A research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, in partial fulfilment of the requirements for the degree of Master of Science in Physiotherapy Johannesburg, 2012 / Introduction: Stroke continues to be a major public health problem for both the developed and developing world despite the various advances in health care. The economic burden of stroke is ever increasing and in light of this, return to work post-stroke is becoming an important area of research for therapists. Thus the main aims of the study were to establish the rate of return to work of patients following stroke and to establish the patients’ perceived factors which influence their return to work after stroke. Method This was a quantitative cross sectional study. A sample of convenience of ninety seven participants were selected from a list of patients obtained from various rehabilitation units and government clinics within the Gauteng province. A self-designed questionnaire was used. A pilot study was conducted to determine the questionnaire’s reliability and validity, and the validity, inter-rater and intra-rater reliability were all found to be satisfactory. The participants were contacted and interviewed at their homes or a location suitable for them using a self-designed questionnaire. The percentage of patients that returned to work was determined and reasons for returning or not returning to work were summarised using frequencies and percentages. A univariate and then multivariate analysis was performed to establish perceived factors which had an influence on return to work Results The study sample had more males than females with an overall mean age of 51years. They were mostly black and between 18 months and 24 months post-stroke. The most common co-morbidities were fatigue and hypertension. There was a 34% return to work rate, with 3% stopping work after a period of time, leaving 31% of the sample working at the time of interview. The most common reasons for returning to work were financial (77%), enjoyment of work (77%) and personal development (73%). For those who did not return to v work the two most common reasons were upper limb dysfunction (61%) and walking difficulties (53%). The main factors that decreased likelihood of return to work included depression and not paying life insurance or monthly car repayments. Conclusion The return to work rate following stroke in this study group is in line with other countries around the world, although it is still relatively low with less than a third of patients with stroke returning to work. Enjoyment of work was shown to be as important a motivating factor for return to work as finances, and physical fallout was the most demotivating factor. Depression was the most likely factor to decrease return to work.

Practicing Motor Imagery (MI) with a Brain-Computer Interface (BCI) has shown promise in promoting motor recovery in stroke patients. A BCI records a person’s brain activity and provides feedback to the person in real time, which allows the person to practice his or her brain activity. By imagining a movement (performing MI) such as gripping with their hand, cortical areas in the brain are activated that largely overlaps with those activated during the actual hand movement. A BCI can provide positive feedback when the hand-related cortical areas are activated during MI, which helps a person to learn how to perform MI. Despite evidence that stroke patients may recover some motor function from practicing MI with BCI feedback thanks to the feedback provided from a BCI, the effectiveness and reliability of BCI-based rehabilitation are still poor.  A BCI can detect MI by analyzing patterns of features from the brain activity. The most common features are extracted from the oscillatory activity in the brain.  In BCI research, MI is often treated as a static pattern of features, which is detected by using machine learning algorithms to assign activity into a binary state. However, this model of MI may be inaccurate. Analyzing brain activity as dynamically varying over time and with a continuous measure of strength could better represent the cortical activity related to MI.  In this Licentiate thesis, I explore a method for analyzing the temporal dynamic of MI-activity with a continuous measure of strength. Brain activity was recorded with electroencephalography (EEG) and subject-specific feature patterns were extracted from a group of healthy subjects while they performed MI of two opposing hand movements: opening and closing the hand. Although MI of the two same-hand movements could not be discriminated, the continuous output from a machine learning algorithm was shown to correlate well with MI-related feature patterns. The temporal analysis also revealed that MI is dynamically encoded early, but later stabilizes into a more static pattern of brain activity. Last, to accommodate for higher temporal resolution of MI, I designed and evaluated a BCI framework by its feedback delay and uncertainty as a function of the stress on the system and found a non-linear correlation. These results could be essential for developing a BCI with time-critical feedback. To summarize, in this Licentiate thesis I propose a promising method for analyzing and extracting a temporal representation of MI, enabling relevant and continuous neurofeedback which may contribute to clinical advances in BCI-based stroke rehabilitation.

