Abstract About one third of home-dwelling older people suffer a fall ech year. The most consuming falls occur when the person is alone and unable to get up, resulting in long lies which are associated with institutionalisation and high morbidity-mortality rate. Even though personal emergency response systems provide applications to call for help, older people are not always able or willing to activate them. Hence, an automatic fall detection system is an important setting. Even though pilot applications and commercial fall detection systems exist, the real-life validation of these systems is scant. The aim of this study was to develop a validated acceleration-based method for fall detection to be adapted for real-life applications among older people. Methods capable of discriminating between falls and activities of daily living (ADL) were determined based on laboratory tests. The threshold-based algorithms were validated with intentional falls in 20 middle-aged test persons and ADL in 20 middle-aged and 21 older people. The algorithm for the waist with impact and end posture detection was able to discriminate falls from ADL with 97% sensitivity and 100% specificity. In order to validate the fall detection system, a field test was performed with 16 residents in elderly care units wearing a wireless sensor. During the 6-month test period, acceleration data from five real-life falls were collected. One of the falls resulted in a hip fracture. These falls showed similar features as intentional falls. However, high pre impact velocity was detected in the case with a fracture, but not in all falls with preventative actions. The system had a fall detection sensitivity of 71.4% with a false alarm rate of 1.1 alarms over a 24-hour time period in this real-life pilot test. The data from real-life falls provide important material for further development of fall detection and studies on fall mechanism and fall prevention. / Tiivistelmä Kotona asuvista yli 65-vuotiaista kaatuu vuosittain kolmannes. Mikäli kaatunut ei kykene nousemaan omin neuvoin, avun saaminen saattaa viivästyä. Tämä suurentaa sekä laitoshoitoon joutumisen todennäköisyyttä että kuoleman riskiä. Erilaisia hälytysjärjestelmiä on kyllä saatavilla, mutta ikääntyneet eivät aina kykene käyttämään niitä tai eivät jostain syystä halua tehdä hälytystä. Tämän vuoksi automaattiselle kaatumishälyttimelle on tarvetta. Tässä tutkimuksessa kehitettiin ja testattiin ikääntyneiden tarpeisiin soveltuva kiihtyvyysanturiin perustuva kaatumisen tunnistumenetelmä. Aineisto koottiin laboratorio-olosuhteissa kokeilla, joihin osallistui sekä nuoria että keski-ikäisiä. Raja-arvoon perustuvia tunnistusalgoritmeja testattiin 20 keski-ikäisen ohjeistettujen testikaatumisten sekä 20 keski-ikäisen ja 21 ikääntyneen arkisten askareiden tuottamalla datalla. Kaatumistapahtuman impaktin ja loppuasennon tunnistaminen vyötäröltä mitatuista kiihtyvyysarvoista erotteli kaatumisen muusta liikkeestä 95 % sensitiivisyydellä ja 100 % spesifisyydellä. Tunnistusmenetelmää testattiin kenttäkokeessa, jossa 16 ikääntynyttä hoitokodin asukasta piti vyötäröllään mittauslaitetta. Kuuden kuukauden aikana kiihtyvyyssignaali saatiin viidestä kaatumisesta. Yksi niistä aiheutti lonkkamurtuman. Analyysin mukaan näiden todellisten kaatumisten kiihtyvyyssignaalit muistuttivat testikaatumisia. Lonkkamurtumatapauksessa ennen impaktia mitattu nopeus oli erittäin korkea. Vastaavaa ei havaittu tapauksissa, joissa oli merkkejä siitä, että kaatumista oli yritetty estää. Kenttäkokeessa kaatumishälytysjärjestelmän sensitiivisyys oli 71.4 % ja vääriä hälytyksiä oli 1.1 vuorokaudessa. Tutkimuksessa saatua tietoa tosielämän kaatumistapahtumista voidaan käyttää hyväksi kehitettäessä kaatumisten ehkäisyä, niiden mekanismin tutkimista sekä kaatumisen tunnistusta kiihtyvyysanturien avulla.

