A dose prediction model for treatment planning was generated using U-Net architecture. The model was generated for advanced stage cancer patients. The U- Net architecture was created with depth=6 and kernel=6. The model architecture was successful to reduce the input image size (192X192) to feature map (6X6) which helped to extract the low level features. The dose prediction of the model was trained with depth=6, kernel=6, MSE loss, Adam optimizer, 1000 epochs and a batch size of 4. The predicted dose was rescaled for gamma analysis to quantify accuracy of the model. The renormalized predicted dose was quantified using gamma analysis with a 3mm, 3% dose tolerance. The gamma map was generated to visualize the regions where dose distributions failed. The gamma percentage values obtained on the training set were acceptable. The mean and standard deviation values of gamma pass percentage obtained on training dataset were 97.5% and 1.24% respectively, which concluded that training process was successful and was an almost perfect match of true dose and predicted dose. However, gamma pass percentage values obtained on validation set was not a good representation of the true dose. Nevertheless, the validation dataset was able to predict the approximate highest dose region. A gamma analysis with a 5mm, 5% dose tolerance was performed to test the the level of discrepancy between the predicted and true dose in the validation set. This increased the gamma pass percentage compared to the 3mm, 3% analysis to a mean gamma pass percentage of 26.2 ± 7.47%. Although the predicted dose was not of sufficient accuracy for clinical use, there technique studied in this work show promise for further development. / Thesis / Master of Science (MSc)

The emergence of Online Social Networking service which have introduced new paradigms of sharing have changed the way people interact with each other. They provide an easy and creative way to connect with people. As a result there are all kinds of people like working professionals, students, businessmen and many more who participate actively in such networks. Social Networking services are also turning out to be a medium for promoting small scale business, books, events etc. Apart from this there is substantial literature which suggests that social connections can be created, maintained and strengthened through computer mediated channels of communication with special importance being given to the role of internet in maintaining social relations. Thus the social networking services build digital connections among people who are socially connected. Though the users can leverage a host of applications/services like messaging, chatting, file sharing, discussions in groups, activities etc from these online social networking applications, there is no relevance given to the context/location of a user in the existing setup. With the technology moving from personal computers to hand held and intelligent devices with embedded technology and connectivity there is a rise in the development of pervasive computing applications which provide easy access to required information and applications through a class of ubiquitous equipment that operate when and where needed. This adds value to the existing infrastructure by providing services which are “anytime and anywhere”. Social networking applications hold a vast amount of personal data that can be used to improve the experience of users in the physical space. With the changing environment of users and also that of the people in their social network a system with a capability of examining the environment and providing highly personalized services to the user is highly desirable. Thus the work described in this document will project a social networking service that gives significance to the location of a user. In other words we will see how social networking services can migrate from networked personal computers to more ubiquitous hand held devices like PDA’s, Smart Phones which would involve the location information of the physical space in order to make social experiences more meaningful. The primary goals of this work are:- 1.To develop a social network service and deploy it to a select set of users and monitor the level of interaction. 2.Propose an architecture which could be used to implement location awareness in the social network service developed. 3.Develop a prototype and provide a “proof of concept” of the architecture and the system designed. 4.Survey privacy issues that are of great importance in such systems. This work will produce a framework and perception of designing a social networking service that is location aware. The resulting system will be deployed to a set of users’ in order to understand the usage pattern of such systems. This would give a good picture of how such systems would be perceived and compared to the existing online social network services that are insensitive to the change in location of its users.

The goal of this work has been to create an interactive visualization of weather data for the company Biometria. The weather data has been collected over a period of 2 years using a measuring device placed in a forest cabin. The forest cabin has been moved to different logging sites, and continuously sent measurement results. To understand the structure and organization of the data that would be visualized, a preliminary study and review have been conducted. Furthermore, a plan for the setup and a preliminary design of a prototype have been created. The preliminary work has formed the basis for the website developed using the ASP .NET Core framework. A login system has been developed using ASP .NET Core Identity and implemented on the website. The interactivity and visualization have been constructed by generating markers on a map when selecting a range between two dates. When clicking on a marker, a box with information such as date, temperature, and humidity is displayed. The position of the marker and its associated information are retrieved through calls to an external Application Programming Interface (API). The markers are displayed in different colors depending on the measured temperature. The construction has been developed in the Visual Studio development environment. / Målet med detta arbete har varit att skapa en interaktiv visualisering av väderdata åt företaget Biometria. Väderdata har samlats in under 2 års tid via en mätenhet som suttit på en skogskoja. Skogskojan har flyttats runt till olika avverkningsplatser och kontinuerligt skickat mätresultat. För att få förståelse om uppbyggnaden och strukturen av det data som skulle visualiseras, har en förstudie och granskning utförts. Vidare har en planering av upplägget och ett förarbete med formgivning av en prototyp skapats. Förarbetet har legat till grund för den webbplats som utvecklats med ramverket ASP .NET Core. Ett inloggningssystem har tagits fram med hjälp utav ASP .NET Core Identity och implementerats i webbplatsen. Interaktiviteten och visualiseringen har konstruerats genom att det vid val av ett spann mellan två datum, genereras markeringar på en karta. Vid klick på markeringen visas en ruta med information datum, temperatur och luftfuktighet. Positionen för markeringen och tillhörande information hämtas via anrop till ett externt API (Application Programming Interface). Markeringarna visas med olika färger beroende på mätenhetens uppmätta temperatur. Konstruktionen har tagits fram i utvecklingsmiljön Visual Studio.

