EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

MIMIC-BP: A curated dataset for blood pressure estimation

Ivandro Sanches iD, Victor V Gomes iD, Carlos Caetano, Lizeth S B Cabrera, Vinicius H Cene iD, Thomas Beltrame iD, Wonkyu Lee, Sanghyun Baek, Otávio A B Penatti iD

Scientific Data · 15.11.2024 · 11 (1)

DOI: 10.1038/s41597-024-04041-1 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Scientific Data, online erschienen am 15.11.2024. Verfasst von Ivandro Sanches und 8 weiteren Autor:innen. Beteiligte Einrichtung: AI R&D Team, Samsung R&D Institute Brazil (SRBR), Campinas, São Paulo, 13097-160, Brazil. ivandro.s@samsung.com. Hauptthemen laut MEDLINE: Deep Learning, Blood Pressure, Blood Pressure Determination, Photoplethysmography. Die Arbeit wurde bisher 33-mal zitiert (Crossref) und nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Blood pressure (BP) is one of the most prominent indicators of potential cardiovascular disorders. Traditionally, BP measurement relies on inflatable cuffs, which is inconvenient and limit the acquisition of such important health-related information in general population. Based on large amounts of well-collected and annotated data, deep-learning approaches present a generalization potential that arose as an alternative to enable more pervasive approaches. However, most existing work in this area currently uses datasets with limitations, such as lack of subject identification and severe data imbalance that can result in data leakage and algorithm bias. Thus, to offer a more properly curated source of information, we propose a derivative dataset composed of 380 hours of the most common biomedical signals, including arterial blood pressure, photoplethysmography, and electrocardiogram for 1,524 anonymized subjects, each having 30 segments of 30 seconds of those signals. We also validated the proposed dataset through experiments using state-of-the-art deep-learning methods, as we highlight the importance of standardized benchmarks for calibration-free blood pressure estimation scenarios.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningBlood PressureBlood Pressure DeterminationPhotoplethysmographyHumansElectrocardiographyMale

Autor:innen und Institutionen

  • AI R&D Team, Samsung R&D Institute Brazil (SRBR), Campinas, São Paulo, 13097-160, Brazil. ivandro.s@samsung.com.
  • AI R&D Team, Samsung R&D Institute Brazil (SRBR), Campinas, São Paulo, 13097-160, Brazil.
  • Health H/W R&D Group, Samsung Electronics Co Ltd, Suwon, 497335, South Korea.
  • AI R&D Team, Samsung R&D Institute Brazil (SRBR), Campinas, São Paulo, 13097-160, Brazil. o.penatti@samsung.com.

ORCID-Profile: Ivandro Sanches, Victor V Gomes, Vinicius H Cene, Thomas Beltrame, Otávio A B Penatti

Zitieren

APA

Sanches, I., Gomes, V. V., Caetano, C., Cabrera, L. S. B., Cene, V. H., Beltrame, T., Lee, W., Baek, S., & Penatti, O. A. B. (2024). MIMIC-BP: A curated dataset for blood pressure estimation. Scientific Data, 11(1). https://doi.org/10.1038/s41597-024-04041-1

Vancouver

Sanches I, Gomes VV, Caetano C, Cabrera LSB, Cene VH, Beltrame T, et al. MIMIC-BP: A curated dataset for blood pressure estimation. Scientific Data. 2024;11(1). doi:10.1038/s41597-024-04041-1

BibTeX
@article{Sanches2024,
  author = {Sanches, Ivandro and Gomes, Victor V and Caetano, Carlos and Cabrera, Lizeth S B and Cene, Vinicius H and Beltrame, Thomas and Lee, Wonkyu and Baek, Sanghyun and Penatti, Otávio A B},
  title = {MIMIC-BP: A curated dataset for blood pressure estimation},
  journal = {Scientific Data},
  year = {2024},
  volume = {11},
  number = {1},
  doi = {10.1038/s41597-024-04041-1}
}
Kostenlos starten →