Improving Neonatal Hip Screening With Artificial Intelligence

NCT07794644 | Not Yet Recruiting FDA Device

• 1 other identifier: 124710
• Study Type: interventional
• Target: 100
• Locations: 0 countries
• Sites: N/A

Brief Summary

The goal of this trial is to pilot a portable ultrasound device that uses artificial intelligence to screen for hip dysplasia. Researchers will gather data to understand the feasibility of performing a larger trial in birthing hospitals. It will also seek to collect initial data on how well the scan compares to the standard hip screening performed soon after birth. Participants will:

• Have the portable ultrasound performed on their baby before they are discharged from hospital
• Have a diagnostic ultrasound performed on their baby at 6-weeks of age
• Complete a short questionnaire about the experience of having the measurement performed on their baby

Conditions

Developmental Dysplasia of Hip

Keywords

hip dysplasia; artificial intelligence; screening; ultrasound; infant

Outcome Measures

Primary Outcomes (4)

• Feasibility of the AI-ultrasound method as determined by a study-specific questionnaire administered to care givers at Day 1

  Caregiver perspectives will be captured via a study-specific questionnaire administered at Day 1 and this will enable determination of the feasibility of the AI-ultrasound method.

  Day 1

• Number of infants unable to be scanned with the AI ultrasound

  The proportion of infants unable to be successfully scanned with the AI-ultrasound will be calculated.

  Day 1

• Reasons for failure to obtain AI ultrasound scan as determined by the performing research assistant

  The reasons why infants were unable to be scanned as determined by the research assistant performing the AI scan will be documented as follows: Unsettled baby, technical failure, body habitus or other.

  Day 1

• Proportion of infants lost to follow-up between the AI-ultrasound and 6-week (corrected) diagnostic scan

  The proportion of infants that had an initial AI scan at Day 1 but did not return for a scan at Week 6 will be calculated.

  Week 6

Secondary Outcomes (4)

• Specificity of AI-ultrasound device as determined by comparison of the geometric measures (femoral head coverage and alpha angle) and expert reviews between the Day 1 and Week 6 scans

  Day 1, Week 6

• Sensitivity of AI-ultrasound measure determined by comparison of the geometric measures (femoral head coverage and alpha angle) and expert reviews between the Day 1 and Week 6 scans

  Day 1, Week 6

• The correlation between the alpha angle degree as reported by the AI-ultrasound analysis and expert analysis of the 6-week diagnostic ultrasound scan

  Day 1, Week 6

• The correlation between the percentage femoral head coverage as reported by the AI-ultrasound analysis and expert analysis of the 6-week diagnostic ultrasound scan

  Day 1, Week 6

Other Outcomes (2)

• Difference in specificity of standard neonatal screening and the AI-ultrasound scan to discriminate for DDH

  Day 1

• Difference in specificity of Retuve and the AI-ultrasound scan to discriminate for DDH

  Day 1

Study Arms (1)

All active participants

EXPERIMENTAL

All infants will undergo the AI-augmented ultrasound measure

Device: Artificial intelligence augmented ultrasound

Interventions

Artificial intelligence augmented ultrasound DEVICE

The hip ultrasound is performed using a handheld device (Exo Iris) that is a pocket-sized ultrasound probe and is run through an application on an IoS (Apple mobile) operation system. A real-time algorithm detects and records the anatomical landmarks.

All active participants

Eligibility Criteria

• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)

You may qualify if:

• Enrolled in the Victorian Hip Dysplasia Registry (VicHip) study
• Infant born at term (≥37 weeks gestation)
• Infant and caregiver admitted to the post-natal ward
• Caregivers indicate they are willing to attend a 6-week ultrasound
• Caregivers can provide a signed and dated informed consent form and is a legally acceptable representative capable of understanding the informed consent document and providing consent on the infant's behalf.

You may not qualify if:

• Any known congenital anomalies in the infant precluding examination of the hips
• Any known congenital neuromuscular conditions in the infant

Sponsors & Collaborators

• Murdoch Childrens Research Institute: lead
• Monash Health: collaborator
• University of Alberta: collaborator

Related Publications (5)

• McArthur A, Wichuk S, Burnside S, et al. Retuve: Automated multi-modality analysis of hip dysplasia with open source AI. Software Impacts. 2025/10/01/ 2025;26:100791. doi:https://doi.org/10.1016/j.simpa.2025.100791

  BACKGROUND

• Jaremko JL, Hareendranathan A, Bolouri SES, Frey RF, Dulai S, Bailey AL. AI aided workflow for hip dysplasia screening using ultrasound in primary care clinics. Sci Rep. 2023 Jun 7;13(1):9224. doi: 10.1038/s41598-023-35603-9.

  PMID: 37286559 BACKGROUND

• from 3D ultrasound using artificial intelligence: A two-center multi-year study. Informatics in Medicine Unlocked. 09/01 2022;33:101082. doi:10.1016/j.imu.2022.101082

  BACKGROUND

• Studer K, Williams N, Antoniou G, Gibson C, Scott H, Scheil WK, Foster BK, Cundy PJ. Increase in late diagnosed developmental dysplasia of the hip in South Australia: risk factors, proposed solutions. Med J Aust. 2016 Apr 4;204(6):240. doi: 10.5694/mja15.01082.

  PMID: 27031400 BACKGROUND

• Johnson MD, Kuschel C, Donnan L. Neonatal clinical examination and selective ultrasound screening are not reliable for the early diagnosis of hip dysplasia: A retrospective cohort study. J Paediatr Child Health. 2023 Oct;59(10):1146-1151. doi: 10.1111/jpc.16472. Epub 2023 Aug 7.

  PMID: 37545325 BACKGROUND

MeSH Terms

Conditions

Hip Dislocation

Condition Hierarchy (Ancestors)

Joint Dislocations; Joint Diseases; Musculoskeletal Diseases; Wounds and Injuries; Hip Injuries

Study Officials

• Brian Loh, MBBS BBiomedSc FRACS FAOrthoA

  Murdoch Children's Research Institute & Monash Health

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Brian Loh

CONTACT

+61383416200; brian.loh@rch.org.au

Natalie Hyde

CONTACT

+61383416200; natalie.hyde@mcri.edu.au

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: NA
• Masking: NONE
• Purpose: SCREENING
• Intervention Model: SINGLE GROUP
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Model Details: All enrolled infants will be assigned to the intervention in addition to standard care.

Study Record Dates

• First Submitted: August 5, 2026
• First Posted: August 31, 2026
• Study Start: August 1, 2026
• Primary Completion (Estimated): February 1, 2027
• Study Completion (Estimated): March 1, 2027
• Last Updated: August 31, 2026

Record last verified: 2026-08

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, ICF
• Time Frame: 12 months after the publication of data until the data is deleted (at minimum 1 year after the youngest participant turns 25 years of age).
• Access Criteria: The data must be obtained from the Murdoch Children's Research Institute. Prior to releasing any data the following are required: a data access agreement must be signed between relevant parties, the Study Management Group must see and approve the data analysis plan describing how the data will be analysed, there must be an agreement around appropriate acknowledgment and any additional costs involved must be covered. Should the Study Management Group be unavailable, this role is delegated to the Murdoch Children's Research Institute. Data will only be shared with a recognized research organisation which has approved the proposed analysis plan.