This project follows a Research through Design approach and through autobiographical design explores the question of how Interaction Design can support Lipoedema patients, by helping them cope with their body image in everyday life. Building on the concept of self-management for people with chronic medical conditions and a conscious connection between body and mind, I created a set of artefacts. The set of artefacts was connected to specific a interaction for each artefact, giving them a more meaningful purpose. Through living with these three artefacts, I was able to identify tensions revolving around themes of self-acceptance, discomfort and vulnerability. These are important when designing, not only for people diagnosed with Lipoedema, but also for any user group struggling with similar issues, such as body image. / Detta projekt följer ett forskning-genom-design tillvägagångssätt och genom autobiografisk design utforskas frågan om hur interaktionsdesign kan stötta Lipödem patienter genom att hjälpa dom förbättra sin kroppsbild i vardagen. Genom att bygga på själv-hanterings konceptet för människor med kroniska sjukdomar och med en medveten koppling mellan kropp och sinne, skapade jag en uppsättning av artefakter. Vardera artefakt var kopplade till specifika interaktioner för att på så sätt ge dom en djupare betydelse. Genom att leva med dessa tre artefakter, kunde jag utforska teman rörandes självacceptans, obehag och sårbarhet. Dessa är framför allt viktiga när man designar för människor med Lipödem men också för andra grupper av människor som kämpar med liknande problem, såsom kroppsbild.

The sharing of personal information among health and social service organisations is a complex issue and problematic process in present-day England. Organisations which provide services to children face enormous challenges on many fronts. Internal ways of working, evolving best practice, data protection applications, government mandates and new government agencies, rapid changes in technology, and increasing costs are but a few of the challenges with which organisations must contend in order to provide services to children while keeping in step with change. This thesis is an exploration into the process of sharing personal information in the context of public sector reforms. Because there is an increasing emphasis of multi-agency collaboration, this thesis examines the information sharing processes both within and among organisations, particularly those providing services to children. From the broad principles which comprise a socio-technical approach of information sharing, distinct critical factors for successful information sharing and best practices are identified. These critical success factors are then used to evaluate the emerging national database, ContactPoint, highlighting particular areas of concern. In addition, data protection and related issues in the information sharing process are addressed. It is argued that one of the main factors which would support effective information sharing is to add a timeline to the life of a dataset containing personal information, after which the shared information would dissolve. Therefore, this thesis introduces Dynamic Multi-Agency Collaboration (DMAC), a theoretical model of effective information sharing using a limited-life dataset. The limited life of the DMAC dataset gives more control to information providers, encouraging effective information sharing within the parameters of the Data Protection Act 1998.

<p dir="ltr">Generating images from training samples solves the challenge of imbalanced data. It provides the necessary data to run machine learning algorithms for image classification, anomaly detection, and pattern recognition tasks. In medical settings, having imbalanced data results in higher false negatives due to a lack of positive samples. Generative Adversarial Networks (GANs) have been widely adopted for image generation. GANs allow models to train without computing intractable probability while producing high-quality images. However, evaluating GANs has been challenging for the researchers due to a need for an objective function. Most studies assess the quality of generated images and the variety of classes those images cover. Overfitting of training images, however, has received less attention from researchers. When the generated images are mere copies of the training data, GAN models will overfit and will not generalize well. This study examines the ability to detect overfitting of popular metrics: Maximum Mean Discrepancy (MMD) and Fréchet Inception Distance (FID). We investigate the metrics on two types of data: handwritten digits and chest x-ray images using Analysis of Variance (ANOVA) models.</p>

