Quality Considerations for Institutional Perioperative Data and AI Model Development in Airway and Circulatory Management
THE JOURNAL OF JAPAN SOCIETY FOR CLINICAL ANESTHESIA · 15.11.2025 · 45 (6) · 571-576
DOI: 10.2199/jjsca.45.571 ↗Kurz gefasst
Fachartikel in der Zeitschrift THE JOURNAL OF JAPAN SOCIETY FOR CLINICAL ANESTHESIA, erschienen am 15.11.2025. Verfasst von Toshiyuki NAKANISHI.
Abstract (Originalsprache)
In daily clinical practice, anesthesiologists record diverse information, including patient vital signs, procedural events such as intubation/extubation and surgical timestamps, as well as medication and fluid administration data. While substantial volumes of time-series data are electronically accumulated daily, only a limited portion can be effectively utilized for analysis. Recent advances in artificial intelligence and machine learning have generated considerable interest in developing high-precision predictive models using perioperative data. However, models developed from low-quality data frequently demonstrate inadequate performance when implemented in clinical settings. This article outlines key considerations for analyzing data collected within institutional settings and presents case studies where the author utilized institutional data for AI research applications. These examples demonstrate both the challenges and opportunities in leveraging perioperative data for developing clinically relevant AI models.
Zitieren
APA
NAKANISHI, T. (2025). Quality Considerations for Institutional Perioperative Data and AI Model Development in Airway and Circulatory Management. THE JOURNAL OF JAPAN SOCIETY FOR CLINICAL ANESTHESIA, 45(6), 571-576. https://doi.org/10.2199/jjsca.45.571
Vancouver
NAKANISHI T. Quality Considerations for Institutional Perioperative Data and AI Model Development in Airway and Circulatory Management. THE JOURNAL OF JAPAN SOCIETY FOR CLINICAL ANESTHESIA. 2025;45(6):571-576. doi:10.2199/jjsca.45.571
BibTeX
@article{NAKANISHI2025,
author = {NAKANISHI, Toshiyuki},
title = {Quality Considerations for Institutional Perioperative Data and AI Model Development in Airway and Circulatory Management},
journal = {THE JOURNAL OF JAPAN SOCIETY FOR CLINICAL ANESTHESIA},
year = {2025},
volume = {45},
number = {6},
pages = {571--576},
doi = {10.2199/jjsca.45.571}
}
EUVIMED