A Tomografia de Impedância Elétrica (TIE) é uma técnica de imageamento, ainda em desenvolvimento, por meio da qual são extraídas imagens correspondentes à distribuição da impedância elétrica de uma seção transversal de um objeto sob análise a partir de medidas elétricas realizadas em sua superfície. Apesar de seus benefícios e vantagens sobre outras técnicas de imageamento, suas imagens não oferecem uma boa resolução espacial. Em imagens TIE de tórax, um dos maiores desafios reside no tratamento da perfusão pulmonar, pois as perspectivas de uso clínico são inúmeras. Assim sendo, melhorar a localização de seus principais órgãos é um dos grandes objetivos. Com o intuito de melhorar a resolução anatômica foi desenvolvida uma metodologia para identificação de região cardíaca em imagens de tomografia de impedância elétrica de perfusão pulmonar por meio da transformada wavelet, utilizando imagens TIE de porcos. Inicialmente foi realizada uma série de estudos de diferentes abordagens com vistas a identificar aquelas características que pudessem indicar similaridades ou diferenças intrínsecas entre os pixels de diferentes regiões anatômicas. Com base nestes estudos, cinco métodos baseados na análise wavelet foram desenvolvidos. Um primeiro conjunto de experimentos, realizado em um único porco, foi utilizado no desenvolvimento e aperfeiçoamento dos métodos. Posteriormente, outros experimentos, envolvendo quatro porcos em diferentes condições fisiológicas, foram utilizados na avaliação de desempenho destes métodos. As imagens de perfusão wavelet foram comparadas com as imagens de perfusão obtidas pelo método de injeção de uma solução hipertônica, considerada como padrão de referência das imagens de perfusão. A metodologia wavelet proposta por este trabalho foi o método, dentre os cinco desenvolvidos, que obteve os melhores resultados. Ela foi capaz de identificar a região cardíaca de cinco porcos submetidos a diversas condições fisiológicas, demonstrando robustez e resultados muito satisfatórios, não apenas em termos quantitativos, com uma área média da curva ROC de 0,86, mas também na qualidade das imagens obtidas, onde os contornos delimitando a região cardíaca ficaram bem definidos e de formato circular, de acordo com o que se esperava. Portanto, o objetivo maior deste trabalho que era melhorar a resolução espacial de imagens TIE de perfusão pulmonar, foi atingido com excelentes resultados e vantagens adicionais, como por exemplo, a possibilidade de sua implementação em equipamentos TIE de monitoramento do tórax e desta forma colaborar no aperfeiçoamento de sistemas de apoio à decisão médica em ambientes críticos, como é o caso das Unidades de Terapia Intensiva (UTIs). / Electrical Impedance Tomography (EIT) is an imaging technique, still in development, which allows imaging of the distribution of conductivity in a cross section of an object under analysis from electrical measures made on its surface. Despite its benefits and advantages over other imaging techniques, its images still do not offer a good spatial resolution. One of the biggest challenges in EIT thorax images is the treatment of the lung perfusion because the perspectives for clinical use are numerous. Thus, there is a great interest in improving the localization of its main organs. In order to improve the anatomical resolution was developed a methodology for identification of cardiac region in images of electrical impedance tomography of pulmonary perfusion by means of wavelet transform, using EIT images of pigs. Some preliminary studies of different approaches were performed in order to identify those characteristics that would indicate intrinsic similarities or differences among the pixels of different anatomical regions. These studies propitiated the development of five methods based on wavelet transform. A first set of experiments, performed in a single pig, was used in developing and improving of the methods. Subsequently, other experiments, involving four pigs in different physiological conditions, were performed to evaluate the performance of these methods. The wavelet perfusional images were compared with the perfusional images obtained by the method of injection of a hypertonic solution, considered as the perfusional reference standard images. Amongst the five developed methods, the best of them was selected as the wavelet methodology proposed by this work. It was capable to identify the heart region of five pigs under different physiological conditions, demonstrating to robustness and very satisfactory results, not only in quantitative terms, with an average area of the ROC curve of 0.86, but also in the quality of the images, where the contours delimiting the cardiac region were well defined and of circular format, according to what was expected. Therefore, the main objective of this work, that was to improve the spatial resolution of EIT images of pulmonary perfusion, was reached with excellent results and additional benefits such as the possibility of its implementation in EIT equipments for monitoring thorax and thus collaborate in improving of medical decision support systems in critical environments, as for example the Intensive Care Units (ICUs).