Polymicrogyria (PMG) is one brain disease that mainly occurs in the pediatric brain. Heavy PMG will cause seizures, delayed development, and a series of problems. For this reason, it is critical to effectively identify PMG and start early treatment. Radiologists typically identify PMG through magnetic resonance imaging scans. In this study, we create and open a pediatric MRI dataset (named PPMR dataset) including PMG and controls from the Children's Hospital of Eastern Ontario (CHEO), Ottawa, Canada. The difference between PMG MRIs and control MRIs is subtle and the true distribution of the features of the disease is unknown. Hence, we propose a novel center-based deep contrastive metric learning loss function (named cDCM Loss) to deal with this difficult problem. Cross-entropy-based loss functions do not lead to models with good generalization on small and imbalanced dataset with partially known distributions. We conduct exhaustive experiments on a modified CIFAR-10 dataset to demonstrate the efficacy of our proposed loss function compared to cross-entropy-based loss functions and the state-of-the-art Deep SAD loss function. Additionally, based on our proposed loss function, we customize a deep learning model structure that integrates dilated convolution, squeeze-and-excitation blocks and feature fusion for our PPMR dataset, to achieve 92.01% recall. Since our suggested method is a computer-aided tool to assist radiologists in selecting potential PMG MRIs, 55.04% precision is acceptable. To our best knowledge, this research is the first to apply machine learning techniques to identify PMG only from MRI and our innovative method achieves better results than baseline methods.

No description available.

Breast cancer is one of the deadliest cancers for women. In the US, 1 in 8 women will be diagnosed with breast cancer within their lifetimes. Detection and diagnosis play an important role in saving lives. To this end, many classifiers with varying structures have been designed to classify breast cancer histopathological images. However, randomly partitioning data, like many previous works have done, can lead to artificially inflated accuracies and classifiers that do not generalize. Data leakage occurs when researchers assume that every image in a dataset is independent of each other, which is often not the case for medical datasets, where multiple images are taken of each patient. This work focuses on convolutional neural network binary classifiers using the BreakHis dataset. Previous works are reviewed. Classifiers from previous literature are tested with patient partitioning, where individual patients are placed in the training, testing and validation sets so that there is no overlap. A classifier which previously achieved 93% accuracy consistently, only achieved 79% accuracy with the new patient partition. Robust data augmentation, a Sigmoid output layer and a different form of min-max normalization were utilized to achieve an accuracy of 89.38%. These improvements were shown to be effective with the architectures used. Sigmoid Model 1.1 is shown to perform well compared to much deeper architectures found in literature.

University of Bradford through the International Development Fund / The full text will be available at the end of the embargo: 13th Oct 2024

越來越多帶有時間序列的資料普遍的存在醫學工程、商業統計、財務金融等各領域，例如：在財務金融分析領域中已知的形狀樣式用以預測未來價格趨勢做出買賣的決策。由於時序性資料通常非常的龐大，領域的專家看法也未必相同，所描述出新的形狀樣式剛開始也都是比較粗略的，必須透過不斷的修正才會得到比較精準的結果。有鑒於此，我們實做了一套時序性資料集的形狀查詢語言，透過簡單的語言描述，讓使用者簡便快速的定義出屬於自己的形狀樣式。此外我們也實作出互動式的環境並實際有效率應用於台灣證券交易市場。 / There are more and more time series data in the fields of medical engineering, commerce statistics, finance, etc. For example, in financial analysis, we can forecast the price trends by using some well known chart patterns. People want to find out some new patterns for making their purchase decisions fast and easily. However, it is technical challenging to implement a high-level pattern description language. This thesis implemented a shape query language for time-series datasets. Through the simple syntax, field users can find out there own shape patterns by using a more realistic, easily and fast way. We have also developed an interactive environment that users can apply our shape query language to the data of Taiwan Stock Market efficiently.