Magister Educationis - MEd / One of the key ideas in the research on mathematics education is that the mathematical knowledge that learners acquire is strongly tied to the particular situation in which it is learnt. This study investigated the contexts that learners in grades eight, nine and ten prefer to deal with in the learning of mathematics based on their personal, social, societal, cultural and contextual concerns or affinities. The study is situated in the large-scale project called the Relevance of School Mathematics Education II (ROSMEII), which is concerned with the application and the use of mathematical knowledge and processes in real life situations. It is based on a survey of learners from ten (10) secondary schools in the Oshana and Khomas regions of Namibia. The ten schools that were sampled represent the spectrum of schools in Namibia in both urban and semi-urban areas. The Rasch model of data analysis is employed to provide some insight into the contextual situations learners would like to deal with in their mathematical learning. The data obtained for this study was analysed using the WINSTEPS Version 3.65.0 suite of computer programs. The current study arises from a concern about the absence of the voices of learners in the contextual situations in mathematics selected by adults such as mathematics teachers, inspectorates and curriculum and materials developers. The assumption is that the inclusion of learners’ insights into mathematics curricular might enhance mathematical learning. The study reveals that school children have an intrinsic desire to learn about mathematical issues embedded in real-life contexts. Several items which Namibian learners have shown interest in are issues they experience in life out of school which are not directly dealt with in school, such as managing personal and financial affairs, health matters, technology, construction, engineering and government financial matters. However, learners indicated to have a low preference in contexts like lotteries and gambling, national and international politics, cultural products, all kinds of pop music and dancing. This thesis contends that the inclusion of contexts in the mathematics curriculum which are of interest to learners will go a long way in facilitating good performance of learners in mathematics. / South Africa

The use of non-invasive signal acquisition methods is today the standard for testing pattern recognition algorithms in prosthetic control. Such research had shown consecutively high performance on both prerecorded and real time data, yet when tested in real life they deteriorate. To investigate why, the author who is a congenital amputee, wore a prosthetic system utilizing pattern recognition control on a daily basis for a five-day period. The system generated one new classification every 50 ms and movement execution was made continuously; for classifying open/close; and by winning a majority vote; for classifying side grip, fine grip and pointer. System data was continuously collected and errors were registered through both a manual and an automatic log system. Calculations on extracted data show that grip classifications had an individual accuracy of 47%- 70% while open/close got 95%/98%, but if classified according to a majority vote, grips increased their accuracy to above 90% while open/close dropped to 80%. The conclusion was that majority vote might help complex classifications, like fine grips, while simpler proportional movements is exacerbated by majority voting. Major error sources were identified as signal similarities, electrode displacements and socket design. After the daily monitoring ended the systems functionality was tested using the "Assessment of Capacity for Myoelectric Control". The ACMC results showed that the system has similar functionality to commercial threshold control and thus is a possible viable option for both acquired and congenital amputees. / Användningen av icke-invasiva signalavläsningsmetoder är för nuvarande standarden inom utvärderingar av mönsterigenkännings-algoritmer för proteskontroll. Forskning inom området har konsekvent visat på hög prestanda för både ansamlat och realtids data, men när algoritmerna testas i verkliga livet fungerar de ej väl. För att undersöka varför har författaren, som har en kongenital amputation, burit en protes vilken använder mönsterigenkänningskontroll i sitt vardagliga liv under en femdagars period. Systemet genererade en ny klassificering var 50 ms och rörelse-utförande skedde antingen kontinuerligt; för öppna/stäng; eller genom att vinna en majoritetsröstning, för att klassificera greppen sidogrepp, fingrepp samt peka. Data insamlades kontinuerligt och felklassificeringar registrerades genom både ett manuellt och ett automatiskt markeringssystem. Beräkningar på insamlade data visade att för grepp låg den individuella träffsäkerheten på 47%- 70% medan öppna/stäng var 95%/98%, men om data grupperades ökade träffsäkerheten för greppen till 90% medan för öppna/stäng minskade den till 80%. Slutsatsen blev då att majoritetsröstning hjälper mer komplexa rörelser som grepp, men är hindrande för mer väldefinierade proportionella öppna/stäng rörelser. De största felkällorna identifierades som likheter i signaler, elektrodavbrytningar och design av proteshylsan. Efter slutförd daglig övervakning undersöktes funktionaliteten hos systemet med hjälp av funktionalitetstestet "Assessment of Capacity for Myoelectric Control". ACMC testen visade att systemet hade liknande funktionalitet som kommersiell tröskelkontrol och därmed kan ses som ett möjligt alternativ för kontroll, både hos dem med förvärvade och kongenitala amputationer.

