NewbornTime – Improved newborn care based on video and artificial intelligence
| ISRCTN | ISRCTN12236970 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN12236970 |
| Secondary identifying numbers | NRC 320968 |
- Submission date
- 16/02/2023
- Registration date
- 22/02/2023
- Last edited
- 09/03/2023
- Recruitment status
- Recruiting
- Overall study status
- Ongoing
- Condition category
- Pregnancy and Childbirth
Plain English Summary
Background and study aims
Approximately 3-8% of all newborns require time-critical resuscitation. NewbornTime will produce a timeline describing events and activities performed on a newborn. Accurate time of birth will be detected using artificial intelligence (AI) models from thermal videos collected in the delivery room. Activity recognition will be performed using AI in the form of deep convolutional neural networks (CNN) on thermal and visual light video from the resuscitation. The system will be designed to recognize multiple time-overlapping activities. Care will be given to make the AI models robust, reliable, general, and adaptive for use in different hospitals and settings. The timelines will be used to evaluate compliance with guidelines and identify successful resuscitation activity patterns. It can further be useful in a de-briefing and quality improvement tool.
Who can participate?
All mothers giving birth at Stavanger University hospital and their newborns
What does the study involve?
Improve newborn resuscitation based on video and artificial intelligence.
What are the possible benefits and risks of participating?
The long-term benefit of participating is that the system can provide new knowledge and a solution for quality improvement. This is an observational study and does not pose any risk to the mother or newborn. Data are stored encrypted and with strict access control.
Where is the study run from?
The project is a collaboration between the University of Stavanger (UiS), Stavanger University Hospital (SUS), Laerdal Medical and BitYoga. UiS, SUS and Laerdal have long experience in collaborative research on newborn care. They have documented promising results in detecting activities using resuscitation videos from a hospital in Tanzania. In NewbornTime the data collection will be performed at SUS. BitYoga and Laerdal will ensure smart GDPR-compliant data contracts and data platforms. UiS will develop site-adaptive AI methods for activity recognition in video. The project has been recommended by Sikt – Norwegian Agency for Shared Services in Education and Research, formerly known as NSD (number 816989).
When is the study starting and how long is it expected to run for?
September 2020 to August 2025
Who is funding the study?
1. Norwegian Research Council (NRC) (project number 320968) (Norway)
2. Helse Vest (Norway)
3. Fondation Idella (Liechtenstein)
4. Helse Campus UiS (Norway)
Who is the main contact?
Prof Kjersti Engan, kjersti.engan@uis.no (Norway)
Prof Siren Rettedal, siren.irene.rettedal@sus.no (Norway)
Contact information
Principal Investigator
Department of Electrical Engineering and Computer Science
University of Stavanger
Postbox 8600
Stavanger
4036
Norway
 ORCID ID |
0000-0002-8970-0067 |
|---|---|
| Phone | +47 92869060 |
| kjersti.engan@uis.no |
Scientific
Department of research
Stavanger University Hospital
Gerd-Ragna Bloch Thorsens gate 8
Stavanger
4019
Norway
| ORCID ID |
0000-0002-4305-141X |
|---|---|
| Phone | +47 45235742 |
| siren.irene.rettedal@sus.no |
Study information
| Study design | Observational cohort study |
|---|---|
| Primary study design | Observational |
| Secondary study design | Cohort study |
| Study setting(s) | Hospital |
| Study type | Other |
| Participant information sheet | Patient information material can be found at https://www.uis.no/en/newborntime/participant |
| Scientific title | NewbornTime – Improved newborn care based on video and artificial intelligence |
| Study acronym | NewbornTime |
| Study hypothesis | The rationale for the study is to improve newborn care using artificial intelligence (AI) for activity and event recognition taken from thermal and visual light videos in the time both during and immediately after birth |
| Ethics approval(s) | Approved 09/03/2021, Regional ethical committee west (REK Vest) (Haukelandsveien 28, 5009 Bergen, Norway; +47 55589711; rek-vest@uib.no), ref: 222455 |
| Condition | Newborn resuscitation |
| Intervention | This study investigates newborn resuscitation because birth asphyxia is a primary cause of death in newborns, and immediate resuscitation of the newborn is crucial to reduce the risk. Currently, there exists no method to automatically record the time of birth or provide objective measurements for the series of activities that took place at a newborn resuscitation. The NewbornTime project aims to utilize video recordings from births and newborn resuscitations to develop an artificial intelligence (AI)-based system, NewbornTimeline, for automatic timeline generation of birth and resuscitation activities. The timeline is put together from the time of birth and the start and stop times of resuscitation activities if any until the end of the resuscitation episode. The timeline documents what events took place so that healthcare professionals