Indiana University-Purdue University Indianapolis (IUPUI)

Bakgrund Stroke är idag en av de vanligaste sjukdomarna som drabbar ca 25 000 människor varje år i Sverige. Flertalet komplikationer, så kallade sekvele kan uppstå relaterat till stroken. En vanlig sekvele är att drabbas av smärta i olika former. Smärtan i sin tur kan påverka människors vardag och hela dennes liv. Idag fokuserar Socialstyrelsens rekommendationer på de farmakologiska åtgärderna vid smärta efter stroke. Det råder även brist i de nationella riktlinjerna för smärta efter stroke vilket kan försvåra för vårdpersonalen att välja rätt behandling för rätt patientgrupp. De icke-farmakologiska åtgärderna är ett stort outforskat ämne med många möjligheter och appliceras under ramen för omvårdnad. Syfte Syftet var att studera de icke-farmakologiska åtgärderna och dess effekter på smärta relaterat till stroke. Metod Studien genomfördes i form av en icke-systematisk litteraturöversikt över kvantitativa studier. Utifrån de urvalskriterierna baserades studien på 18 kvantitativa artiklar hämtade från databaserna CINAHL och PubMed. En kvalitetsgranskning gjordes utifrån Sophiahemmets bedömningskriterier för att säkerställa artiklarnas validitet. Samtliga artiklar analyserades utifrån en integrerad analysmetod. Resultat Studien visar att det finns flertal olika icke-farmakologiska åtgärder som kan användas vid smärta efter stroke. Åtgärderna delades in i kategorierna motirriterande åtgärder, kropp och själsåtgärder och perifera mekanismer. Majoriteten av de ingående studierna har erhållit förbättrat resultat avseende smärta med statistisk signifikans. Patienter erhöll även en förbättrad självständighet och livskvalitet relaterad till åtgärderna. Även skillnader i effektiviteten av åtgärderna kunde påvisas hos olika typer av smärtklassifikationer ur dess karaktär och tidsaspekt. Slutsats Erhållna resultatet tyder på att det finns ett stort antal icke-farmakologiska åtgärder som uppvisar god effekt och skulle kunna användas mer inom sjukvården. Samtliga åtgärder kan sjuksköterskan utföra alternativt göra en behovsbedömning och därefter remittera vidare/delegera. Genom att tillämpa en evidensbaserad omvårdnadsprocess kan patienternas behov av omvårdnad anpassas för en optimal smärtlindring. / Background Stroke is one of the most common diseases today, affecting 25 000 people annually in Sweden. Complications, known as sequelae, can occur in relation to the stroke. A common one is to suffer from pain in various forms and can affect people's everyday life. The National Board of Health and Welfare's recommendations focus on the pharmacological measures for pain after a stroke. There is also a lack of national guidelines for post-stroke pain, which can make it difficult for healthcare professionals to choose the right treatment for the right patient group. The non-pharmacological measures are an unexplored topic with many possibilities and are applied in the context of nursing. Aim The aim was to study the non-pharmacological interventions and their effects on stroke- related pain. Method The study was conducted as a non-systematic literature review of quantitative studies. Within the selection criteria, the study was based on 18 quantitative articles retrieved from the databases CINAHL and PubMed. A quality review was carried out based on Sophiahemmets assessment criteria to ensure the validity. An integrated analyse-method was used. Results The study shows that there are several different non-pharmacological measures that can be used in the treatment of stroke pain. The measures were divided into the category's counterirritant measures, body and mind measures and as peripheral mechanisms. Most of the studies have obtained improved pain outcomes. Differences in the effectiveness of the interventions could also be demonstrated in different types of pain. Conclusions The results indicate that there is large number of non-pharmacological measures that show effect and could be used more in the health care system. All measures can be carried out by nurse, to assess needs and therefore delegate. By applying an evidence-based nursing process, patients' nursing needs can be adapted for optimal pain relief.

Introduction: Early supported discharge intervention offers strokerehabilitation in the home environment. Digitalisation in society may make it demanding to perform everyday activities in traditional ways. Considering that the use of technology now is part of our daily activities, occupational therapists need to maintain a client-centred practice and make sure that their own competencies in using the technology is adequate. Aim: The aim with this study is to explore occupational therapists' perceptions and experiences of interventions involving digital technology with post-stroke patients in early supported discharge settings. Method: A qualitative study with aphenomenological approach was carried out. Four interviews of occupational therapists were conducted. A qualitative content analysis approach was done for the analysis. Findings: Two main categories emerged which were interpreted into the theme Digitalization requires clinical reasoning. Many factors that involved the digitalization of today's society affected both the patients' occupations pre- and post-stroke and therefore also affected the OTs roles and work process. Conclusions: This study highlights that more knowledge is required regarding digital technology and enhancing competence and clinical reasoning within the occupational therapy practice process. Significance: This3study may provide a better understanding for the potential effects that digitalization has within occupational therapy practise.