Background: The objective of this study is to determine the difference in customer attitudes towards CGI advertisements pictured in real-life scenes and traditional digital advertising. Furthermore, this study aims to discover how customer attitudes towards CGI advertisements pictured in real-life scenes influence brand experience. Method: The study developed 6 hypotheses examining customer attitudes towards advertisements. These hypotheses focused on 5 factors: credibility, irritation, entertainment, creativity, and informativeness. How customers perceive these factors and their attitudes towards CGI advertisements pictured in real-life scenes compared to traditional digital advertisements were examined. One of the hypotheses further addressed the question of whether attitudes towards CGI advertisements pictured in real-life scenes influence customers' brand experience. A questionnaire was used to test the hypotheses and answered by 139 respondents. Findings: The result of the study showed that customers perceive the factors of credibility andinformativeness more positively in traditional digital advertisements. In contrast, factors such as creativity, entertainment, and lack of irritation were evaluated more favorably for CGI advertisements pictured in real-life scenes. The study also discovered that positive attitudes toward CGI advertisements pictured in real-life scenes lead to more positive brand experiences. Conclusion: Age groups between 18 and 55 prefer advertising that is entertaining, creative, and innovative. CGI advertising in real scenes should not stand on its own but rather serve to capture the attention of the customer. This can help companies to become more visible and engage the target audience, which can then lead to a positive brand experience. This is due to its attractiveness and perception as an engaging, attention-grabbing, and more brand-recalling advertisement compared to traditional digital advertising. However, it is important to note that CGI advertising should not serve as themainsource of information about a product or service, but rather as a tool to grab attention.

This research sought to examine the lived experiences and minds of entrepreneurs who had influenced society in ways beyond the domain of the private sector. An entrepreneur was defined as an individual who pursues opportunity or is driven by the perception of opportunity. It was proffered that a divide exists in our understanding of an entrepreneur—as person—and an entrepreneur—as business starter. To a great extent, this divide has limited exploration of the connections and benefits that entrepreneurial thinking might bring to education. This was a qualitative and interpretivist study based on interviews with a purposive sample of entrepreneurs who have had a major influence in private, public, and more generally in the plural sectors in and around Canada. Nine entrepreneurs were invited to tell their life stories, to provide their perceptions of how entrepreneurs think and to relate how they believed they had developed their skills. Using in-depth interviews, the researcher was able to discover, in part, the essence of how these entrepreneurs thought, how they made decisions, and how they viewed themselves and their world. The findings and insights add to the growing body of literature on entrepreneurship. The researcher argued that the dissertation had given light to the lived experience of entrepreneurs. Understanding how these persons had become entrepreneurial and how they had recognized opportunities for their own mentorship were addressed by the research. How these entrepreneurs thought, their motivations, and interests, as well as how they perceived their role in the world provided valuable insights, especially as one seeks to understand how to cultivate or develop, nurture and encourage entrepreneurs, and to foster entrepreneurial behaviour throughout the education sector. The exploratory insights from these rich interviews serve to further foster fruitful conversations on potential alignments between the ways and thinking of entrepreneurs and educational systems.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Notre recherche étudie la pratique des professeurs des écoles débutants à partir d'une approche par la description de leur vécu, de leur point de vue, c'est-à-dire en première personne. La méthode de recueil des données choisie est l'entretien d'explicitation de Vermersch, dans le but de mieux comprendre comment se construit le « je » professionnel chez les professeurs des écoles débutants. Cet accompagnement amène la personne interviewée à recontacter, au plus près, son vécu passé. Les verbalisations ainsi recueillies concernent, selon notre choix, des descriptions de situations de classe. À partir de l'analyse de ces données, nous avons mis en évidence des gestes professionnels d'ajustement (Jorro) des professeurs des écoles débutantes qui sont le résultat de prises d'information/prises de décision très brèves en situation. Les principaux indices qui attirent l'attention des professeurs des écoles débutantes sont relatifs aux élèves : leurs mouvements, leurs paroles, leur expressivité (le regard, l'absence de réaction, les sourires, etc.). Nous avons mis en évidence des compétences émergentes chez les professeurs des écoles débutantes, qui relèvent de compétences transversales. Pour nous, une compétence émergente est un savoir-faire professionnel mû par une intention, un sous-but, qui est en cours de construction chez le professeurs des écoles, qui, souvent n'est pas encore conscientisé, mais qui est présent dans la description en première personne d'un vécu de classe. Ces savoir-faire appartiennent au préréfléchi de l'action et vont donner lieu à des prises de conscience par le processus cognitif de réfléchissement. Les compétences émergentes que nous avons pu identifier sont au nombre de quatre : faire verbaliser les élèves, favoriser les interactions entre les élèves, écouter et