In Brief | Artificial Intelligence | Paediatrics & Neonatology
The Feasibility of Monitoring Neonatal Arousal States with AI-Based Monitoring: A Study Overview
Time to read: 02:51
Time to listen: 06:31
Originally Published: 23 January 2025
Source: JAMA Network Open
Type of article: In Brief
MedED Catalogue Reference: MAIB002
Category: Artificial Intelligence
Cross Reference: Paediatrics & Neonatology
Keywords: AI, neonatology, NICU, Monitoring
Key Takeaway
This article is a review of recent studies originally published in the JAMA Network Open,25 January 2025. This article does not represent the original research, nor is it intended to replace the original research. Access the full Disclaimer Information.
Watch the 60 Second Summary
(Please let us know if you like this format by clicking here)
Listen to Full Article
The neonatal period is marked by the most dramatic physiological changes in human life. It is also a period that carries significant risk for the infant.
Neonatal mortality rates (NMR) vary significantly by region, with the highest rates in low- and middle-income countries and the lowest in high-income countries.3
It is estimated that a child born in Sub-Saharan Africa is 10 times more likely to die in the first month of life than a child born in a high-income country. In 2020, for example, Australia reported an NMR of 2.4 per 1,000 live births, while Sub-Saharan Africa recorded 27 per 1,000 live births. 2,3
Assessing newborn developmental function is crucial for identifying conditions of risk and predicting long-term outcomes.
A key component of this assessment is the state system, a fundamental developmental process that reflects an infant’s level of arousal and ability to respond to stimuli.
The state system is a fundamental component of newborn neurodevelopmental function. It refers to the complex biochemical interactions involving multiple brain regions, cellular mechanisms, and immune responses, and reflects the newborn's underlying neurologic and behavioral competence.1
Various physiological and environmental factors influence the state, including hunger, hydration, sleep cycle stage, illness, and external stimuli, such as noise in a neonatal intensive care unit (NICU).
The Brazelton Neonatal Behavioral Assessment Scale (BNAS) categorizes neonatal state into six levels:2
Manually assessing these states is labour-intensive and impractical in clinical settings, and yet their clinical significance cannot be overstated.
Study Purpose
The researchers of this 2025 study investigated the potential of artificial intelligence (AI)–based video monitoring using a smartphone as a more efficient method for state assessment.
Study Methodology
Given that the head and hands are the most visible body parts in NICU settings,the researchers developed a model to detect and track these specific areas.
Three NICU nurses annotated the neonatal state using the Neonatal Behavioral Assessment Scale (NBAS). The nurses used a visual analogue scale to event-code the segments corresponding to the same arousal state and time segments were included only when at least two out of three nurses agreed on the state
The median value, which reflects the agreement of at least two nurses, was used to represent the arousal state for each second.
The body detection model was then applied to the videos, and the mean speed and variance for each second were extracted.
Each video segment was then classified based on their labeled states into three categories: 1 or 2, 3 or 4, and 5 or 6. Forty-nine-time points were included in the final State analysis.
Findings
Specifically, using the technology, the researchers were able to record that:
Movement speed was significantly higher in states 3 (quiet alert) and 4 (active alert) compared to other states, indicating increased activity during wakefulness without crying
?Variability in movement direction was highest in states 5 (crying) and 6 but did not show significant changes across developmental stages.
Discussion
This study highlights the feasibility of AI-based video monitoring as an efficient and objective alternative to manual state assessment.
However, the researchers noted that the increase in movement detected by AI may be influenced by other developmental transitions, such as the shift from preterm to writhing movements. Furthermore, the small sample size (27 neonates) limits generalisability, and further larger studies would be necessary to confirm their findings.
Conclusion
The study demonstrates that smartphone videos can accurately assess neonatal arousal states, enabling practical, scalable, and cost-effective neonatal monitoring in clinical settings
In resource-limited settings such as South Africa, where high-risk neonates require close monitoring, AI-based smartphone video assessments may offer a low-cost, accessible alternative to traditional methods. The ubiquity of smartphones makes this approach particularly applicable in under-resourced hospitals and rural healthcare facilities, supporting improved neonatal care.
References
1. Doherty TM, Hu A, Salik I. Physiology, Neonatal. [Updated 2023 Apr 24]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 Jan-. Available from: https://www.ncbi.nlm.nih.gov/books/NBK539840/
3. VandenBerg, K. A. (2007). State systems development in high-risk newborns in the neonatal intensive care unit: Identification and management of sleep, alertness, and crying. Journal of Perinatal & Neonatal Nursing, 21(2), 130–139. Wolters Kluwer Health | Lippincott Williams & Wilkins. https://nidcap.org/wp-content/uploads/2013/12/VandenBerg-2007-State-systems.pdf
3. Goga, A., Feucht, U., Zar, H. J., Vanker, A., etal. (2019). Neonatal, infant and child health in South Africa: Reflecting on the past towards a better future. South African medical journal , 109(11b), 83–88
Disclaimer
This article is compiled from several resources researched and compiled by the contributor. It is in no way presented as an original work. Every effort has been made to attribute quotes and content correctly. Where possible all information has been independently verified. The Medical Education Network bears no responsibility for any inaccuracies which may occur from the use of third-party sources. If you have any queries regarding this article contact us
Fact-checking Policy
The Medical Education Network makes every effort to review and fact-check the articles used as source material in our summaries and original material. We have strict guidelines in relation to the publications we use as our source data, favouring peer-reviewed research wherever possible. Every effort is made to ensure that the information contained here is an accurate reflection of the original material. Should you find inaccuracies, out of date content or have any additional issues with our articles, please make use of the contact us form to notify us.
