Deep learning model predicts pulse pressure variation from peripheral vein waveforms
Tests a computer model reading peripheral vein waveforms in adults receiving mechanical ventilation.
Plain-language title generated from the registry record
Registry title: Deep Learning Model for Predicting a Peripheral Venous Waveform-based Pulse Pressure Variation
Key facts
- Where
- Seongnam-si, South KoreaSee the site
Could this be for you?
Answer for yourself. This is not a decision: only the study team can say whether you can take part.
- Are you between 19 and 80 years old?
Source: NCT06734650, who can take part
Where in South Korea
- Seoul National University Bundang HospitalSeongnam-si, South KoreaNo longer taking part
Questions to ask the study team
- How often do I need to come in, and for how long each time?
- Are travel costs paid?
- Can I keep my own doctor?
- What happens if I want to stop?
Trial details
- Who runs it
- Seoul National University Bundang Hospital
- Trial code
- NCT06734650
- Record updated
- 16 December 2024
- Original description
Pulse pressure variation is a monitoring index that indicates the response to fluid therapy in patients receiving mechanical ventilation, and is used as a reference for patients with unstable hemodynamic conditions. However, it is invasive because it requires arterial puncture to collect it. In a previous study by the investigators, the investigators developed and verified an artificial intelligence model that predicts stroke volume variation, in real time using only the central venous pressure waveform. However, since a large vein such as the jugular vein must be punctured to collect the central venous pressure waveform, it is still invasive, and its clinical utility is low. Therefore, in this study, the investigators collected waveforms from peripheral veins that are less invasive and can be a wide range of applications because all surgical patients have them. The investigators aimed to develop and verify an artificial intelligence model that predicts pulse pressure variation obtained from peripheral venous waveforms .Read full original descriptionShow less
Development and Validation of a Peripheral Venous Waveform-based Pulse Pressure Variation Calculating Deep Learning Model
Pulse pressure variation is a monitoring index that indicates the response to fluid therapy in patients receiving mechanical ventilation, and is used as a reference for patients with unstable hemodynamic conditions. However, it is invasive because it requires arterial puncture to collect it. In a previous study by the investigators, the investigators developed and verified an artificial intelligence model that predicts stroke volume variation, in real time using only the central venous pressure waveform. However, since a large vein such as the jugular vein must be punctured to collect the central venous pressure waveform, it is still invasive, and its clinical utility is low. Therefore, in this study, the investigators collected waveforms from peripheral veins that are less invasive and can be a wide range of applications because all surgical patients have them. The investigators aimed to develop and verify an artificial intelligence model that predicts pulse pressure variation obtained from peripheral venous waveforms .
Inclusion: Patients who voluntarily agreed and signed the written informed consent form before participating in this study; Adult aged 19 years or older; American Society of Anesthesiologists physical class (ASA) 1-3; Patients scheduled for elective hepatectomy under general anesthesia; Patients who require arterial pressure monitoring and additional peripheral venous access for routine anesthesia preparation; Non-smokers with normal pulmonary function.
Exclusion: Patients with abnormal findings on electrocardiogram before surgery; Patients who cannot undergo peripheral venous puncture.