prendre en compte la parole des élèves, utiliser son corps et son positionnement dans la classe. Deux catégories de compétences émergentes se dégagent : l'importance du corps de l'enseignant par les professeurs des écoles débutantes, d'une part, et l'importance de la prise en compte de la parole des élèves, d'autre part. Ces deux objets de prise de conscience semblent fondamentaux dans la conduite de la classe, pour instaurer un climat de confiance et pour assurer une relation pédagogique propice à l'apprentissage des élèves. Notre recherche met en évidence des compétences conscientisées par les professeurs des écoles débutantes. Elles sont déjà présentes à la conscience des professeurs des écoles débutantes interviewées, ou ont fait l'objet d'une prise de conscience au cours de notre recherche. Elles sont constituées de ce que la professeurs des écoles sait qu'elle sait faire. Ces compétences conscientisées font partie des gestes professionnels (Jorro). Ces gestes professionnels contiennent les savoir-faire historiques du métier et intègrent des compétences que le professionnel ajuste selon ses perceptions et les relations inter-subjectives, en situation. Nous avons montré que les compétences émergentes et les compétences conscientisées, constituant des gestes professionnels, étaient sous-tendues par des valeurs pour les professeurs des écoles débutantes de notre corpus. Ces valeurs sont : créer un climat de confiance dans la classe, de donner confiance en eux aux élèves, d'aider les élèves en difficulté et de rendre les élèves autonomes. Ces valeurs sous-jacentes sont incarnées par les professeurs des écoles débutantes à travers une posture d'écoute et de respect des élèves, sachant manier aussi l'humour, une posture d'accompagnement des élèves (dans le sens d'être « à côté », de « rassurer » et d'« encourager »). Ces valeurs représentent les deux piliers du « je » professionnel en construction chez les professeurs des écoles débutantes : l'importance de la relation aux élèves et l'importance du lâcher-prise. Nous formulons l'hypothèse que ce « je » professionnel évolue tout au long de la carrière des professeurs des écoles. / Our research studies the practice of the primary school teachers beginners starting from an approach by description from their real-life experience, their point of view, i.e. in first person. The selected method of collection of the data is the elicitation interview method of Vermersch, with an aim of better understanding how the professional “I” of the primary school teachers beginners in construction. This accompaniment brings the person interviewed to recontact, with close, its past real life experience. The entries of charge thus collected concern, according to our choice, descriptions of situations of class. Starting from the analysis of these data, we highlighted professional gestures of adjustment (Jorro) of the primary school teachers beginners which are the result of acquisitions of information and very short decision makings in situation. The principal indices which draw the attention of the primary school teachers beginners are relating to the pupils: their movements, their words, their expressivity (the glance, the absence of reaction, smiles, etc.). We highlighted emergent competences of the primary school teachers beginners, which concern transverse competences. For us, an emergent competence is a professional know-how drived by an intention, a under-goal, which is in construction at the primary school teachers beginners, which, often is not conscientized yet, but which is present in description in first person of the real life experience on class. This know-how belongs to the given cognitive process of the action and will cause awakenings by the cognitive process of reflecting. Emergent competences that we could identify are four : to make the pupils to verbalize, support the interactions between the pupils, to listen to and to take into account the word of the pupils, to use teacher's body and its positioning in the classroom. Two categories of emergent competences appear: importance of the body of the teacher by the primary school teachers beginners, on the one hand, and the importance of the taking into account of the word of the pupils, on the other hand. These two objects of awakening seem fundamental in the control of the class, to found a climate of trust and to ensure a teaching relation favourable with the training of the pupils. Our research highlights the competences conscientized by the beginners primary school teachers. They are already present at the conscience of the primary school interviewed teachers, or were the object of an awakening during our research. They consist of what the primary school teachers beginners know that they can do. These conscientized competences are part of the professional gestures (Jorro). These occupational gestures contain historical know-how of the trade and integrate competences which the professional adjusts according to his perceptions and the intersubjective relations, in situation. We showed that emergent competences and conscientized competences, constituting professional gestures, were underlain by values for the primary school teachers beginners of our corpus. These values are : to create a climate of trust in the class, to give confidence in them to the pupils, to help the pupils in difficulty and to make the pupils autonomous. These subjacent values are incarnated by the primary school teachers beginners through a posture of listening and respect of the pupils, knowing to also handle humour, a posture of accompaniment of the pupils (in the direction to be “at side”, “to reassure” and “to encourage”). These values represent the two pillars of the professional “I” in construction of the primary school teachers beginners: importance of the relation to the pupils and importance of the release-catch. We formulate the assumption that this “I” professional evolves throughout the career of the primary school teachers.

