Data Platforms and Ecosystems in Healthcare

Permanent URI for this collectionhttps://hdl.handle.net/10125/107479

Browse

Recent Submissions

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item type: Item ,
    Privacy and User Satisfaction in Digital Health Applications
    (2024-01-03) Schewina, Kai
    Digital health applications are promising to healthcare due to their important position in the digital health ecosystem. User satisfaction with apps is crucial for their widespread use, but there are substantial concerns regarding the usage of users’ health data preventing their dissemination. To study the relationship between privacy and user satisfaction, we utilize a dataset encompassing almost 94,412 apps from the Apple App Store. We apply propensity score matching to estimate the average treatment effects of multiple privacy settings that describe what data is collected and to what extent. We find that not collecting data on users is associated with a decrease in user satisfaction. Linking data to users’ profiles and tracking users’ data across applications increases user satisfaction. Our findings challenge the currently assumed relationship between the two variables and provide nuanced information for decision-makers about data collection in digital health applications.
  • Thumbnail Image
    Item type: Item ,
    Leveraging Large Language Models for Simplified Patient Summary Generation, Literature Retrieval and Medical Information Summarization: A Health CASCADE Study
    (2024-01-03) Balaskas, Georgios; Papadopoulos, Homer; Korakis, Antonis
    In the evolving healthcare landscape, integrating advanced technologies such as machine learning and natural language processing has become vital. This paper presents an innovative system that leverages modern Natural Language Processing (NLP) capabilities to extract information from Electronic Health Records (EHRs) and generate simplified patient summaries (SPS). These SPS are subsequently used to provide clinicians with summaries of relevant academic literature, improving their ability to access pertinent information efficiently. The system architecture employs Large Language Models (LLMs) to generate SPSs and summarize relevant information, while dense vector retriever models are used for information retrieval from document corpus, which is created by combining parts of publicly available datasets such as PubMed, the CORD19 dataset, and more. The presented system has the potential to significantly reduce the time and effort required by clinicians to access relevant patient information, allowing them to concentrate more on patient care and contribute to improved patient outcomes.
  • Thumbnail Image
    Item type: Item ,
    Data Size Matters: The Impact of Message Framing in Different Health Scenarios on the Donation of Personal Health Information
    (2024-01-03) Klein, Julia; Masuch, Kristin; Schulze, Laura; Trang, Simon
    Health data donation allows individuals to share their personal health information for the greater good. As privacy concerns hinder many individuals from disclosing such sensitive information, this study investigates how benefit appeals, attribute framing, and health conditions can influence the intention to donate personal health information. We conduct a scenario-based online experiment and answer our research question using data from a German sample (n=208). We used a vignette design with a 2 (benefit appeal) x 2 (attribute framing) x 2 (health condition) mixed-subject design. Our results indicate that benefit appeals, attribute framing, and health condition statistically significantly influence the intention to donate personal health information. Our findings contribute to health in information systems and the privacy literature stream by extending knowledge regarding phenomena with multi-layered benefit structures and by opening future research possibilities in the context of health data donation.
  • Thumbnail Image
    Item type: Item ,
    Lean Study Host: Towards an Automated Pipeline for Multi-Center Study Hosting
    (2024-01-03) Heine, Lukas; Hörst, Fabian; Nasca, Enrico; Siveke, Jens; Egger, Jan; Kim, Moon; Bahnsen, Fin Hendrik; Kleesiek, Jens
    Medical studies are an essential part of advancing research. A uniform, flexible software infrastructure that allows for straightforward data management stands at the core of studies that involve multiple sites. Such a solution must accommodate the specific technical needs of clinical practitioners and researchers, such as uploading, viewing, downloading, annotating, and sharing image material in various forms. The current tool landscape needs a solution that bridges the gap between intuitive data governance and usability without introducing undesired technical and legal overhead. We present "Lean Study Host'' (LSH), a novel, open-source approach to clinical study data management that caters to clinicians, technical staff, and data protection officers. It seeks to reduce technical, administrative, and legal overhead to allow studies to focus more efforts on research. It combines a cloud-native, microservice-based architecture, deidentification, and on-premises hosting to keep data sovereignty within the local institution.
  • Thumbnail Image
    Item type: Item ,
    Introduction to the Minitrack on Data Platforms and Ecosystems in Healthcare
    (2024-01-03) Fürstenau, Daniel; Thiebes, Scott; Sunyaev, Ali; Braune, Katarina