Consolidating patient and clinical data to support better-informed clinical decisions remains a primary function of electronic health records (EHRs). In the United States, nearly 6 million patients receive care from an accountable care organization (ACO). Knowledge of clinical decision support (CDS) tool design for use by physicians participating in ACOs remains limited. The purpose of this quantitative study was to examine whether a significant correlation exists between characteristics of alert content and alert timing (the independent variables) and physician perceptions of improved ACO quality measure adherence during electronic ordering (the dependent variable). Sociotechnical theory supported the theoretical framework for this research. Sixty-nine physician executives using either a Cerner Incorporated or Epic Systems EHR in a hospital or health system affiliated ACO participated in the online survey. The results of the regression analysis were statistically significant, R2 = .108, F(2,66) = 3.99, p = .023, indicating that characteristics of alert content and timing affect physician perceptions for improving their adherence to ACO quality measures. However, analysis of each independent variable showed alert content highly correlated with the dependent variable (p = .007) with no significant correlation found between workflow timing and the dependent variable (p = .724). Understanding the factors that support physician acceptance of alerts is essential to third-party software developers and health care organizations designing CDS tools. Providing physicians with improved EHR-integrated CDS tools supports the population health goal of ACOs in delivering better patient care.

The aim of this study is to examine the impact of a cloud-based CDS alerting system for SIRS, a precursor to sepsis, and sepsis itself, on adult patient and process outcomes at VCU Health System. The two main hypotheses are: 1) the implementation of cloud-based SIRS and sepsis alerts will lead to lower sepsis-related mortality and lower average length of stay, and 2) the implementation of cloud-based SIRS and sepsis alerts will lead to more frequent ordering of the Sepsis PowerPlan and more recording of sepsis diagnoses. To measure these outcomes, a pre-post study was conducted. A pre-implementation group diagnosed with sepsis within the year leading up to the alert intervention consisted of 1,551 unique inpatient visits, and the three-year post-implementation sample size was 9,711 visits, for a total cohort of 11,262 visits. Logistic regression and multiple linear regression were used to test the hypotheses. Study results showed that sepsis-related mortality was slightly higher after the implementation of SIRS alerts, but the presence of sepsis alerts did not have a significant relationship to mortality. The average length of stay and the total number of recorded sepsis diagnoses were higher after the implementation of both SIRS and sepsis alerts, while ordering of the Sepsis Initial Resuscitation PowerPlan was lower. There is preliminary evidence from this study that more sepsis diagnoses are made as a result of alert adoption, suggesting that clinicians can consider the implementation of these alerts in order to capture a higher number of sepsis diagnoses.

An ongoing objective in healthcare is the development of tools to improve patient decision-making and surgical outcomes for patients with breast cancer that have undergone or plan to undergo breast reconstruction. In keeping with the bioethical concept of autonomy, these decision models are patient-oriented and expansive, covering a range of different patient decision-makers. In pursuit of these goals, this dissertation contributes to the development of a prototype shared decision support system that will guide patients with breast cancer and their physicians in making decisions about breast reconstruction. This dissertation applies principles in decision analysis to breast reconstruction decision-making. In this dissertation, we examine three important areas of decision-making: (1) the options available to decision-makers, (2) the validity of probabilistic information assessed from reconstructive surgeons, and (3) the feasibility of applying multiattribute utility theory. In addition, it discusses the influences of breast aesthetics and proposes a measure for quantifying such influences. The dissertation concludes with a fictional case study that demonstrates the integration of the findings and application of decision analysis in patient-oriented shared breast reconstruction decision-making. Through the implementation of decision analysis principles, cognitive biases and emotion may be attenuated, clearing the decision-maker’s judgment, and ostensibly leading to good decisions. While good decisions cannot guarantee good outcomes at the individual level, they can be expected to improve outcomes for patients with breast cancer as a whole. And regardless of the outcome, good decisions yield clarity of action and grant the decision-maker a measure of peace in an otherwise uncertain world. / text