A major dilemma that is being addressed in the current project is the discrepancies between healthcare system's expectations for a rapid and successful rehabilitation process and patients after having a stroke ability to meet these expectations while striving to adapt to the calamitous event in their life. Emphasizing a more biomedical approach and under implementation of psychosocial approaches, poor acknowledging of patients' basic psychological needs lead to poor motivation, therapeutic disengagement and may lead to a rehabilitation failure. To cope with this gap in the process of stroke rehabilitation, an educational program aiming for occupational therapists working with patients after having a stroke in their acute and post-acute rehabilitation phases was constructed. The program guides practitioners for effective communication with their patients, building a needs-supportive environment and addressing their patients' basic psychological needs in light of the selfdetermination theory, theories of adaptation from occupational therapy perspectives and considering occupational justice and the ICF model. A clinical reasoning, step-by-step problem solving is introduced using adaptation of known models and innovated models for interventions that were created for this purpose. Program delivery through a series of 4-webinar modules is illustrated with their learning objectives, assignments and discussions. The program evaluation and implementation are expected to be the initiator of a change in the health and rehabilitation climate and in Israel.

Stroke is one of the leading causes of long-term motor disability, including oro-facial impairments which affect speech and swallowing. Over the last decades, rehabilitation programs have evolved from utilizing mainly compensatory measures to focusing on recovering lost function. In the continuing effort to improve recovery, the concept of biofeedback has increasingly been leveraged to enhance self-efficacy, motivation and engagement during training. Although both speech and swallowing disturbances resulting from oro-facial impairments are frequent sequelae of stroke, efforts to develop sensing technologies that provide comprehensive and quantitative feedback on articulator kinematics and kinetics, especially those of the tongue, and specifically during post-stroke speech and swallowing therapy have been sparse. To that end, such a sensing device needs to accurately capture intraoral tongue motion and contact with the hard palate, which can then be translated into an appropriate form of feedback, without affecting tongue motion itself and while still being light-weight and portable. This dissertation proposes the use of an intraoral sensing principle known as optopalatography to provide such feedback while also exploring the design of optopalatographic devices itself for use in dysphagia and dysarthria therapy. Additionally, it presents an alternative means of holding the device in place inside the oral cavity with a newly developed palatal adhesive instead of relying on dental retainers, which previously limited device usage to a single person. The evaluation was performed on the task of automatically classifying different functional tongue exercises from one another with application in dysphagia therapy, whereas a phoneme recognition task was conducted with application in dysarthria therapy. Results on the palatal adhesive suggest that it is indeed a valid alternative to dental retainers when device residence time inside the oral cavity is limited to several tens of minutes per session, which is the case for dysphagia and dysarthria therapy. Functional tongue exercises were classified with approximately 61 % accuracy across subjects, whereas for the phoneme recognition task, tense vowels had the highest recognition rate, followed by lax vowels and consonants. In summary, retainer-free optopalatography has the potential to become a viable method for providing real-time feedback on tongue movements inside the oral cavity, but still requires further improvements as outlined in the remarks on future development.:1 Introduction 1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Problem statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Goals and contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.4 Scope and limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2 Basics of post-stroke speech and swallowing therapy 2.1 Dysarthria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2 Dysphagia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3 Treatment rationale and potential of biofeedback . . . . . . . . . . . . . . . . . 13 2.4 Summary and conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3 Tongue motion sensing 3.1 Contact-based methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3.1.1 Electropalatography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3.1.2 Manometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.1.3 Capacitive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.2 Non-contact based methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.2.1 Electromagnetic articulography . . . . . . . . . . . . . . . . . . . . . . . 23 3.2.2 Permanent magnetic articulography . . . . . . . . . . . . . . . . . . . . 24 3.2.3 Optopalatography (related work) . . . . . . . . . . . . . . . . . . . . . . 25 3.3 Electro-optical stomatography . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.4 Extraoral sensing techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.5 Summary, comparison and conclusion . . . . . . . . . . . . . . . . . . . . . . . 29 4 Fundamentals of optopalatography 4.1 Important radiometric quantities . