnose and better understand diseases like cardiac arrhythmias, epilepsy, muscular degeneration and many more. In this thesis, we study different types of cutaneous electrodes based on organic materials, from conception to pre-clinical evaluation. Our approach is based on the usage of PEDOT:PSS conducting polymer and ionic gels in order to reduce impedance at the skin-electrode interface. Moreover, the substrate of our electrodes is made with different materials such as thin and conformable plastics and textiles. Our devices are then flexible, motion resistant and can be integrating into clothes. We developed new fabrication processes, considering the different substrates and organic materials specifics. The electrodes were characterized and then tested on human volunteers to show their excellent performance in comparison to standard medical electrodes. The evaluation of noise reduction capabilities and possibilities to perform long-term recordings were established on signals coming from muscles, heart and brain. Furthermore, we present a hundred micrometer-small “active” electrode, based on the organic electrochemical transistor. It enables in situ amplification and filtering of recorded signals. The wearable organic electrodes developed in this work are of great industrial and clinic interest. Future work will aim to integrate these technologies into state-of-the-art medical devices.

Information and communication technologies hope to revolutionize the healthcare industry with innovative and affordable solutions with a focus on pervasive care. Wearable sensors products can provide monitoring in a natural environment with a constant stream of information, enriching healthcare practices and enabling better pervasive care. Wearable sensor technologies could monitor patients' mobility, gait, tremor, daily activity and other health indicators in real time that could allow for simple, non-invasive, tracking of spine care that may lead to increased patient engagement, integration, feedback, post-surgery analysis, monitoring of patient's condition, patient's data extraction and analysis and possibly aiding in better diagnosis, intervention, adherence to treatment for the betterment of quality of care. This research focuses on the assessment of technology adoption potential of medical devices particular to tracking the mobility of patients of neurosurgery and orthopedics. Wearable medical devices that track the mobility of patients after spinal procedures could help surgeons in providing post-operative care, analysis of treatment outcomes and patient mobility. The assessment of those devices by physicians is a complex process associated with various perspectives and criteria. Therefore, the objective of this research is to assess the potential for technology adoption of those wearable medical devices through development of a hierarchical decision-making model (HDM) that incorporates the relevant perspectives and criteria encompassing the needs of hospital neurological surgery and orthopedics departments. The proposed research builds on an existing body of knowledge researched through literature review and background of the field and expands the health technology assessment field by implementation of a holistic, comprehensive and multi-perspective approach to technology assessment in wearable sensor products adoption for pervasive care in neurosurgery and orthopedics. The Hierarchical Decision Model (HDM) approach is used to break the problem down into hierarchical levels and then calculate the alternatives using pairwise comparison scales and a judgment quantification technique. Inconsistencies, disagreement, sensitivity and scenario analysis are performed as well. HDM research software is created with Ruby and R to facilitate the computation of some of these important model parameters to higher precision than is available in current statistical analysis software packages or extensions targeted for decision making. Patient perspective dominates as the main perspective for the technology adoption potential of wearable devices for pervasive care in neurosurgery and orthopedics, followed by technical and financial perspectives. Valedo, a wearable device aimed to relieve back pain through exercises, motivation and mobility tracking, received the highest ranking for adoption potential, while other devices also received high relative scores. The framework could serve as a supplementary technology assessment tool and could be tested in other settings: private, small clinic etc. with the experts and special needs of physicians in particular healthcare departments.

