Thesis (MTech (Information Technology))--Cape Peninsula University of Technology, 2018. / Healthcare facilities in South Africa accumulate big data, daily. However, this data is not being utilised to its full potential. The healthcare sector still uses traditional methods to store, process, and analyse data. Currently, there are no big data analytics tools being used in the South African healthcare environment. This study was conducted to establish what factors hinder the effective use of big data in the South African healthcare environment. To fulfil the objectives of this research, qualitative methods were followed. Using the case study method, two healthcare organisations were selected as cases. This enabled the researcher to find similarities between the cases which drove them towards generalisation. The data collected in this study was analysed using the Actor-Network Theory (ANT). Through the application of ANT, the researcher was able to uncover the influencing factors behind big data analytics in the healthcare environment. ANT was essential to the study as it brought out the different interactions that take place between human and non-human actors, resulting in big data. From the analysis, findings were drawn and interpreted. The interpretation of findings led to the developed framework in Figure 5.5. This framework was developed to guide the healthcare sector of South Africa towards the selection of appropriate big data analytics tools. The contribution of this study is in twofold; namely, theoretically and practically. Theoretically, the developed framework will act as a useful guide towards the selection of big data analytics tools. Practically, this guide can be used by South African healthcare practitioners to gain better understanding of big data analytics and how they can be used to improve healthcare service delivery.

published_or_final_version / Nursing Studies / Master / Master of Nursing in Advanced Practice

Objective The objective of this study was to use machine learning and health standards to address the problem of clinical data interoperability across healthcare institutions. Addressing this problem has the potential to make clinical data comparable, searchable and exchangeable between healthcare providers. Data sources Structured and unstructured data has been used to conduct the experiments in this study. The data was collected from two disparate data sources namely MIMIC-III and NHanes. The MIMIC-III database stored data from two electronic health record systems which are CareVue and MetaVision. The data stored in these systems was not recorded with the same standards; therefore, it was not comparable because some values were conflicting, while one system would store an abbreviation of a clinical concept, the other would store the full concept name and some of the attributes contained missing information. These few issues that have been identified make this form of data a good candidate for this study. From the identified data sources, laboratory, physical examination, vital signs, and behavioural data were used for this study. Methods This research employed a CRISP-DM framework as a guideline for all the stages of data mining. Two sets of classification experiments were conducted, one for the classification of structured data, and the other for unstructured data. For the first experiment, Edit distance, TFIDF and JaroWinkler were used to calculate the similarity weights between two datasets, one coded with the LOINC terminology standard and another not coded. Similar sets of data were classified as matches while dissimilar sets were classified as non-matching. Then soundex indexing method was used to reduce the number of potential comparisons. Thereafter, three classification algorithms were trained and tested, and the performance of each was evaluated through the ROC curve. Alternatively the second experiment was aimed at extracting patient’s smoking status information from a clinical corpus. A sequence-oriented classification algorithm called CRF was used for learning related concepts from the given clinical corpus. Hence, word embedding, random indexing, and word shape features were used for understanding the meaning in the corpus. Results Having optimized all the model’s parameters through the v-fold cross validation on a sampled training set of structured data ( ), out of 24 features, only ( 8) were selected for a classification task. RapidMiner was used to train and test all the classification algorithms. On the final run of classification process, the last contenders were SVM and the decision tree classifier. SVM yielded an accuracy of 92.5% when the and parameters were set to and . These results were obtained after more relevant features were identified, having observed that the classifiers were biased on the initial data. On the other side, unstructured data was annotated via the UIMA Ruta scripting language, then trained through the CRFSuite which comes with the CLAMP toolkit. The CRF classifier obtained an F-measure of 94.8% for “nonsmoker” class, 83.0% for “currentsmoker”, and 65.7% for “pastsmoker”. It was observed that as more relevant data was added, the performance of the classifier improved. The results show that there is a need for the use of FHIR resources for exchanging clinical data between healthcare institutions. FHIR is free, it uses: profiles to extend coding standards; RESTFul API to exchange messages; and JSON, XML and turtle for representing messages. Data could be stored as JSON format on a NoSQL database such as CouchDB, which makes it available for further post extraction exploration. Conclusion This study has provided a method for learning a clinical coding standard by a computer algorithm, then applying that learned standard to unstandardized data so that unstandardized data could be easily exchangeable, comparable and searchable and ultimately achieve data interoperability. Even though this study was applied on a limited scale, in future, the study would explore the standardization of patient’s long-lived data from multiple sources using the SHARPn open-sourced tools and data scaling platforms / Information Science / M. Sc. (Computing)

Since the early 20th century, the view has developed that high quality health care can be delivered only when all the pertinent data about the health of a patient is available to the clinician. Various types of health records have emerged to serve the needs of healthcare providers and more recently, patients or consumers. These health records include, but are not limited to, Personal Health Records, Electronic Heath Records, Electronic Medical Records and Payer-Based Health Records. Payer-Based Health Records emerged to serve the needs of medical aids or health care plans. Electronic Medical Records and Electronic Health Records were targeted at the healthcare provider market, whereas a gap developed in the patient market. Personal Health Records were developed to address the patient market, but adoption was slow at first. The success of online social networking reignited the flame that Personal Health Records needed and online consumer-based Personal Health Records were developed. Despite all the various types of health records, there still seems to be a lack of meaningful use of personal health records in modern society. The purpose of this dissertation is to propose a framework for Personal Health Records in online social networking, to address the issue of a lack of a central, accessible repository for health records. In order for a Personal Health Record to serve this need it has to be of meaningful use. The capability of a PHR to be of meaningful use is core to this research. In order to determine whether a Personal Health Record is of meaningful use, a tool is developed to evaluate Personal Health Records. This evaluation tool takes into account all the attributes that a Personal Health Record which is of meaningful use should comprise of. Suitable ratings are allocated to enable measuring of each attribute. A model is compiled to facilitate the selection of six Personal Health Records to be evaluated. One of these six Personal Health Records acts as a pilot site to test the evaluation tool in order to determine the tool’s utility and effect improvements. The other five Personal Health Records are then evaluated to measure their adherence to the attributes of meaningful use. These findings, together with a literature study on the various types of health records and the evaluation tool, inform the building blocks used to present the framework. It is hoped that the framework for Personal Health Records in online social networking proposed in this research, may be of benefit to provide clear guidance for the achievement of a central or integrated, accessible repository for health records through the meaningful use of Personal Health Records.