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.1.1 Solid angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.1.2 Radiant flux and radiant intensity . . . . . . . . . . . . . . . . . . . . . 33 4.1.3 Irradiance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.1.4 Radiance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.2 Sensing principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.2.1 Analytical models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 4.2.2 Monte Carlo ray tracing methods . . . . . . . . . . . . . . . . . . . . . . 37 4.2.3 Data-driven models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.2.4 Model comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.3 A priori device design consideration . . . . . . . . . . . . . . . . . . . . . . . . 41 4.3.1 Optoelectronic components . . . . . . . . . . . . . . . . . . . . . . . . . 41 4.3.2 Additional electrical components and requirements . . . . . . . . . . . . 43 4.3.3 Intraoral sensor layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 5 Intraoral device anchorage 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 5.1.1 Mucoadhesion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 5.1.2 Considerations for the palatal adhesive . . . . . . . . . . . . . . . . . . . 48 5.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 5.2.1 Polymer selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 5.2.2 Fabrication method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 5.2.3 Formulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.2.4 PEO tablets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.2.5 Connection to the intraoral sensor’s encapsulation . . . . . . . . . . . . 50 5.2.6 Formulation evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 5.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 5.3.1 Initial formulation evaluation . . . . . . . . . . . . . . . . . . . . . . . . 54 5.3.2 Final OPG adhesive formulation . . . . . . . . . . . . . . . . . . . . . . 56 5.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 6 Initial device design with application in dysphagia therapy 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 6.2 Optode and optical sensor selection . . . . . . . . . . . . . . . . . . . . . . . . . 60 6.2.1 Optode and optical sensor evaluation procedure . . . . . . . . . . . . . . 61 6.2.2 Selected optical sensor characterization . . . . . . . . . . . . . . . . . . 62 6.2.3 Mapping from counts to millimeter . . . . . . . . . . . . . . . . . . . . . 62 6.2.4 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 6.3 Device design and hardware implementation . . . . . . . . . . . . . . . . . . . . 64 6.3.1 Block diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 6.3.2 Optode placement and circuit board dimensions . . . . . . . . . . . . . 64 6.3.3 Firmware description and measurement cycle . . . . . . . . . . . . . . . 66 6.3.4 Encapsulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 6.3.5 Fully assembled OPG device . . . . . . . . . . . . . . . . . . . . . . . . 67 6.4 Evaluation on the gesture recognition task . . . . . . . . . . . . . . . . . . . . . 69 6.4.1 Exercise selection, setup and recording . . . . . . . . . . . . . . . . . . . 69 6.4.2 Data corpus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 6.4.3 Sequence pre-processing . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 6.4.4 Choice of classifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 6.4.5 Training and evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 6.4.6 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 6.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 7 Improved device design with application in dysarthria therapy 7.1 Device design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 7.1.1 Design considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 7.1.2 General system overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 7.1.3 Intraoral sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 7.1.4 Receiver and controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 7.1.5 Multiplexer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 7.2 Hardware implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 7.2.1 Optode placement and circuit board layout . . . . . . . . . . . . . . . . 87 7.2.2 Encapsulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 7.3 Device characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 7.3.1 Photodiode transient response . . . . . . . . . . . . . . . . . . . . . . . 91 7.3.2 Current source and rise time . . . . . . . . . . . . . . . . . . . . . . . . 91 7.3.3 Multiplexer switching speed . . . . . . . . . . . . . . . . . . . . . . . . . 92 7.3.4 Measurement cycle and firmware implementation . . . . . . . . . . . . . 93 7.3.5 In vitro measurement accuracy . . . . . . . . . . . . . . . . . . . . . . . 95 7.3.6 Optode measurement stability . . . . . . . . . . . . . . . . . . . . . . . 96 7.4 Evaluation on the phoneme recognition task . . . . . . . . . . . . . . . . . . . . 98 7.4.1 Corpus selection and recording setup . . . . . . . . . . . . . . . . . . . . 98 7.4.2 Annotation and sensor data post-processing . . . . . . . . . . . . . . . . 98 7.4.3 Mapping from counts to millimeter . . . . . . . . . . . . . . . . . . . . . 99 7.4.4 Classifier and feature selection . . . . . . . . . . . . . . . . . . . . . . . 100 7.4.5 Evaluation paradigms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 7.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 7.5.1 Tongue distance curve prediction . . . . . . . . . . . . . . . . . . . . . . 105 7.5.2 Tongue contact patterns and contours . . . . . . . . . . . . . . . . . . . 105 7.5.3 Phoneme recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 7.