This paper is concerned with a study of approximation order and construction of locally supported elements for the space S41 (Δ) of Cl quartic pp (piecewise polynomial) functions on a triangulation Δ of a connected polygonal domain Ω in R2. It is well known that, when Δ is a three-directional mesh Δ(1), the order of approximation of S41(Δ(1)) is only 4, not 5. Though a local Clough-Tocher refinement procedure of an arbitrary triangulation A yields the optimal (fifth) order of approximation from the space S41(Δ) (see [1]), it needs more data points in addition to the vertex set of the triangulation A. In this paper, we will introduce a particular mixed three-directional mesh Δ(3)) and construct so-called mixed three-directional elements. We prove that the space S41(Δ(3)) achieves its optimal-order of approximation by constructing an interpolation scheme using mixed three-directional elements.

<p> A simple model, expressed in terms of cloud amounts and heights, and air temperature (recorded hourly at Ft. St. John, B.C.) and daily radiosonde records (from St. Nelson, B.C.), is used in an attempt to approximate incoming solar and net radiation values at a nearby pasture site, where hourly measurements were recorded. Results from the model indicate that measured values on a daily basis were estimated within 20%, while five and ten day running means were calculated within 10% of the measured means. </p> / Thesis / Bachelor of Arts (BA)

Petri nets are a graphical construction with clearly de ned semantics which can model concurrent communicating systems in a formal manner similar to the way that automata theory can model formal language theory(Petri, 1962). As Dr. Carl Petri found the existing automata insu cient or too cumbersome for describing communicating systems others have found Petri Nets to be too cumbersome for e ectively reasoning about sophisticated, real world systems. In some cases these di culties were overcome by extending the theory of Petri Nets. Dr. Kurt Jensen developed the theory of Coloured Petri Nets (Jensen, 1981) for the purpose of generalizing and simplifying complex Petri Net models. This work incorporates Coloured Petri Nets and other theoretical extensions to describe a real world automated steel cutting system. During the course of this investigation the paper will formalize colours in the language of algebras and examine patterns related to timing conditions. / Thesis / Master of Science (MSc) / Petri nets can provide a graphical explanation of computer systems that have sophisticated communications. The graph has a precise mathematical meaning which allows it to be formally analyzed to prove many interesting properites of the net. There have been many extensions of the theory, some of which are incorporated to this model of an automated steel cutting machine. This thesis also presents the idea of colours, or data types, in the language of algebras.

‘Nature-positive’ (NP) is emerging as an increasingly used term, intended to encompass goals aimed at reversing the global decline in biodiversity and the incessant destruction of ecosystems. The 1.5-degree goal of the Paris agreement has become a unifying target in the climate discourse, accompanied by the guiding concept of ‘net-zero’ emissions. Anequivalent, widely accepted target for biodiversity has been called for, with NP being named a contender to take this position. However, the concept NP, and net-zero alike, have been subject to criticism. Building on an extensive literature search, this study seeks to identify and analyse the most frequently employed definitions of NP, and who promotes them. It also aims to draw conclusions from a comparative analysis where definitions of NP are mapped onto the IPBES- promoted Nature Futures Framework (NFF). For this purpose, an assessment tool (‘the NFF Mixing Triangle’) is presented. The study finds that NP is frequently used without specifying its meaning, in scientific and non-scientific literature alike. While existing dominant definitions are largely aligned on the conceptual level, there is a lack of consensus on a common definition among stakeholders. The Global Goal for Nature (GG4N) stands out as the most referenced source for defining NP, and is found to have the strongest alignment with NFF. While problematising and discussing criticism, the study suggests a number of focus areas that could serve the future potential of NP. This includes converging NP definitions, balancing definitions across the NFF value dimensions, and considering applicability across multiple scales. An aspiration to consolidate guiding frameworks, use of the Mitigation Hierarchy to strengthen the robustness of NP, and focus on stewardship inside corporations as well as across sectors and value chains, is proposed. Meanwhile, the study suggests that the external drivers of legislation and regulation, and the current growth paradigm and consumption patterns, need to be addressed in concert if NP is to have a real contribution to sustainability transformation.