A large amount of the information in conversations come from non-verbal cues such as facial expressions and body gesture. These cues are lost when we don't communicate face-to-face. But face-to-face communication doesn't have to happen in person. With video communication we can at least deliver information about the facial mimic and some gestures. This thesis is about video communication over distances; communication that can be available over networks with low capacity since the bitrate needed for video communication is low. A visual image needs to have high quality and resolution to be semantically meaningful for communication. To deliver such video over networks require that the video is compressed. The standard way to compress video images, used by H.264 and MPEG-4, is to divide the image into blocks and represent each block with mathematical waveforms; usually frequency features. These mathematical waveforms are quite good at representing any kind of video since they do not resemble anything; they are just frequency features. But since they are completely arbitrary they cannot compress video enough to enable use over networks with limited capacity, such as GSM and GPRS. Another issue is that such codecs have a high complexity because of the redundancy removal with positional shift of the blocks. High complexity and bitrate means that a device has to consume a large amount of energy for encoding, decoding and transmission of such video; with energy being a very important factor for battery-driven devices. Drawbacks of standard video coding mean that it isn't possible to deliver video anywhere and anytime when it is compressed with such codecs. To resolve these issues we have developed a totally new type of video coding. Instead of using mathematical waveforms for representation we use faces to represent faces. This makes the compression much more efficient than if waveforms are used even though the faces are person-dependent. By building a model of the changes in the face, the facial mimic, this model can be used to encode the images. The model consists of representative facial images and we use a powerful mathematical tool to extract this model; namely principal component analysis (PCA). This coding has very low complexity since encoding and decoding only consist of multiplication operations. The faces are treated as single encoding entities and all operations are performed on full images; no block processing is needed. These features mean that PCA coding can deliver high quality video at very low bitrates with low complexity for encoding and decoding. With the use of asymmetrical PCA (aPCA) it is possible to use only semantically important areas for encoding while decoding full frames or a different part of the frames. We show that a codec based on PCA can compress facial video to a bitrate below 5 kbps and still provide high quality. This bitrate can be delivered on a GSM network. We also show the possibility of extending PCA coding to encoding of high definition video.

當今運動健身蔚為風潮,加上網路資源普及,許多人藉由數位教學影片鍛鍊體 魄,不但能自由安排時間,也能在家參與課程,成為新世代的學習方式。傳統上,運 動健身較好的方式為教練在旁協助,除了口頭給予即時的指示,還能直接以身體觸碰 學員姿勢不良的部位,引導其肢體伸展、調整身體重心、提醒放鬆過於緊繃之部位, 然而新世代的學習方式透過教學影片無法立即對運動者當下的肢體動作做出反應,過 程中全倚賴運動者本身對於肢體的認知,往往和影片中教練的動作有落差而不自知, 因此,本研究期望藉由觸覺回饋輔助使用者做出正確的姿勢或動作。 過去針對觸覺回饋的研究相當地多,甚至可回溯至1950年代,然而將觸覺回饋應 用於運動指引的研究近年才漸漸出現,其應用層面僅限於提示作用,無法引導使用者 該如何動作,且目前回饋方式仍需倚賴受測者記憶動作與觸覺回饋的對應關係,無法 直覺做出反應。本研究模仿肌肉群收縮帶動肢體運動之方式,設計人工外部肌肉引導 手臂旋前旋後動作,人工外部肌肉包含步進馬達產生拉力、魚線及鬆緊帶模擬肌肉分 佈、收縮以及袖套包覆手臂帶動旋轉,系統設計歷經三版本的演進,最終設計出一套 具引導效果的觸覺回饋穿戴式裝置。 系統評估共邀請10位受測者進行實驗,結果證實此套裝置能有效提供方向性指示 (正確率98%),且受測者普遍反應裝置提供的回饋方式相當直覺,手臂會有被帶動 的感覺,能馬上知道該如何轉動手臂。實驗更進一步測試引導手臂轉動特定角度,實 驗結果效果也相當好,平均誤差在3度以內,此外,亦探討實驗過程中受測者對觸覺回 饋的行為反應,作為日後系統改良或觸覺回饋設計的參考。 / Nowadays, exercise and fitness have become a growing trend. Since the access to the internet resources is very easy and popular, many people choose to do exercise through digital online videos, which not only they can arrange their own exercising schedule, but also they can learn the courses at home. Traditionally, a better way for exercise learning is getting assistance from a professional coach, who can give instruction immediately, and adjust by direct body contact right away while the exercisers act incorrectly. However, the online video can not accomplish the purpose. On the condition that the exercisers rely only on the cognition of their own bodies they might not notice their posture different from the video. This research aimed to provide guidelines to do the correct posture or movement through tactile feedback. From past till now, the researches of tactile feedback are of considerable numbers, we can find the related researches back to 1950s. Recently it starts to be applied in exercising guiding. However, the applications only provide passive instructions, which require users to memorize the relationship between the tactile feedbacks and the correspond actions. Users are unable to react by instinct. In this research, we imitate the way of body movement driven by the muscles contraction. We design artificial external muscles on a sleeve to guide forearm pronation and supination. The wearable tactile feedback sleeve consists of stepper motors to provide pulling force, fishing wires and elastics to imitate muscle contraction to drive the forearm to roll. This system design has been revised three times, and we finally established a wearable tactile feedback device which has guiding effect. 10 participants are recruited for the experiments. The result showed that this device can guide forearm rolling successfully (the accuracy is 98%). The participants commented that the feedback is very close to instinct. They felt their arm was guided by the device, and knew the exact moment to roll their forearm. In the second experiment, we tried to guide the forearm rolling for several target angles and the result was quite promising. The mean error is within 3 degrees. We also reported the participants’ reactions through our tactile feedback system. We will expand the system to guide the other parts of human body in the future.