The presented thesis discusses bilevel programming problems with the focus on solution algorithms. Bilevel programming problem is a hierarchical programming problem, where constraints contain another programming problem. We formulate basic bilevel optimization theory and describe three types of so- lution algorithms for bilevel programming problems: Algorithms based on KKT reformulation where the lower level is replaced by its KKT conditions, algorithms based on optimal value function where the bilevel programming problem is re- duced to a single level problem using the optimal value function of the lower level problem, and algorithms solving linear bilevel programming problems. Using real data for portfolio optimization bilevel programming problems, we compare ability to solve the problems and computing time of some of the pre- sented algorithms. 1

INTRODUÇÃO: No Brasil, no período compreendido entre 2013 a 2015, a terapia tripla com boceprevir (BOC) ou telaprevir (TVR) foi o tratamento padrão para pacientes infectados pelo vírus da hepatite C (VHC) genótipo 1. OBJETIVOS: Avaliar a efetividade e a segurança da terapia tripla com BOC ou TVR, no contexto de vida real, em pacientes portadores de hepatite C crônica genótipo 1 tratados em centros de referência no Brasil. Foi também objetivo deste estudo, investigar os fatores preditores para obtenção da resposta viral sustentada (RVS) e fatores preditores da ocorrência de eventos adversos sérios (EAS) associados a essa modalidade terapêutica. MÉTODOS: Trata-se de um estudo multicêntrico, observacional e retrospectivo, que incluiu pacientes monoinfectados pelo VHC genótipo 1 que iniciaram o tratamento com interferon peguilado (Peg-IFN), ribavirina e BOC ou TVR . Dados demográficos, clínicos, virológicos e eventos adversos (EA) foram coletados durante o tratamento e seguimento. A análise dos fatores preditores da RVS e de EAS foi realizada por meio do modelo de regressão de Poisson com variância robusta. RESULTADOS: Dos 715 pacientes analisados, 56,1% eram do sexo masculino, com média de idade de 54,1 ± 10,1 anos, 59% eram portadores de cirrose hepática e 67,1% tinham sido tratados previamente. Foram tratados com TVR, 557 pacientes (77,9%), e com BOC, 158 (22,1%). Pela análise de intenção de tratamento, a RVS geral foi de 56,6% (IC 95%, 52,9 - 60,3), com uma efetividade semelhante em ambos os grupos (51,9% [BOC] versus (vs.) 58% [TVR], P = 0,190). Os pacientes cirróticos tiveram uma menor taxa de RVS comparado aos não cirróticos (46,9% vs. 70,6%, P < 0,001). Na análise múltipla, a obtenção de RVS foi associada com a ausência de cirrose hepática (P < 0,001), antecedente de recidiva viral após tratamento prévio (P < 0,001), contagem de plaquetas no pré-tratamento acima de 100.000/mm3 (P < 0,001) e obtenção de resposta viral rápida (P < 0,001). A taxa de descontinuação do tratamento foi de 34,8%, sendo a ocorrência de EA (16,1%) e a falha virológica (15,9%) as principais causas. Em relação à segurança do tratamento, 90,1% dos pacientes (IC 95%, 87,6 - 92,2) apresentaram algum EA durante o tratamento, sendo que 44,2% dos pacientes (IC 95%, 40,5 - 47,9) apresentaram EAS. A anemia foi o EA mais comum em ambos os grupos (59,5% [BOC] vs. 74,5% [TVR], P < 0,001). Houve maior ocorrência de EAS nos pacientes cirróticos comparados aos não cirróticos (50,7% vs. 34,8%, P < 0,001). Sexo feminino (P < 0,001), idade acima de 65 anos (P = 0,008), diagnóstico de cirrose hepática (P = 0,019), concentração de hemoglobina pré-tratamento alterada (P < 0,001) e contagem de plaquetas abaixo de 100.000/mm3 no pré-tratamento (P < 0,001) foram associados à ocorrência de EAS. Seis óbitos (0,8%) ocorreram. CONCLUSÃO: A terapia tripla com BOC ou TVR apresentou uma taxa de RVS superior ao tratamento padrão anterior (Peg-IFN e ribavirina). Contudo, este tratamento apresentou uma alta taxa de EAS, principalmente em pacientes com doença hepática avançada / INTRODUCTION: In Brazil, from 2013 until 2015, triple therapy with boceprevir (BOC) or telaprevir (TVR) was the standard treatment for patients infected with hepatitis C virus (HCV) genotype 1. OBJECTIVES: To evaluate the effectiveness and safety of triple therapy with BOC or TVR, in real life context, in patients with chronic hepatitis C genotype 1 treated at reference centers in Brazil. It was also the objective of this study to investigate the predictive factors for obtaining sustained viral response (SVR) and factors predictive of the occurrence of serious adverse events (SAE) associated with this therapeutic modality. METHODS: This was a multicenter, observational, retrospective study that included HCV infected patients genotype 1 who started treatment with pegylated interferon, ribavirin and BOC or TVR from July 2013 until April 2014, from 15 centers of reference in Brazil. Demographic, clinical, virological, and adverse events (AE) data were collected during treatment and follow-up. The analysis of predictive factors of SVR and SAE was performed using the Poisson regression model with robust variance. RESULTS: Of the 715 patients analyzed, 56.1% were males, mean age was 54.1 ± 10.1 years, 59% had hepatic cirrhosis and 67.1% had been previously treated. They were treated with TVR, 557 patients (77.9%), and with BOC, 158 (22.1%). By intention-to-treat analysis, overall SVR was 56.6% (95%CI, 52.9 - 60.3), with similar effectiveness in both groups (51.9% [BOC] versus (vs.) 58% [TVR], P = 0.190). Cirrhotic patients had a lower SVR rate compared to non-cirrhotic patients (46.9% vs. 70.6%, P < 0.001). In the multiple analysis, SVR acquisition was associated with absence of hepatic cirrhosis (P < 0.001), previous viral relapse after previous treatment (P < 0.001), pre-treatment platelet count above 100,000/mm3 (P < 0.001) and rapid viral response (P < 0.001). The treatment discontinuation rate was 34.8%, with the occurrence of AE (16.1%) and virological failure (15.9%) being the main causes. Regarding treatment safety, 90.1% of the patients (95%CI, 87.6 - 92.2) presented some AE during treatment, and 44.2% of the patients (95%CI, 40.5 - 47,9) presented SAE. Anemia was the most common AE in both groups (59.5% [BOC] vs. 74.5% [TVR], P < 0.001). There was a higher occurrence of SAE in cirrhotic patients compared to non-cirrhotic patients (50.7% vs. 34.8%, P < 0.001). Female gender (P < 0.001), age above 65 years (P = 0.008), diagnosis of hepatic cirrhosis (P = 0.019), altered pre-treatment hemoglobin concentration (P < 0.001) and platelet count below 100,000/mm3 in the pre-treatment (P < 0.001) were associated with the occurrence of SAE. Six deaths (0.8%) occurred. CONCLUSION: Triple therapy with BOC or TVR showed a higher SVR rate than the previous standard treatment (Peg-IFN and ribavirin). However, this treatment had a high EAS rate, especially in patients with advanced liver disease

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal