Assessing AI-enhanced portable ultrasound screening for infant hip dysplasia
| ISRCTN | ISRCTN33985096 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN33985096 |
- Submission date
- 11/07/2024
- Registration date
- 23/07/2024
- Last edited
- 23/07/2024
- Recruitment status
- Recruiting
- Overall study status
- Ongoing
- Condition category
- Musculoskeletal Diseases
Plain English summary of protocol
Background and study aims
About 1% of children are born with a loose, abnormally shaped hip joint, known as a dysplastic hip. Early detection in infancy allows for successful treatment with a harness or brace. If missed, there is a high risk of developing early osteoarthritis. Current clinical tests for hip dysplasia primarily detect severe cases, and traditional ultrasound screening has several limitations: it can be overly sensitive to minor shape changes, measurements on images are unreliable, and it is challenging for technicians to perform on squirming infants.
Early tests indicate that the 3D shape of the hip joint can be easily scanned using the same 3D ultrasound probe commonly used for fetal scans. These 3D images allow for optimal viewing angles without relying on the technician’s skill. 3D sequences are now included in some routine scans in Edmonton, and pseudo-3D models can also be created from video 'sweep' images recorded during conventional 2D ultrasound scans.
3D hip models are being developed based on existing ultrasound data. This study aims to further test and develop 3D hip ultrasound by determining the severity of hip dysplasia (if any) at each child's first visit, and by following each child's hip development and outcomes over several years. The researchers believe the 3D method will be easier to use at the bedside, measure hips more accurately, and provide valuable information about abnormal hip shapes. They are also testing artificial intelligence (AI) to automatically detect abnormal hips from 3D ultrasound or 2D "sweep" images. If successful, these methods could be applied to handheld portable ultrasound devices used by lightly trained personnel, such as primary care clinic nurses, potentially enabling widespread DDH screening.
Who can participate?
Infants aged from newborn to 12 weeks born in Alberta attending wellness checks (around 6-8 weeks) at participating sites
What does the study involve?
Each participant will undergo an initial 3D ultrasound scan to screen for hip dysplasia. Follow-up assessments will be conducted at regular intervals over a 5-year period to monitor hip development and clinical outcomes. All assessments involve non-invasive ultrasound imaging and clinical evaluations.
What are the possible benefits and risks of participating?
The potential benefits include more accurate and reliable detection of hip dysplasia, which can lead to early and effective treatment. There are minimal risks associated with the study as it involves non-invasive ultrasound imaging.
Where is the study run from?
University of Alberta (Canada)
When is the study starting and how long is it expected to run for?
July 2012 to December 2032
Who is funding the study?
1. Arthritis Society Canada (Canada)
2. TD Better Health Program (Canada)
3. Women and Children's Health Research Institute (WCHRI) (Canada)
4. Alberta Machine Intelligence Institute (AMII) (Canada)
Who is the main contact?
Dr Jacob Jaremko, jjaremko@ualberta.ca
Contact information
Principal Investigator
8303 112 St NW
Edmonton
T6G 2T4
Canada
 ORCID ID |
0000-0001-5314-2297 |
|---|---|
| Phone | +1 (0)780 493 2773 |
| jjaremko@ualberta.ca |
Scientific
8303 112 St NW
Edmonton
T6G 2T4
Canada
| ORCID ID |
0000-0001-6201-9258 |
|---|---|
| Phone | +1 (0)780 493 2773 |
| hareendr@ualberta.ca |
Public
8303 112 St NW
Edmonton
T6G 2T4
Canada
| Phone | +1 (0)780 493 2773 |
|---|---|
| cgallan1@ualberta.ca |
Study information
| Study design | Multicenter prospective longitudinal observational study |
|---|---|
| Primary study design | Observational |
| Secondary study design | Longitudinal study |
| Study setting(s) | Community, GP practice, Hospital |
| Study type | Diagnostic, Prevention, Screening, Efficacy |
| Participant information sheet | 45772_PIS.pdf |
| Scientific title | Assessment of developmental dysplasia of the hip screening in infants by AI-enhanced 3D point-of-care ultrasound: correlation with clinical and imaging findings at initial infant screening and follow-up |
| Study objectives | Three-dimensional (3D) ultrasound, either obtained with 3D probe or by cine sweep with conventional 2D probe, enhanced with artificial intelligence (AI), could improve the accuracy and reliability of developmental dysplasia of the hip (DDH) screening in infants, compared to current single-image two-dimensional (2D) ultrasound methods, by providing more comprehensive imaging and enabling automated diagnosis even when performed by minimally trained personnel. |
| Ethics approval(s) |