This thesis investigates inkjet printing of fully-integrated modules fabricated on organic substrates as a system-level solution for ultra-low-cost and eco-friendly mass production of wireless sensor modules. Prototypes are designed and implemented in both traditional FR-4 substrate and organic substrate. The prototype on organic substrate is referred to as a Smart Wireless Integrated Module (SWIM). Parallels are drawn between FR-4 manufacturing and inkjet printing technology, and recommendations are discussed to enable the potential of inkjet printing technology. Finally, this thesis presents novel applications of SWIM technology in the area of wearable and implantable electronics. Chapter 1 serves as an introduction to inkjet printing technology on organic substrates, wireless sensor networks (WSNs), and the requirements for low-power consumption, low-cost, and eco-friendly technology. Chapter 2 discusses the design of SWIM and its implementation using traditional manufacturing techniques on FR-4 substrate. Chapter 3 presents a benchmark prototype of SWIM on paper substrate. Challenges in the manufacturing process are addressed, and solutions are proposed which suggest future areas of research in inkjet printing technology. Chapter 4 presents novel applications of SWIM technology in the areas of implantable and wearable electronics. Chapter 5 concludes the thesis by discussing the importance of this work in creating a bridge between current inkjet printing technology and its future.

The purpose of this thesis is to explore the subject of industrial exoskeleton in accord-ance to the applicability of the technology preventing musculoskeletal disorders within the automotive industry. The modern technology of exoskeletons has a limited field of research and knowledge and is in need to be studied to provide organisations with proper findings for understanding the applicability of the technology. In the auto-motive industry musculoskeletal disorders (MSDs) is one of the most common disor-ders among employees and industries work constantly to decrease and prevent MSDs within their work environments. By conducting literature reviews, the status of exo-skeleton research and development concluded that academic research mostly focuses on technological development of exoskeletons, and not laboratory and/or field testing of currently available industrial exoskeletons. However, through database and website searches, twenty-four available industrial exoskeletons were identified which could be applicable within the automotive industry. Through literature and a case illustration, a number of potential causes for MSDs within the automotive industry were identified and a framework was developed in order to match appropriate available industrial ex-oskeleton to be used in potentially preventing common MSDs. The discussion of the thesis highlights the benefits and challenges of implementing an industrial exoskele-ton within an industry. Proper research on the currently available industrial exoskele-tons is lacking and creates questions of reliability for the technology. However, devel-opment of industrial exoskeletons have shown to focus on prevention of the most common causes of MSDs within industries in their design and development, making the applicability of industrial exoskeletons highly possible.