Health care providers internationally are facing challenges surrounding the delivery of high quality, cost effective services. The use of integrated electronic information systems is seen by many people working in the health sector as a way to address some of the associated issues. In New Zealand the primary health care sector has been restructured to follow a population based care model and provides services through not-for-profit Primary Health Organisations (PHOs). PHOs, together with their District Health Boards (DHBs), contributing service providers, and local communities, are responsible for the care of their enrolled populations. The Ministry of Health (MoH) is streamlining information sharing in this environment through improvements to computer based information systems (IS). By providing health professionals with improved access to required information within an appropriate time frame, services can be targeted efficiently and effectively and patient health outcomes potentially improved. However, the adoption of IS in health care has been slower than in other industries. Therefore, a thorough knowledge of health care professionals’ attitudes to, and use of, available IS is currently needed to contribute to the development of appropriate systems. This research employs a multiple case study strategy to establish the usage of IS by three New Zealand PHOs and their member primary health care providers (PHPs), with a focus on the role of IS in clinical decision support (CDS). A mixed method approach including semi-structured interviews and postal surveys was used in the study. Firstly, the research develops and applies a survey tool based on an adaptation of an existing framework, for the study of IT sophistication in the organisations. This provides the foundation for an in-depth study of the use of computerised CDS (eCDS) in the PHO environment. Secondly, a conceptual model of eCDS utilisation is presented, illustrating the variation of eCDS use by member general practitioner (GP) practices within individual organisations. Thirdly, five areas of importance for improving eCDS utilisation within PHO’s are identified, contributing information of use to organisations, practitioners, planners, and systems developers. Lastly, the research provides a structure for the study of the domain of eCDS in PHOs by presenting a research approach and information specific for the area.

abstract: Clinical Decision Support (CDS) is primarily associated with alerts, reminders, order entry, rule-based invocation, diagnostic aids, and on-demand information retrieval. While valuable, these foci have been in production use for decades, and do not provide a broader, interoperable means of plugging structured clinical knowledge into live electronic health record (EHR) ecosystems for purposes of orchestrating the user experiences of patients and clinicians. To date, the gap between knowledge representation and user-facing EHR integration has been considered an “implementation concern” requiring unscalable manual human efforts and governance coordination. Drafting a questionnaire engineered to meet the specifications of the HL7 CDS Knowledge Artifact specification, for example, carries no reasonable expectation that it may be imported and deployed into a live system without significant burdens. Dramatic reduction of the time and effort gap in the research and application cycle could be revolutionary. Doing so, however, requires both a floor-to-ceiling precoordination of functional boundaries in the knowledge management lifecycle, as well as formalization of the human processes by which this occurs. This research introduces ARTAKA: Architecture for Real-Time Application of Knowledge Artifacts, as a concrete floor-to-ceiling technological blueprint for both provider heath IT (HIT) and vendor organizations to incrementally introduce value into existing systems dynamically. This is made possible by service-ization of curated knowledge artifacts, then injected into a highly scalable backend infrastructure by automated orchestration through public marketplaces. Supplementary examples of client app integration are also provided. Compilation of knowledge into platform-specific form has been left flexible, in so far as implementations comply with ARTAKA’s Context Event Service (CES) communication and Health Services Platform (HSP) Marketplace service packaging standards. Towards the goal of interoperable human processes, ARTAKA’s treatment of knowledge artifacts as a specialized form of software allows knowledge engineers to operate as a type of software engineering practice. Thus, nearly a century of software development processes, tools, policies, and lessons offer immediate benefit: in some cases, with remarkable parity. Analyses of experimentation is provided with guidelines in how choice aspects of software development life cycles (SDLCs) apply to knowledge artifact development in an ARTAKA environment. Portions of this culminating document have been further initiated with Standards Developing Organizations (SDOs) intended to ultimately produce normative standards, as have active relationships with other bodies. / Dissertation/Thesis / Doctoral Dissertation Biomedical Informatics 2018