Doctor of Philosophy / Department of Civil Engineering / Yacoub M. Najjar / Access to databases after the digital revolutions has become easier because large databases are progressively available. Knowledge discovery in these databases via intelligent data analysis technology is a relatively young and interdisciplinary field. In engineering applications, there is a demand for turning low-level data-based knowledge into a high-level type knowledge via the use of various data analysis methods. The main reason for this demand is that collecting and analyzing databases can be expensive and time consuming. In cases where experimental or empirical data are already available, prediction models can be used to characterize the desired engineering phenomena and/or eliminate unnecessary future experiments and their associated costs. Phenomena characterization, based on available databases, has been utilized via Artificial Neural Networks (ANNs) for more than two decades. However, there is a need to introduce new paradigms to improve the reliability of the available ANN models and optimize their predictions through a hybrid decision system. In this study, a new set of ANN modeling approaches/paradigms along with a new method to tackle partially missing data (Query method) are introduced for this purpose. The potential use of these methods via a hybrid decision making system is examined by utilizing seven available databases which are obtained from civil engineering applications. Overall, the new proposed approaches have shown notable prediction accuracy improvements on the seven databases in terms of quantified statistical accuracy measures. The proposed new methods are capable in effectively characterizing the general behavior of a specific engineering/scientific phenomenon and can be collectively used to optimize predictions with a reasonable degree of accuracy. The utilization of the proposed hybrid decision making system (HDMS) via an Excel-based environment can easily be utilized by the end user, to any available data-rich database, without the need for any excessive type of training.

Au cours des dernières années, la disponibilité toujours croissante de données de télédétection multi-source de l'océan a été un facteur clé pour améliorer notre compréhension des dynamiques de la surface de l'océan. A cet égard, il est essentiel de mettre au point des approches efficaces pour exploiter ces ensembles de données. En particulier, la décomposition des processus géophysiques en modes pertinents est une question clé pour les problèmes de caractérisation, de prédiction et de reconstruction. Inspirés par des progrès récents en séparation aveugle des sources, nous visons, dans la première partie de cette thèse, à étendre les modèles de séparation aveugle de sources sous contraintes de non-négativité au problème de la caractérisation et décomposition d'opérateurs ou fonctions de transfert entre variables d'intérêt. Nous développons des schémas computationnels efficaces reposant sur des fondations mathématiques solides. Nous illustrons la pertinence des modèles de décomposition proposés dans différentes applications impliquant l'analyse et la prédiction de dynamiques géophysiques. Par la suite, étant donné que la disponibilité toujours croissante d'ensembles de données multi-sources supporte l'exploration des approches pilotées par les données en tant qu'alternative aux formulations classiques basées sur des modèles, nous explorons des approches basées sur les données récemment introduits pour l'interpolation des champs géophysiques à partir d'observations satellitaires irrégulièrement échantillonnées. De plus, en vue de la future mission SWOT, la première mission satellitaire à produire des observations d'altimétrie par satellite complètement bidimensionnelles et à large fauchée, nous nous intéressons à évaluer dans quelle mesure les données SWOT permettraient une meilleure reconstruction des champs altimétriques. / In the last few decades, the ever-growing availability of multi-source ocean remote sensing data has been a key factor for improving our understanding of upper ocean dynamics. In this regard, developing efficient approaches to exploit these datasets is of major importance. Particularly, the decomposition of geophysical processes into relevant modes is a key issue for characterization, forecasting and reconstruction problems. Inspired by recent advances in blind source separation, we aim, in the first part of this thesis dissertation, at extending non-negative blind source separation models to the problem of the observation-based characterization and decomposition of linear operators or transfer functions between variables of interest. We develop mathematically sound and computationally efficient schemes. We illustrate the relevance of the proposed decomposition models in different applications involving the analysis and forecasting of geophysical dynamics. Subsequently, given that the ever-increasing availability of multi-source datasets supports the exploration of data-driven alternatives to classical model-driven formulations, we explore recently introduced data-driven models for the interpolation of geophysical fields from irregularly sampled satellite-derived observations. Importantly, with a view towards the future SWOT mission, the first satellite mission to produce complete two-dimensional wide-swath satellite altimetry observations, we focus on assessing the extent to which SWOT data may lead to an improved reconstruction of altimetry fields.