Implementations of information technology solutions to address specific information problems are only successful when the technology is utilized. The antecedents of technology use involve user, system, task and organization characteristics as well as externalities which can affect all of these entities. However, measurement of the interaction effects between these entities can act as a proxy for individual attribute values. A model is proposed which based upon evaluation of these interaction effects can predict technology utilization. This model was tested with systems being implemented at a pediatric health care facility. Results from this study provide insight into the relationship between the antecedents of technology utilization. Specifically, task time provided significant direct causal effects on utilization. Indirect causal effects were identified in task value and perceived utility constructs. Perceived utility, along with organizational support also provided direct causal effects on user satisfaction. Task value also impacted user satisfaction in an indirect fashion. Also, results provide a predictive model and taxonomy of variables which can be applied to predict or manipulate the likelihood of utilization for planned technology.

Developments in electronics, computing and communication technologies have transformed IT systems from desktop and tightly coupled mainframe computers of the past to modern day highly complex distributed systems. These ICT systems interact with humans at a much advanced level than what was envisaged during the early years of computer development. The ICT systems of today have gone through various phases of developments by absorbing intermediate and modern day concepts such as networked computing, utility, on demand and autonomic computing, virtualisation and so on. We now live in a ubiquitous computing and digital economy era where computing systems have penetrated into the human lives to a degree where these systems are becoming invisible. The price of these developments is in the increased costs, higher risks and higher complexity. There is a compelling need to study these emerging systems, their applications, and the emerging market sectors that they are penetrating into. Motivated by the challenges and opportunities offered by the modern day ICT technologies, we aim in this thesis to explore the major technological developments that have happened in the ICT systems during this century with a focus on developing techniques to manage applied ICT systems in digital economy. In the process, we wish to also touch on the evolution of ICT systems and discuss these in context of the state of the art technologies and applications. We have identified the two most transformative technologies of this century, grid computing and cloud computing, and two application areas, intelligent healthcare and transportation systems. The contribution of this thesis is multidisciplinary in four broad areas. Firstly, a workload model of a grid-based ICT system in the healthcare sector is proposed and analysed using multiple healthcare organisations and applications. Secondly, an innovative intelligent system for the management of disasters in urban environments using cloud computing is proposed and analysed. Thirdly, cloud computing market sectors, applications, and workload are analysed using over 200 real life case studies. Fourthly, a detailed background and literature review is provided on grid computing and cloud computing. Finally, directions for future work are given. The work contributes in multidisciplinary fields involving healthcare, transportation, mobile computing, vehicular networking, grid, cloud, and distributed computing. The discussions presented in this thesis on the historical developments, technology and architectural details of grid computing have served to understand as to how and why grid computing was seen in the past as the global infrastructure of the future. These discussions on grid computing also provided the basis that we subsequently used to explain the background, motivations, technological details, and ongoing developments in cloud computing. The introductory chapters on grid and cloud computing, collectively, have provided an insight into the evolution of ICT systems over the last 50+ years - from mainframes to microcomputers, internet, distributed computing, cluster computing, and computing as a utility and service. The existing and proposed applications of grid and cloud computing in healthcare and transport were used to further elaborate the two technologies and the ongoing ICT developments in the digital economy. The workload models and analyses of grid and cloud computing systems can be used by the practitioners for the design and resource management of ICT systems.

Thesis (MTech (Information Technology))--Cape Peninsula University of Technology, 2016. / Health information policies are constantly becoming a key feature in directing information usage in healthcare. After the passing of the Health Information Technology for Economic and Clinical Health (HITECH) Act in 2009 and the Affordable Care Act (ACA) passed in 2010, in the United States, there has been an increase in health systems innovations. Coupling this health systems hype is the current buzz concept in Information Technology, „Big data‟. The prospects of big data are full of potential, even more so in the healthcare field where the accuracy of data is life critical. How big health data can be used to achieve improved health is now the goal of the current health informatics practitioner. Even more exciting is the amount of health data being generated by patients via personal handheld devices and other forms of technology that exclude the healthcare practitioner. This patient-generated data is also known as Personal Health Records, PHR. To achieve meaningful use of PHRs and healthcare data in general through big data, a couple of hurdles have to be overcome. First and foremost is the issue of privacy and confidentiality of the patients whose data is in concern. Secondly is the perceived trustworthiness of PHRs by healthcare practitioners. Other issues to take into context are data rights and ownership, data suppression, IP protection, data anonymisation and reidentification, information flow and regulations as well as consent biases. This study sought to understand the role of data policies in the process of data utilisation in the healthcare sector with added interest on PHRs utilisation as part of big health data.

Senior Health Care System (SHCS) is created for users to enter participants' conditions and store information in a central database. When users are ready for quarterly assessments the system generates a simple summary that can be reviewed, modified, and saved as part of the summary assessments, which are required by Federal and California law.