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 8 Conclusion and future work 115 9 Appendix 9.1 Analytical light transport models . . . . . . . . . . . . . . . . . . . . . . . . . . 119 9.2 Meshed Monte Carlo method . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 9.3 Laser safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 9.4 Current source modulation voltage . . . . . . . . . . . . . . . . . . . . . . . . . 123 9.5 Transimpedance amplifier’s frequency responses . . . . . . . . . . . . . . . . . . 123 9.6 Initial OPG device’s PCB layout and circuit diagrams . . . . . . . . . . . . . . 127 9.7 Improved OPG device’s PCB layout and circuit diagrams . . . . . . . . . . . . 129 9.8 Test station layout drawing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 Bibliography 152 / Der Schlaganfall ist eine der häufigsten Ursachen für motorische Langzeitbehinderungen, einschließlich solcher im Mund- und Gesichtsbereich, deren Folgen u.a. Sprech- und Schluckprobleme beinhalten, welche sich in den beiden Symptomen Dysarthrie und Dysphagie äußern. In den letzten Jahrzehnten haben sich Rehabilitationsprogramme für die Behandlung von motorisch ausgeprägten Schlaganfallsymptomatiken substantiell weiterentwickelt. So liegt nicht mehr die reine Kompensation von verlorengegangener motorischer Funktionalität im Vordergrund, sondern deren aktive Wiederherstellung. Dabei hat u.a. die Verwendung von sogenanntem Biofeedback vermehrt Einzug in die Therapie erhalten, um Motivation, Engagement und Selbstwahrnehmung von ansonsten unbewussten Bewegungsabläufen seitens der Patienten zu fördern. Obwohl jedoch Sprech- und Schluckstörungen eine der häufigsten Folgen eines Schlaganfalls darstellen, wird diese Tatsache nicht von der aktuellen Entwicklung neuer Geräte und Messmethoden für quantitatives und umfassendes Biofeedback reflektiert, insbesondere nicht für die explizite Erfassung intraoraler Zungenkinematik und -kinetik und für den Anwendungsfall in der Schlaganfalltherapie. Ein möglicher Grund dafür liegt in den sehr strikten Anforderungen an ein solche Messmethode: Sie muss neben Portabilität idealerweise sowohl den Kontakt zwischen der Zunge und dem Gaumen, als auch die dreidimensionale Bewegung der Zunge in der Mundhöhle erfassen, ohne dabei die Artikulation selbst zu beeinflussen. Um diesen Anforderungen gerecht zu werden, wird in dieser Dissertation das Messprinzip der Optopalatographie untersucht, mit dem Schwerpunkt auf der Anwendung in der Dysarthrie- und Dysphagietherapie. Dies beinhaltet auch die Entwicklung eines entsprechenden Gerätes sowie dessen Befestigungsmethode in der Mundhöhle über ein dediziertes Mundschleimhautadhäsiv. Letzteres umgeht das bisherige Problem der notwendigen Anpassung eines solchen intraoralen Gerätes an einen einzelnen Nutzer. Für die Anwendung in der Dysphagietherapie erfolgte die Evaluation anhand einer automatischen Erkennung von Mobilisationsübungen der Zunge, welche routinemäßig in der funktionalen Dysphagietherapie durchgeführt werden. Für die Anwendung in der Dysarthrietherapie wurde eine Lauterkennung durchgeführt. Die Resultate bezüglich der Verwendung des Mundschleimhautadhäsives suggerieren, dass dieses tatsächlich eine valide Alternative zu den bisher verwendeten Techniken zur Befestigung intraoraler Geräte in der Mundhöhle darstellt. Zungenmobilisationsübungen wurden über Probanden hinweg mit einer Rate von 61 % erkannt, wogegen in der Lauterkennung Langvokale die höchste Erkennungsrate erzielten, gefolgt von Kurzvokalen und Konsonanten. Zusammenfassend lässt sich konstatieren, dass das Prinzip der Optopalatographie eine ernstzunehmende Option für die intraorale Erfassung von Zungenbewegungen darstellt, wobei weitere Entwicklungsschritte notwendig sind, welche im Ausblick zusammengefasst sind.:1 Introduction 1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Problem statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Goals and contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.4 Scope and limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2 Basics of post-stroke speech and swallowing therapy 2.1 Dysarthria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2 Dysphagia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3 Treatment rationale and potential of biofeedback . . . . . . . . . . . . . . . . . 13 2.4 Summary and conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3 Tongue motion sensing 3.1 Contact-based methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3.1.1 Electropalatography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3.1.2 Manometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.1.3 Capacitive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.2 Non-contact based methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.2.1 Electromagnetic articulography . . . . . . . . . . . . . . . . . . . . . . . 23 3.2.2 Permanent magnetic articulography . . . . . . . . . . . . . . . . . . . . 24 3.2.3 Optopalatography (related work) . . . . . . . . . . . . . . . . . . . . . . 25 3.3 Electro-optical stomatography . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.4 Extraoral sensing techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.5 Summary, comparison and conclusion . . . . . . . . . . . . . . . . . . . . . . . 29 4 Fundamentals of optopalatography 4.1 Important radiometric quantities . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.1.1 Solid angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.1.2 Radiant flux and radiant intensity . . . . . . . . . . . . . . . . . . . . . 33 4.1.3 Irradiance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.1.4 Radiance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.2 Sensing principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.2.1 Analytical models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 4.2.2 Monte Carlo ray tracing methods . . . . . . . . . . . . . . . . . . . . . . 