Clinical diagnosis of chronic disease is a vital and challenging research problem which requires intensive clinical practice guidelines in order to ensure consistent and efficient patient care. Conventional medical diagnosis systems inculcate certain limitations, like complex diagnosis processes, lack of expertise, lack of well described procedures for conducting diagnoses, low computing skills, and so on. Automated clinical decision support system (CDSS) can help physicians and radiologists to overcome these challenges by combining the competency of radiologists and physicians with the capabilities of computers. CDSS depend on many techniques from the fields of image acquisition, image processing, pattern recognition, machine learning as well as optimization for medical data analysis to produce efficient diagnoses. In this dissertation, we discuss the current challenges in designing an efficient CDSS as well as a number of the latest techniques (while identifying best practices for each stage of the framework) to meet these challenges by finding informative patterns in the medical dataset, analysing them and building a descriptive model of the object of interest and thus aiding in medical diagnosis. To meet these challenges, we propose an extension of conventional clinical decision support system framework, by incorporating artificial immune network (AIN) based hyper-parameter optimization as integral part of it. We applied the conventional as well as optimized CDSS on four case studies (most of them comprise medical images) for efficient medical diagnosis and compared the results. The first key contribution is the novel application of a local energy-based shape histogram (LESH) as the feature set for the recognition of abnormalities in mammograms. We investigated the implication of this technique for the mammogram datasets of the Mammographic Image Analysis Society and INbreast. In the evaluation, regions of interest were extracted from the mammograms, their LESH features were calculated, and they were fed to support vector machine (SVM) and echo state network (ESN) classifiers. In addition, the impact of selecting a subset of LESH features based on the classification performance was also observed and benchmarked against a state-of-the-art wavelet based feature extraction method. The second key contribution is to apply the LESH technique to detect lung cancer. The JSRT Digital Image Database of chest radiographs was selected for research experimentation. Prior to LESH feature extraction, we enhanced the radiograph images using a contrast limited adaptive histogram equalization (CLAHE) approach. Selected state-of-the-art cognitive machine learning classifiers, namely the extreme learning machine (ELM), SVM and ESN, were then applied using the LESH extracted features to enable the efficient diagnosis of a correct medical state (the existence of benign or malignant cancer) in the x-ray images. Comparative simulation results, evaluated using the classification accuracy performance measure, were further benchmarked against state-of-the-art wavelet based features, and authenticated the distinct capability of our proposed framework for enhancing the diagnosis outcome. As the third contribution, this thesis presents a novel technique for detecting breast cancer in volumetric medical images based on a three-dimensional (3D) LESH model. It is a hybrid approach, and combines the 3D LESH feature extraction technique with machine learning classifiers to detect breast cancer from MRI images. The proposed system applies CLAHE to the MRI images before extracting the 3D LESH features. Furthermore, a selected subset of features is fed to a machine learning classifier, namely the SVM, ELM or ESN, to detect abnormalities and to distinguish between different stages of abnormality. The results indicate the high performance of the proposed system. When compared with the wavelet-based feature extraction technique, statistical analysis testifies to the significance of our proposed algorithm. The fourth contribution is a novel application of the (AIN) for optimizing machine learning classification algorithms as part of CDSS. We employed our proposed technique in conjunction with selected machine learning classifiers, namely the ELM, SVM and ESN, and validated it using the benchmark medical datasets of PIMA India diabetes and BUPA liver disorders, two-dimensional (2D) medical images, namely MIAS and INbreast and JSRT chest radiographs, as well as on the three-dimensional TCGA-BRCA breast MRI dataset. The results were investigated using the classification accuracy measure and the learning time. We also compared our methodology with the benchmarked multi-objective genetic algorithm (ES)-based optimization technique. The results authenticate the potential of the AIN optimised CDSS.

L'explosion récente de données d'imagerie cardiaque a été phénoménale. L'utilisation intelligente des grandes bases de données annotées pourrait constituer une aide précieuse au diagnostic et à la planification de thérapie. En plus des défis inhérents à la grande taille de ces banques de données, elles sont difficilement utilisables en l'état. Les données ne sont pas structurées, le contenu des images est variable et mal indexé, et les métadonnées ne sont pas standardisées. L'objectif de cette thèse est donc le traitement, l'analyse et l'interprétation automatique de ces bases de données afin de faciliter leur utilisation par les spécialistes de cardiologie. Dans ce but, la thèse explore les outils d'apprentissage automatique supervisé, ce qui aide à exploiter ces grandes quantités d'images cardiaques et trouver de meilleures représentations. Tout d'abord, la visualisation et l'interprétation d'images est améliorée en développant une méthode de reconnaissance automatique des plans d'acquisition couramment utilisés en imagerie cardiaque. La méthode se base sur l'apprentissage par forêts aléatoires et par réseaux de neurones à convolution, en utilisant des larges banques d'images, où des types de vues cardiaques sont préalablement établies. La thèse s'attache dans un deuxième temps au traitement automatique des images cardiaques, avec en perspective l'extraction d'indices cliniques pertinents. La segmentation des structures cardiaques est une étape clé de ce processus. A cet effet une méthode basée sur les forêts aléatoires qui exploite des attributs spatio-temporels originaux pour la segmentation automatique dans des images 3Det 3D+t est proposée. En troisième partie, l'apprentissage supervisé de sémantique cardiaque est enrichi grâce à une méthode de collecte en ligne d'annotations d'usagers. Enfin, la dernière partie utilise l'apprentissage automatique basé sur les forêts aléatoires pour cartographier des banques d'images cardiaques, tout en établissant les notions de distance et de voisinage d'images. Une application est proposée afin de retrouver dans une banque de données, les images les plus similaires à celle d'un nouveau patient. / The recent growth of data in cardiac databases has been phenomenal. Cleveruse of these databases could help find supporting evidence for better diagnosis and treatment planning. In addition to the challenges inherent to the large quantity of data, the databases are difficult to use in their current state. Data coming from multiple sources are often unstructured, the image content is variable and the metadata are not standardised. The objective of this thesis is therefore to simplify the use of large databases for cardiology specialists withautomated image processing, analysis and interpretation tools. The proposed tools are largely based on supervised machine learning techniques, i.e. algorithms which can learn from large quantities of cardiac images with groundtruth annotations and which automatically find the best representations. First, the inconsistent metadata are cleaned, interpretation and visualisation of images is improved by automatically recognising commonly used cardiac magnetic resonance imaging views from image content. The method is based on decision forests and convolutional neural networks trained on a large image dataset. Second, the thesis explores ways to use machine learning for extraction of relevant clinical measures (e.g. volumes and masses) from3D and 3D+t cardiac images. New spatio-temporal image features are designed andclassification forests are trained to learn how to automatically segment the main cardiac structures (left ventricle and left atrium) from voxel-wise label maps. Third, a web interface is designed to collect pairwise image comparisons and to learn how to describe the hearts with semantic attributes (e.g. dilation, kineticity). In the last part of the thesis, a forest-based machinelearning technique is used to map cardiac images to establish distances and neighborhoods between images. One application is retrieval of the most similar images.