can learn; it can detect deviations and it can identify areas where there is a need for better routines or training. Resuscitation activities include stimulation, clearing airways and performing bag-mask ventilation. The system input will be based on thermal video from the delivery room and visual light and thermal video from the resuscitation table. By using artificial intelligence and video processing, we will develop algorithms to automatically and accurately detect the time of birth and resuscitation activities. The time of birth will be manually marked by a nurse by pressing a “Baby born” button on an iPad used for the research project. The resuscitation activities will be assessed retrospectively by a medical doctor watching the visual light videos and marking the start and stop of all the relevant events and activities, like ventilation, stimulation, suction, and chest compression. |
| Intervention type | Other |
| Primary outcome measure | NewbornTimeline will be evaluated through the following outcome variables: 1. Time of birth measured using the artificial intelligence-based judgement of the video recording of the labor by a thermal camera and manually recorded time as ground truth at the time of birth in the delivery room 2. The resuscitation activities and events measured using the artificial intelligence-based judgement of the visual light video recording from the resuscitation table and manually labeled videos in retrospect as ground truth at the time of resuscitation |
| Secondary outcome measures | 1. Compliance with resuscitation guidelines measured by comparing the manually labeled timelines and the AI-produced timelines with resuscitation guidelines in retrospect towards the end of the project when the AI models are working and the manual labeling is terminated 2. Successful resuscitation activity patterns measured using machine learning on the series of timelines compared with medical records on heart rate and the condition of the newborn at the end of the resuscitation in retrospect towards the end of the project 3. Number of consents, generated study IDs and collected videos of different types collected as a function of time measured using a digital patient consent handling and automated video data collection system to facilitate secure and accountable data collection throughout the data collection period |
| Overall study start date | 01/09/2020 |
| Overall study end date | 31/08/2025 |
Eligibility
| Participant type(s) | Patient |
|---|---|
| Age group | Mixed |
| Sex | Both |
| Target number of participants | The expected sample size for newborns receiving resuscitation is approximately 500. The expected sample size of women giving births is > 500. |
| Participant inclusion criteria | 1. All women giving birth at the hospital 2. All newborns requiring resuscitation |
| Participant exclusion criteria | 1. Non-consent from the mother 2. HCPs refraining from participation within 48 hours of the event 3. Birth in the labour room without cameras installed and no newborn resuscitation necessary |
| Recruitment start date | 15/11/2021 |
| Recruitment end date | 31/08/2025 |
Locations
Countries of recruitment
- Norway
Study participating centres
Stavanger
4019
Norway
Stavanger
4021
Norway
Stavanger
4002
Norway
Stavanger
4021
Norway
Sponsor information
University/education
Kjell Arholms gate 41
Stavanger
4021
Norway
| Phone | +47 51831000 |
|---|---|
| post@uis.no | |
| Website | http://www.uis.no/frontpage/ |
"ROR" |
https://ror.org/02qte9q33 |
Hospital/treatment centre
Gerd-Ragna Bloch Thorsens gate 8
Stavanger
4019
Norway
| Phone | +47 51518000 |
|---|---|
| post@sus.no | |
| Website | http://www.helse-stavanger.no/en/Sider/default.aspx |
|
"ROR" |
https://ror.org/04zn72g03 |
Funders
Funder type
Government
Government organisation / National government
- Alternative name(s)
- Forskningsrådet, Norwegian Research Council, Research Council of Norway
- Location
- Norway
Government organisation / Local government
- Alternative name(s)
- Western Norway Regional Health Authority, WNRHA, Helse Vest Regionalt Helseføretak, Helse Vest RHF
- Location
- Norway
Private sector organisation / Trusts, charities, foundations (both public and private)
- Alternative name(s)
- Foundation Idella
- Location
- Denmark
Private sector organisation / Universities (academic only)
- Alternative name(s)
- University of Stavanger, UiS, NOR
- Location
- Norway
Results and Publications
| Intention to publish date | 31/10/2025 |
|---|---|
| Individual participant data (IPD) Intention to share | No |
| IPD sharing plan summary | Not expected to be made available |
| Publication and dissemination plan | Planned publications in high-impact peer-reviewed technical and medical journals and at conferences |
| IPD sharing plan | The dataset generated during the current study is not expected to be made available due to sensitivity of data and privacy concerns. |
Study outputs
| Output type | Details | Date created | Date added | Peer reviewed? | Patient-facing? |
|---|---|---|---|---|---|
| Protocol article | 08/03/2023 | 09/03/2023 | Yes | No |
Editorial Notes
09/03/2023: Publication reference added.
21/02/2023: Trial's existence confirmed by the Regional Ethical Committee West (REK Vest) (Norway).