37 4.2.3 Data-driven models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.2.4 Model comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.3 A priori device design consideration . . . . . . . . . . . . . . . . . . . . . . . . 41 4.3.1 Optoelectronic components . . . . . . . . . . . . . . . . . . . . . . . . . 41 4.3.2 Additional electrical components and requirements . . . . . . . . . . . . 43 4.3.3 Intraoral sensor layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 5 Intraoral device anchorage 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 5.1.1 Mucoadhesion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 5.1.2 Considerations for the palatal adhesive . . . . . . . . . . . . . . . . . . . 48 5.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 5.2.1 Polymer selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 5.2.2 Fabrication method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 5.2.3 Formulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.2.4 PEO tablets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.2.5 Connection to the intraoral sensor’s encapsulation . . . . . . . . . . . . 50 5.2.6 Formulation evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 5.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 5.3.1 Initial formulation evaluation . . . . . . . . . . . . . . . . . . . . . . . . 54 5.3.2 Final OPG adhesive formulation . . . . . . . . . . . . . . . . . . . . . . 56 5.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 6 Initial device design with application in dysphagia therapy 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 6.2 Optode and optical sensor selection . . . . . . . . . . . . . . . . . . . . . . . . . 60 6.2.1 Optode and optical sensor evaluation procedure . . . . . . . . . . . . . . 61 6.2.2 Selected optical sensor characterization . . . . . . . . . . . . . . . . . . 62 6.2.3 Mapping from counts to millimeter . . . . . . . . . . . . . . . . . . . . . 62 6.2.4 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 6.3 Device design and hardware implementation . . . . . . . . . . . . . . . . . . . . 64 6.3.1 Block diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 6.3.2 Optode placement and circuit board dimensions . . . . . . . . . . . . . 64 6.3.3 Firmware description and measurement cycle . . . . . . . . . . . . . . . 66 6.3.4 Encapsulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 6.3.5 Fully assembled OPG device . . . . . . . . . . . . . . . . . . . . . . . . 67 6.4 Evaluation on the gesture recognition task . . . . . . . . . . . . . . . . . . . . . 69 6.4.1 Exercise selection, setup and recording . . . . . . . . . . . . . . . . . . . 69 6.4.2 Data corpus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 6.4.3 Sequence pre-processing . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 6.4.4 Choice of classifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 6.4.5 Training and evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 6.4.6 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 6.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 7 Improved device design with application in dysarthria therapy 7.1 Device design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 7.1.1 Design considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 7.1.2 General system overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 7.1.3 Intraoral sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 7.1.4 Receiver and controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 7.1.5 Multiplexer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 7.2 Hardware implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 7.2.1 Optode placement and circuit board layout . . . . . . . . . . . . . . . . 87 7.2.2 Encapsulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 7.3 Device characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 7.3.1 Photodiode transient response . . . . . . . . . . . . . . . . . . . . . . . 91 7.3.2 Current source and rise time . . . . . . . . . . . . . . . . . . . . . . . . 91 7.3.3 Multiplexer switching speed . . . . . . . . . . . . . . . . . . . . . . . . . 92 7.3.4 Measurement cycle and firmware implementation . . . . . . . . . . . . . 93 7.3.5 In vitro measurement accuracy . . . . . . . . . . . . . . . . . . . . . . . 95 7.3.6 Optode measurement stability . . . . . . . . . . . . . . . . . . . . . . . 96 7.4 Evaluation on the phoneme recognition task . . . . . . . . . . . . . . . . . . . . 98 7.4.1 Corpus selection and recording setup . . . . . . . . . . . . . . . . . . . . 98 7.4.2 Annotation and sensor data post-processing . . . . . . . . . . . . . . . . 98 7.4.3 Mapping from counts to millimeter . . . . . . . . . . . . . . . . . . . . . 99 7.4.4 Classifier and feature selection . . . . . . . . . . . . . . . . . . . . . . . 100 7.4.5 Evaluation paradigms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 7.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 7.5.1 Tongue distance curve prediction . . . . . . . . . . . . . . . . . . . . . . 105 7.5.2 Tongue contact patterns and contours . . . . . . . . . . . . . . . . . . . 105 7.5.3 Phoneme recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 7.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 8 Conclusion and future work 115 9 Appendix 9.1 Analytical light transport models . . . . . . . . . . . . . . . . . . . . . . . . . . 119 9.2 Meshed Monte Carlo method . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 9.3 Laser safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 9.4 Current source modulation voltage . . . . . . . . . . . . . . . . . . . . . . . . . 123 9.5 Transimpedance amplifier’s frequency responses . . . . . . . . . . . . . . . . . . 123 9.6 Initial OPG device’s PCB layout and circuit diagrams . . . . . . . . . . . . . . 127 9.7 Improved OPG device’s PCB layout and circuit diagrams . . . . . . . . . . . . 129 9.8 Test station layout drawing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 Bibliography 152

