Work

A Tool for integrating dynamic healthcare data sources

Year: 2020
Type: dissertation
Abstract: Currently, there is a growing interest in finding a comprehensive method in ETL process technology that can deal with healthcare data formats. Healthcare centers are very concerned about their sensitive information which mostly has patients' data inside. they avoid using the market-driven ETL tools because they need higher performance and adaption for complex security treatment. On the grounds of this, they are in great need of a stable ETL tool that does their transformation for them privately. One of the common data formats used by various healthcare providers for the exchange of clinical and administrative data among software applications is Health Level Seven (HL7). HL7 refers to a series of international standards that are used to move clinical and administrative data between applications by different healthcare providers. Typically, hospitals and other healthcare providers have several different data structures that are used for everything from billing reports to monitoring clinical data for patients. Standardized HL7 format is one of the widely used cases, with information about the patient identification and results of patient tests. Data analysts in the bioinformatics department typically need to analyze and review these kinds of information carefully. HL7 has a standard format for describing each of this information but it is a time-consuming process to understand and prepare them for analysis. In this thesis, a powerful ETL tool named ETL_for_HL7 is proposed to deal with this type of data. Usually, HL7 data is made available in the XML format file by the laboratory. The main objective of this thesis is to understand the HL7 data format, design a proper data model, create a valid conceptual schema and convert it to a physical data warehouse, implement an ETL tool in the way that the laboratory saves it and sends it to the data management department, transform it into analysis-ready formats (e.g., DW schema) and finally load it into the data warehouse. The tool starts working by getting the patient's ID and the date of patient's test as parameters. It selects the relevant XML file from the input folder. In the second phase, it starts transforming it from encoded XML that has an HL7 data format within to JSON format file and then extracts two types of information from the entire response. They are patient identification and bio values of the patient's test result. The third phase starts when the results are ready in JSON format. The data can now load into the data warehouse that we previously implemented. The data is able to be used by any analysis department and this was our main objective.
Language: English
Open Access status: green
OpenAlex ID: W3136953768
Cites: 22
Cited by:
Related to: 10