Approved 01/10/2012, University of Alberta Health Research Ethics Board (116 St & 85 Ave, Edmonton, T6G 2R3, Canada; +1 (0)780 492 3111; reoffice@ualberta.ca), ref: MS23_Pro00032107 |
| Ethics approval additional information | The University of Alberta Human Research Ethics Board granted a waiver of consent for this project, given that ultrasound is a harmless imaging modality, but the researchers still included a brief assent discussion with each family prior to performing scans. |
| Health condition(s) or problem(s) studied | Diagnostic screening for developmental dysplasia of the hip (DDH) in infants |
| Intervention | Participants undergo initial screening for developmental dysplasia of the hip (DDH) using AI-enhanced 3D ultrasound (by 3D probe and/or cine sweep) at well-baby checkups (6-10 weeks old). All data collection is non-invasive, with ultrasound imaging and clinical evaluations conducted during routine pediatric visits. |
| Intervention type | Device |
| Pharmaceutical study type(s) | Not Applicable |
| Phase | Phase I/II |
| Drug / device / biological / vaccine name(s) | Exo Iris (Exo Inc., Santa Clara, CA, USA) |
| Primary outcome measure | Diagnostic accuracy of DDH screening (using the US/AI model) against gold-standard conventional clinical ultrasound diagnosis, as assessed by sensitivity, specificity, positive and negative predictive value and AUC at the time of baseline ultrasound |
| Secondary outcome measures | 1. Implementation success, assessed using the Consolidated Framework for Implementation Research (CFIR) 5-point framework, measured starting at initial rollout (2021), and every 3 years thereafter until the conclusion of recruitment (2024, 2027, 2030) 2. Feasibility of DDH screening (using the US/AI model) in neonates vs conventional physical assessments conducted by a physician at baseline, assessed by.System Usability Score (SUS), qualitative analysis of user and patient/family feedback from semi-structured interviews, and economic analysis. This is assessed every 3 years as above, based on data collected continuously and reviewed annually. |
| Overall study start date | 01/07/2012 |
| Completion date | 31/12/2032 |
Eligibility
| Participant type(s) | Healthy volunteer |
|---|---|
| Age group | Mixed |
| Lower age limit | 0 Days |
| Upper age limit | 24 Weeks |
| Sex | Both |
| Target number of participants | 20,000 |
| Key inclusion criteria | 1. All infants in Alberta that can be reached with the technology 2. Age range from newborn to 12 weeks |
| Key exclusion criteria | Does not meet the inclusion criteria |
| Date of first enrolment | 01/01/2020 |
| Date of final enrolment | 31/12/2030 |
Locations
Countries of recruitment
- Canada
Study participating centres
Spruce Grove
T7X 3C7
Canada
Spruce Grove
T7X 2V2
Canada
Red Deer
T4R 2P7
Canada
Fort McMurray
T9H 3S5
Canada
Wetaskiwin
T9A 3N3
Canada
Maskwacis
T0C 1N0
Canada
Siksika
T0J 3W0
Canada
Red Deer
T4N 4E7
Canada
Edmonton
T6G 2B7
Canada
Sponsor information
University/education
2-13 Heritage Medical Research Centre
Edmonton
T6G 2R7
Canada
| Phone | +1 (0)780 918 0654 |
|---|---|
| robin.love@ualberta.ca | |
| Website | http://ualberta.ca/ |
"ROR" |
https://ror.org/0160cpw27 |
Funders
Funder type
Charity
Private sector organisation / Associations and societies (private and public)
- Alternative name(s)
- Société de l’arthrite
- Location
- Canada
Government organisation / For-profit companies (industry)
- Alternative name(s)
- TD Bank, N.A.
- Location
- United States of America
Private sector organisation / Other non-profit organizations
- Alternative name(s)
- Women & Children's Health Research Institute, WCHRI
- Location
- Canada
Government organisation / National government
- Alternative name(s)
- AMII
- Location
- Canada
Results and Publications
| Intention to publish date | 01/09/2026 |
|---|---|
| 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 journals, including Implementation Science, Pediatric Imaging, Bone & Joint, Computers in Biology and Medicine, Journal of Ultrasound, and Journal of Pediatric Orthopaedics. Planned dissemination into high-impact conferences, including the Pediatric Orthopaedic Society of North America (POSNA), the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), and the Radiological Society of North America (RSNA). Further dissemination will include sharing reported outcomes of DDH screening using the US/AI tool with Alberta Health Services (AHS). |
| IPD sharing plan | The datasets generated during and/or analysed during the current study are not expected to be made available due to the conditions imposed by our data transfer agreements. Additional information related to the data and analysis are available from the corresponding author on reasonable request. |
Study outputs
| Output type | Details | Date created | Date added | Peer reviewed? | Patient-facing? |
|---|---|---|---|---|---|
| Participant information sheet | 22/07/2024 | No | Yes |
Additional files
Editorial Notes
12/07/2024: Study's existence confirmed by the University of Alberta.