Scoliosis is a common skeletal condition in which a curvature forms along the coronal plane of the spine. Although scoliosis has been long recognized, its pathophysiology and best mode of treatment are still debated. Currently, definitive diagnosis of scoliosis and its progression are performed through anterior-posterior (AP) radiographs by measuring the angle of coronal curvature, referred to as Cobb angle. Cobb angle measurements can be performed by Deep Learning algorithms and are currently being investigated as a possible diagnostic tool for clinicians. This thesis focuses on the role of Deep Learning in the diagnosis and treatment of Scoliosis and proposes a study design using the algorithms to continue to better understand and classify the disease.

The Portable Document Format (PDF) is an ideal format for viewing and printing documents. Today many companies store their documents in a PDF format. However, the conversion from a PDF document to any other structured format is inherently difficult. As a result, a lot of the information contained in a PDF document is not directly accessible - this is problematic. Manual intervention is required to accurately convert a PDF into another file format - this can be deemed as both strenuous and exhaustive work. An automated solution to this process could greatly improve the accessibility to information in many companies. A significant amount of literature has investigated the process of extracting information from PDF documents in a structured way. In recent years these methodologies have become heavily dependent on computer vision. The work on this paper evaluates how the U-Net model handles multi-label segmentation on PDF documents in a medical context - extending on Stahl et al.’s work in 2018. Furthermore, it compares two newer extensions of the U-Net model, MultiResUNet (2019) and SS-U-Net (2021). Additionally, it assesses how each of the models performs in a data-sparse environment. The three models were implemented, trained, and then evaluated. Their performance was measured using the Dice coefficient, Jaccard coefficient, and percentage similarity. Furthermore, visual inspection was also used to analyze how the models performed from a perceptual standpoint. The results indicate that both the U-Net and the SS-U-Net are exceptional at segmenting PDF documents effectively in a data abundant environment. However, the SS-U-Net outperformed both the U-Net and the MultiResUNet in the data-sparse environment. Furthermore, the MultiResUNet significantly underperformed in comparison to both the U-Net and SS-U-Net models in both environments. The impressive results achieved by the U-Net and SS-U-Net models suggest that it can be combined with a larger system. This proposed system allows for accurate and structured extraction of information from PDF documents. / Portable Document Format (PDF) är ett välfungerande format för visning och utskrift av dokument. I dagsläget väljer många företag därmed att lagra sina dokument i PDF-format. Konvertering från PDF format till någon annan typ av strukturerat format är dock svårt, och detta resulterar i att mycket av informationen i PDF-dokumenten är svårtillgängligt, vilket är problematiskt för de företag som arbetar med detta filformat. Det krävs manuellt arbete för att konvertera en PDF till ett annat filformat - detta kan betraktas som både ansträngande och uttömmande arbete. En automatiserad lösning på denna process skulle kunna förbättra tillgängligheten av information för många företag. En stor mängd litteratur har undersökt processen att extrahera information från PDF-dokument på ett strukturerat sätt. På senare tid har dessa metoder blivit starkt beroende av datorseende. Den här forskningen utvärderar hur U-Net-modellen hanterar segmentering av PDF dokument, baserat på flerfaldiga etiketter, i ett medicinskt sammanhang. Arbetet är en utökning av Stahl et al. forskning från 2018. Dessutom jämförs två nyare utökade varianter av U-Net-modellen , MultiResUNet (2019) och SS-U-Net (2021). Utöver detta så utvärderas även varje modell utefter hur den presterar i en gles datamiljö. De tre modellerna implementerades, utbildades och utvärderades. Deras prestanda mättes med hjälp av Dice-koefficienten, Jaccard-koefficienten och procentuell likhet. Vidare så görs även en visuell inspektion för att analysera hur modellerna presterar utifrån en perceptuell synvinkel. Resultaten tyder på att både U-Net och SS-U-Net är exceptionella när det gäller att segmentera PDF-dokument i en riklig datamiljö. SS-U-Net överträffade emellertid både U-Net och MultiResUNet i den glesa datamiljön. Dessutom underpresterade MultiResUNet signifikant i jämförelse med både U-Net och SS-U-Net modellen i båda miljöerna. De imponerande resultaten som uppnåtts av modellerna U-Net och SS-U-Net tyder på att de kan kombineras med ett större system. Detta föreslagna systemet möjliggör korrekt och strukturerad extrahering av information från PDF-dokument.

Previous academic literature examining the performance of private equity funds has documented that the average private equity fund has failed to outperform public equity markets. This underperformance coupled with a greater risk-return trade-off has failed to discourage investment in private equity markets. In fact, private equity firms have enjoyed record amounts of fund raising over the past several years. This phenomenon has been characterized as a puzzle and its investigation within the academic literature has only just begun. Using a unique and current data set covering private equity returns and their underlying cash flows, we examine performance measurements of private equity funds in the context of their relationships with one another and with public markets; as well as, examine the characteristics of the funds and their managers that drive these relationships. Our findings suggest that private equity investors are partially motivated by misinterpreted performance measurements and that this misinterpretation is compounded by fund reported residual values. These findings have important policy implications for both private equity fund managers and investors.