Background. Many older stroke survivors live with multiple (> 2) chronic conditions (MCC), resulting in the need for care by multiple health and social service providers from multiple organizations and sectors. Managing the physical, social and psychological needs related to stroke in addition to other chronic conditions is a complex process that is best served by an interprofessional team of health care providers working collaboratively toward common goals. Interprofessional education (IPE) has been promoted by numerous organizations as a method to enhance collaborative practice. However, many home care providers have not received formal IPE or training to support collaborative practice. Providing IPE in the home care setting is challenging because providers rarely work in a common location, often work in isolation, and spend much of their time driving to provide care to clients in their homes. Moreover, the effectiveness of IPE on collaborative practice for stroke rehabilitation in the home care setting is undetermined. New approaches to IPE for practicing health care providers working in the home care setting are needed. The purpose of this study was to examine the feasibility and acceptability of implementing a new theory-based, IPE intervention, and to explore its effects on collaborative practice in home care for older adult stroke survivors with MCC. Method. This feasibility study involved the use of both a qualitative descriptive and a quantitative (one-group repeated measures) design. The IPE intervention was developed and evaluated within the context of a larger pragmatic randomized controlled trial (RCT), which evaluated the effectiveness of the Aging Community and Health Research Unit Community Partnership Program (ACHRU-CPP). Informed by the W(e) Learn Framework for Interprofessional Education, the National Interprofessional Competency framework, and the literature, the IPE intervention consisted of four key components: (a) an initial three-hour standardized IPE training session; (b) standardized training for care coordinators; (c) collaborative practice reflective huddles; and (d) outreach visits. The primary outcome was the feasibility of the IPE intervention (enrollment rate, attrition rate, implementation barriers/facilitators). Secondary outcomes included the acceptability of the IPE intervention, the feasibility of the study methods (recruitment/retention rates and procedures, eligibility criteria, data collection and analysis methods), and potential effectiveness of the intervention based on three-month changes in collaborative practice, as measured by the Collaborative Practice Assessment Tool (CPAT) and the 19-Item Team Climate Inventory (TCI). Feasibility and acceptability outcomes were based on descriptive statistics for enrollment and attrition rate and qualitative descriptive analysis of focus group content, field notes, and evaluation of training. The potential effectiveness of the IPE intervention was explored using paired t-tests and Cohen’s d, with the results expressed using descriptive statistics and effect estimates (95% confidence intervals). Results. A total of 37 home care providers from two provider agencies and one Community Care Access Centre (CCAC) in Ontario, Canada participated in the study. Participants included registered nurses, physiotherapists, occupational therapists, personal support workers, care coordinators as well as nursing, rehabilitation and personal support worker supervisors. Participants viewed the intervention as feasible and acceptable. It was effective in improving three domains of collaborative practice as measured by the CPAT (communication/information exchange; community linkage and coordination of care; decision-making and conflict management) and one domain of collaborative practice, as measured by the TCI (task orientation) at six months post initial training. Participants perceived many benefits to the intervention, including improved communication and collaboration within their teams, enhanced role understanding, increased learning with and from each other, and increased appreciation and valuing of the expertise of all team members. Facilitators to implementing the intervention included: funding from the larger trial, support from key stakeholders including agency leadership, provision of key resources (e.g., Team Charter, sample agenda), and continuity of the care coordinators. Barriers included unanticipated delays in recruitment of older adult stroke survivor participants into the larger trial, and higher than expected attrition rates. The study methods were feasible and effective in reaching the target population. We established that the intervention could be delivered as planned. Conclusion. The results of this study provide preliminary evidence for the feasibility, acceptability and preliminary effects of the IPE intervention on collaborative practice for an interprofessional stroke-specific team in home care caring for older adult stroke survivors with MCC. The results also provide knowledge of the facilitators and barriers to successfully implementing and sustaining the intervention into home care practice. Further research is warranted to test this intervention in other chronic populations and settings. / Thesis / Doctor of Philosophy (PhD) / This feasibility study used qualitative and quantitative methods to evaluate the implementation of a new theory-based, Interprofessional Education (IPE) intervention and explored its effects on collaborative practice in home care for older stroke survivors with multiple chronic conditions. The IPE intervention was developed and evaluated within the context of a larger pragmatic randomized controlled trial (RCT), which evaluated the effectiveness of the Aging Community and Health Research Unit Community Partnership Program. The six-month IPE intervention consisted of four key components: (a) an initial three-hour standardized IPE training session; (b) standardized training for care coordinators; (c) collaborative practice reflective huddles; and (d) outreach visits. Participants included 37 home care providers including registered nurses, physiotherapists, occupational therapists, personal support workers, care coordinators, and nursing, rehabilitation and personal support worker supervisors from two service provider agencies and one Community Care Access Centre (CCAC) in Ontario, Canada. The intervention was effective in improving collaborative practice (e.g., communication within teams, role understanding, team decision-making and conflict management). Facilitators to implementing the intervention included: funding from the larger trial, leadership support, provision of key resources, and continuity of the care coordinators. Barriers included unanticipated delays in recruitment of older adult stroke survivor participants into the larger trial, and higher than expected attrition rates. This study offers preliminary evidence that the intervention is feasible to deliver, acceptable to providers, and may improve collaboration within an interprofessional stroke-specific team. Further research is necessary to test this intervention in other chronic populations and settings.

No / Multi-disciplinary team members predict each patient’s rehabilitation potential to maximise best use of resources. A lack of underpinning theory about rehabilitation potential makes it difficult to apply this concept in clinical practice. This study theorises about rehabilitation potential drawing on everyday decision-making by Health Care Professionals (HCPs) working in stroke rehabilitation services. Methods: A clinical scenario, checked for face validity, was used in two focus groups to explore meaning and practice around rehabilitation potential. Participants were 12 HCPs working across the stroke pathway. Groups were co-facilitated, audio-recorded and fully transcribed. Analysis paid attention to data grounded in first-hand experience, convergence within and across groups and constructed a conceptual overview of HCPs’ judgements about rehabilitation potential. Results: Rehabilitation potential is predicted by observations of “carry-over” and functional gain and managed differently across recovery trajectories. HCPs’ responses to rehabilitation potential judgements include prioritising workload, working around the system and balancing optimism and realism. Impacts for patients are streaming of rehabilitation intensity, rationing access to rehabilitation and a shifting emphasis between management and active rehabilitation. For staff, the emotional burden of judging rehabilitation potential is significant. Current service organisation restricts opportunities for feedback on the accuracy of previous judgements. Conclusion: Patients should have the opportunity to demonstrate rehabilitation potential by participation in therapy. As therapy resources are limited and responses to therapy may be context-dependent, early decisions about a lack of potential should not limit longer-term opportunities for rehabilitation. Services should develop strategies to enhance the quality of judgements through feedback to HCPs of longer-term patient outcomes.Implications for Rehabilitation Rehabilitation potential is judged at the level of individual patients (rather than population-based predictive models of rehabilitation outcome), draws on different sources of often experiential knowledge, and may be less than reliable. Decisions about rehabilitation potential may have far reaching consequences for individual patients, including the withdrawal of active rehabilitation in hospital or in the community and eventual care placement. A better understanding of what people mean by rehabilitation potential by all team members, and by patients and carers, may improve the quality of joint decision making and communication.

No / We drew on an interdisciplinary research design to examine stroke survivors’ experiences of living with stroke and with technology in order to provide technology developers with insight into values, thoughts and feelings of the potential users of a to-be-designed robotic technology for home-based rehabilitation of the hand and wrist. Method: Ten stroke survivors and their family carers were purposefully selected. On the first home visit, they were introduced to cultural probe. On the second visit, the content of the probe packs were used as prompt to conduct one-to-one interviews with them. The data generated was analysed using thematic analysis. A third home visit was conducted to evaluate the early prototype. Results: User requirements were categorised into their network of relationships, their attitude towards technology, their skills, their goals and motivations. The user requirements were used to envision the requirements of the system including providing feedback on performance, motivational aspects and usability of the system. Participants’ views on the system requirements were obtained during a participatory evaluation. Conclusion: This study showed that prior to the development of technology, it is important to engage with potential users to identify user requirements and subsequently envision system requirements based on users’ views.