EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ChatGPT-based biological and psychological data imputation

Anam Nazir, Muhammad Nadeem Cheeema, Ze Wang

Meta-Radiology · 2023-11 · 1 (3) · 100034

DOI: 10.1016/j.metrad.2023.100034 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Meta-Radiology, 2023-11. Verfasst von Anam Nazir und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIA NIH HHS, NIBIB NIH HHS, NCATS NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Missing data are a common problem for large cohort or longitudinal research and have been handled through data imputation. Based on simplified models such as linear or nonlinear interpolations, current imputation methods may not be accurate for real-life data such as biological and behavioral data. The purpose of this work was to explore the capability of ChatGPT, a powerful Large Language Model (LLM) developed by OpenAI, for biological and psychological data imputation. We tested the feasibility using data from the Human Connectome Project. Performance was evaluated by comparing the imputed data against known ground truth (GT) and measured with metrics like Pearson correlation coefficient (r), relative accuracy (MP), and mean absolute error (MAE). Comparative analyses with traditional imputation techniques are also conducted to demonstrate the superior efficacy of the ChatGPT as a data imputer. In summary, through customized data-to-text prompting engineering, ChatGPT can successfully capture intricate patterns and dependencies within biological data, resulting in precise imputations. Fine-tuning ChatGPT with domain-specific biological vocabulary with human in-loop as an interpreter enhances the accuracy and relevance of the imputations.

Schlagwörter

Stichwörter der Autor:innen

Biological data imputationChatGPTLarge language modelMedical ImagingMissing values

Autor:innen und Institutionen

  • Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine.

Förderung

  • NIA NIH HHS
  • NIBIB NIH HHS
  • NCATS NIH HHS
  • University of Maryland, Baltimore County
  • National Institutes of Health
  • Institute for Clinical and Translational Research, University of Wisconsin, Madison

Zitieren

APA

Nazir, A., Cheeema, M. N., & Wang, Z. (2023). ChatGPT-based biological and psychological data imputation. Meta-Radiology, 1(3), 100034. https://doi.org/10.1016/j.metrad.2023.100034

Vancouver

Nazir A, Cheeema MN, Wang Z. ChatGPT-based biological and psychological data imputation. Meta-Radiology. 2023;1(3):100034. doi:10.1016/j.metrad.2023.100034

BibTeX
@article{Nazir2023,
  author = {Nazir, Anam and Cheeema, Muhammad Nadeem and Wang, Ze},
  title = {ChatGPT-based biological and psychological data imputation},
  journal = {Meta-Radiology},
  year = {2023},
  volume = {1},
  number = {3},
  pages = {100034},
  doi = {10.1016/j.metrad.2023.100034}
}
Kostenlos starten →