Rebooting Infant Pain Assessment: Using Machine Learning to Exponentially Improve Neonatal Intensive Care Unit Practice
- Conditions
- Acute Pain
- Registration Number
- NCT05579496
- Lead Sponsor
- York University
- Brief Summary
A multi-national multidisciplinary team will be working collaboratively to build a machine learning algorithm to distinguish between preterm infant distress states in the Neonatal Intensive Care Unit.
- Detailed Description
Unmanaged pain in hospitalized infants has serious long-term complications. Our international team of knowledge users and health/natural science/engineering/social science researchers have come together to build a machine learning algorithm that will learn how to discriminate invasive and non-invasive distress. A sample of 400 preterm infants (300 from Mount Sinai Hospital and 100 from University College London Hospital \[UCLH\]) and their mothers will be followed during a routine painful procedure (heel lance). Pain indicators (facial grimacing \[behavioural indicators\], heart rate, oxygen saturation levels \[physiologic indicators\], brain electrical activity) during the painful procedure will be used to train the algorithm to discriminate between different types of distress (pain-related and non-pain related).
Recruitment & Eligibility
- Status
- RECRUITING
- Sex
- All
- Target Recruitment
- 400
- parents of a child currently in the NICU or
- health professionals currently working in the NICU.
- Participants who cannot communicate fluently in English
- QUANTITITATIVE DATA CAPTURE (video, eeg, ecg, SPo2)
- Inclusion Criteria:
- Infants born between 28 0/7 weeks 32 6/7 weeks gestational age
- Infants who are within 6 weeks postnatal age
- Infants who are undergoing a routine heel lance
- Exclusion Criteria:
- Infants with congenital malformations
- Infants receiving analgesics or sedatives at the time of study (aside from sucrose),
- Infants with history of perinatal hypoxia/ischemia at the time of study.
- Infants with diaper rash or excoriated buttocks
Study & Design
- Study Type
- OBSERVATIONAL
- Study Design
- Not specified
- Primary Outcome Measures
Name Time Method Cortical Correlate of Distress For 2 hours surrounding Painful procedure (time locked to heel lance; approximately 1 hour before to 1 hour after heel lance) To be analyzed using machine learning via bedside monitoring: Continuous EEG data capture
Oxygen Saturation Correlate of Distress Over 2 hours surrounding Painful procedure (time locked to heel lance; approximately 1 hour before to 1 hour after heel lance) To be analyzed using machine learning via bedside monitoring: amount of oxygen-carrying hemoglobin in the blood relative to the amount of hemoglobin not carrying oxygen
Cardiac Correlates of Distress Over 2 hours surrounding Painful procedure (time locked to heel lance) To be analyzed using machine learning via bedside monitoring: Heart Rate, Heart Rate Variability
Behavioural Correlate of Distress NFCS-P coded in 1-5 minute epochs, over 2 hour surrounding painful procedure (time locked to heel lance; approximately 1 hour before to 1 hour after heel lance) To be analyzed using machine learning via bedside videography: Facial Grimacing using Neonatal Facial Coding System(NFCS-P subset; Bucsea et al., in preparation)
- Secondary Outcome Measures
Name Time Method Semi-Structured Interview These interviews are occurring at the beginning of the study and will be qualitatively analyzed. They are not linked to infants whose data we are collecting primary outcomes. Health Professionals and Caregivers will be asked about their thoughts on using AI for infant pain assessment
Trial Locations
- Locations (2)
University College London Hospital
🇬🇧London, No Province, United Kingdom
Mount Sinai Hospital
🇨🇦Toronto, Ontario, Canada