Applying Large Language Models to Enhance Data Extraction for Clinical Research

Project: Research

Project participants

Description

Acute heart failure including cardiogenic shock is a life-threatening condition with high 30-day mortality up to 60%.
In order to understand these critical conditions better, large registries are being established. These are most valuable primarily in generating hypotheses for further assessment. This assessment is usually performed in randomised controlled trials (RCTs) which permit insights into causal relationships between medical interventions (like the use of novel medical drugs or mechanical circulatory support devices like the veno-arterial extracorporeal membrane oxygenation, so-called VA-ECMO) and patient outcomes. Unfortunately, building the registries as well as performing RCTs are labour-intensive endeavours and, thus, both time-consuming and costly.
In the applicants’ opinion, leveraging existing data collected from clinical routine is of utmost importance to advance the research in understanding the critical conditions. Also, increasing the rate of patient recruitment by better embedding research-related tasks into clinical routine will reduce RCT durations necessary to obtain sufficient numbers of patients according to power calculations. On top, building larger registries from automatically collected data in a multi-centre setting will be simplified.
Applying artificial intelligence (AI) through large language models (LLMs) addresses this need: Much information of interest to the clinical researcher is contained in discharge letters from hospitals in a more or less structured way. Instead of spending labour force to collect these data, applying AI is a valuable and cost-saving alternative.
StatusActive
Period01.09.2428.02.26

Recently viewed

Publications

  1. The Population Trajectories Both of the Wild Rabbit (Oryctolagus cuniculus) and the Iberian Lynx (Lynx pardinus) in Spain
  2. The Optimal Choice of a Forwarding Agency
  3. "Der siebente Brunnen". Fred Wanders Versuch einer anderen Darstellung der Shoah in der DDR-Literatur
  4. Wir müssen reden. Roundtablegespräch zur documenta 12
  5. Thermomechanical characterization of Portevin-Le Châtelier bands in AlMg3 (AA5754) and modeling based on a modified Estrin-McCormick approach
  6. Organizational identity and its applications
  7. Quantencomputer. Taktlos
  8. Productivity and the product scope of multi-product firms:
  9. Influence of variables of the combined coagulation–Fenton-sedimentation process in the treatment of trifluraline effluent
  10. Assessing Trust by Disclosure in Online Social Networks
  11. Computer perception of constitutional (topological) symmetry:
  12. Pia und die Dinge
  13. Forest history from a single tree species perspective
  14. Root-emitted volatile organic compounds:
  15. Microstructural, mechanical and corrosion characterization of an as-cast Mg–3Zn–0.4Ca alloy for biomedical applications
  16. Measures to Safeguard and Enhance Soil-Related Ecosystem Services
  17. Im Netz der Dinge
  18. Knowledge production and distribution of higher education institutions in the sway of global development trends
  19. The universe of group representation in Germany
  20. Unfair wage perceptions and sleep: Evidence from German survey data
  21. European welfare states constructing “Unaccompanied Minors”
  22. Cost and schedule overruns in large hydropower dams
  23. Professionalisierung des Lehrerberufs