Making complex medical decisions is becoming an increasingly challenging task due to the growing amount of available evidence to consider and the higher demand for personalized treatment and patient care. IT systems for the provision of clinical decision support (CDS) can provide sustainable relief if decisions are automatically evaluated and processed. In this paper, we propose an approach for quantifying similarity between new and previously recorded medical cases to enable significant knowledge transfer for reasoning tasks on a patient-level. Methodologically, 102 medical cases with oropharyngeal carcinoma were analyzed retrospectively. Based on independent disease characteristics, patient-specific data vectors including relevant information entities for primary and adjuvant treatment decisions were created. Utilizing the ϕK correlation coefficient as the methodological foundation of our approach, we were able to determine the predictive impact of each characteristic, thus enabling significant reduction of the feature space to allow for further analysis of the intra-variable distances between the respective feature states. The results revealed a significant feature-space reduction from initially 19 down to only 6 diagnostic variables (ϕK correlation coefficient ≥ 0.3, ϕK significance test ≥ 2.5) for the primary and 7 variables (from initially 14) for the adjuvant treatment setting. Further investigation on the resulting characteristics showed a non-linear behavior in relation to the corresponding distances on intra-variable level. Through the implementation of a 10-fold cross-validation procedure, we were further able to identify 8 (primary treatment) matching cases with an evaluation score of 1.0 and 9 (adjuvant treatment) matching cases with an evaluation score of 0.957 based on their shared treatment procedure as the endpoint for similarity definition. Based on those promising results, we conclude that our proposed method for using data-driven similarity measures for application in medical decision-making is able to offer valuable assistance for physicians. Furthermore, we consider our approach as universal in regard to other clinical use-cases, which would allow for an easy-to-implement adaptation for a range of further medical decision-making scenarios.

Background: Laboratory results are of central importance for clinical decision making. The time span between availability and review of results by clinicians is crucial to patient care. Clinical decision support systems (CDSS) are computational tools that can identify critical values automatically and help decrease treatment delay. Objective: With this work, we aimed to implement and evaluate a CDSS that supports health care professionals and improves patient safety. In addition to our experiences, we also describe its main components in a general manner to make it applicable to a wide range of medical institutions and to empower colleagues to implement a similar system in their facilities. Methods: Technical requirements must be taken into account before implementing a CDSS that performs laboratory diagnostics (labCDSS). These can be planned within the functional components of a reactive software agent, a computational framework for such a CDSS. Results: We present AMPEL (Analysis and Reporting System for the Improvement of Patient Safety through Real-Time Integration of Laboratory Findings), a labCDSS that notifies health care professionals if a life-threatening medical condition is detected. We developed and implemented AMPEL at a university hospital and regional hospitals in Germany (University of Leipzig Medical Center and the Muldental Clinics in Grimma and Wurzen). It currently runs 5 different algorithms in parallel: hypokalemia, hypercalcemia, hyponatremia, hyperlactatemia, and acute kidney injury. Conclusions: AMPEL enables continuous surveillance of patients. The system is constantly being evaluated and extended and has the capacity for many more algorithms. We hope to encourage colleagues from other institutions to design and implement similar CDSS using the theory, specifications, and experiences described in this work.