ClinicalTrials.gg
Not yet recruitingNCT07941700Updated Nov 17, 2026

AI-Guided eFAST After Liver Surgery

An interventional study of Real-time AI guidance for eFAST and Nurse-performed eFAST with AI guidance in Hemoperitoneum, Ascites and Intra-Abdominal Fluid Collection, sponsored by Oslo University Hospital. Not yet recruiting at 1 site in Norway. Open to participants aged 18 Years and older. Per ClinicalTrials.gov, last updated 2026-11-17.

Sponsored by Oslo University Hospital · Not applicable, Interventional, and Diagnostic

Updated Nov 17, 2026Newly registeredGo to Updates ↓

Help build ClinicalTrials.gg — short surveys, remote interviews or testing, in your language. Take part →

Phase
Not applicable
Study type
Interventional
Enrollment
30
Allocation
Not applicable
Ages
18 Years and older
Sex
All
01

Study summary

This study evaluates whether nurses with real-time artificial intelligence (AI) guidance, and without a remote expert, can perform extended Focused Assessment with Sonography in Trauma (eFAST) with diagnostically usable image quality in patients who have recently undergone liver surgery. Each patient is examined twice: first by a nurse using a handheld ultrasound probe with AI guidance, then by a radiologist as reference standard. The study measures image quality, agreement on free fluid between the nurse and the reference standard, and examination duration.

Read the detailed description

In the precursor study (NCT06594562), nurses performed eFAST after liver surgery with real-time guidance from a remote expert. This study replaces the remote expert with real-time AI guidance.

Design: prospective single-arm diagnostic agreement study at Oslo University Hospital, Rikshospitalet, target 30 patients. Each patient is examined twice. A ward nurse performs AI-guided eFAST with a Clarius handheld probe (index test). An experienced radiologist performs an independent eFAST within one hour (reference standard). CT replaces the radiologist's eFAST as reference when clinically indicated and performed within 24 hours.

AI model: developed by SINTEF, trained on recordings from the precursor study, and locked before enrollment. It provides real-time image recognition to guide the operator.

Analysis: the abdomen (right and left upper quadrant, suprapubic) is analysed per patient and per window. Results are compared descriptively with the precursor cohort. Reporting follows STARD-AI and DECIDE-AI. Enrollment ends at 30 patients or after three months.

02

Conditions studied

  • Hemoperitoneum
  • Ascites
  • Intra-Abdominal Fluid Collection
  • Postoperative Complications

Keywords

  • Artificial Intelligence
  • eFAST
  • Point-of-Care Ultrasound
  • Nurse-Performed Ultrasound
  • Handheld Ultrasound
03

In context

Ascites

248 studies on the registry are indexed under Ascites; 57 are open to participants now.

This study's planned enrollment of 30 is below the median of 54 across 180 interventional studies indexed under Ascites.

Browse Ascites studies →

Lead sponsor

Oslo University Hospital is the lead sponsor of 835 studies on the registry; 150 are open to participants now.

Counted across the registry records on this site, refreshed daily.

04

Who can participate

Ages eligible
18 Years and older
Sexes eligible
All
Accepts healthy volunteers
No

Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.

Inclusion criteria

  • Adults (18 years or older) admitted for postoperative monitoring after laparoscopic liver surgery at Oslo University Hospital, Rikshospitalet
  • Examined within 72 hours after surgery
  • Written

Exclusion criteria

Exclusion Criteria:

  • Allergy to ultrasound gel
  • Conversion to open surgery
  • High risk of contact transmission of resistant pathogens (ESBL, MRSA, VRE)

Operators are five or more ward nurses with limited point-of-care ultrasound experience, who give separate consent to registration of performance data. Reference examinations are performed by experienced radiologists at Oslo University Hospital.

Terms in these criteria, in plain language(1)

Underlined terms in the criteria open these too. Explained by clinicaltrials.gg, not part of the registry record. The study team decides who can take part.

Informed consent

Before you join, the study team explains what taking part involves — its purpose, what happens, the risks and possible benefits — and answers your questions, so you can decide freely. Consent is usually recorded on a form (the informed consent form, ICF) signed by you or, where the law allows, by someone legally authorized to decide for you.

You can keep asking questions during the study, you are told of new information that matters, and you can choose to leave at any time.

05

Study design

Phase
Not applicable
Primary purpose
Diagnostic
Allocation
Not applicable
Intervention model
Single group
Masking
None (open label)
Enrollment
30 participants (estimated)

Study arms

  • Experimental
    AI-guided nurse-performed eFAST

    All enrolled patients receive an AI-guided eFAST by a ward nurse, followed by a reference examination by a radiologist, or CT when clinically indicated.

    Device: Real-time AI guidance for eFAST · Diagnostic Test: Nurse-performed eFAST with AI guidance · Diagnostic Test: Reference standard: radiologist eFAST or CT

Interventions

  • DeviceReal-time AI guidance for eFAST

    Software model developed by SINTEF, running on a laptop connected to a Clarius wireless handheld ultrasound probe. Provides real-time structural image recognition to guide the operator.

  • Diagnostic testNurse-performed eFAST with AI guidance

    Index test. eFAST performed by a ward nurse with limited ultrasound experience, guided in real time by the AI model, without a remote expert. Windows: right and left upper quadrant, suprapubic (sagittal and transverse), subxiphoid pericardial and bilateral lung. The nurse records a yes/no free fluid assessment per window.

  • Diagnostic testReference standard: radiologist eFAST or CT

    Independent bedside eFAST performed by an experienced radiologist within one hour after the index test, recorded yes/no per window. CT is the reference when clinically indicated and performed within 24 hours of the ultrasound examinations, and takes precedence when CT and radiologist eFAST disagree. The radiologist also scores image quality of the nurse's stored clips.

06

What researchers measure

Primary outcomes

  1. Image quality per abdominal eFAST window

    Likert scale 1 to 5 per window (right upper quadrant, left upper quadrant, suprapubic), scored by the reference radiologist on stored clips from the nurse's examination, based on how confidently the clip can be interpreted. Same scale as in the precursor study (NCT06594562). Threshold for interpretation: median of 3 or higher per window.

    Time frame: Within 1 hour after the index examination

  2. Agreement on free fluid in the abdomen

    Percent agreement (concordance rate, p\_o) between the nurse's AI-assisted yes/no assessment and the reference standard (radiologist eFAST or CT), per patient. The abdomen is positive if fluid is seen in at least one of the right upper quadrant, left upper quadrant or suprapubic windows. Threshold for interpretation: 80 percent or higher.

    Time frame: Within 24 hours after the index examination (reference eFAST within 1 hour, CT within 24 hours)

  3. Examination duration

    Time from probe on patient to end of examination, minutes:seconds, reported as median with interquartile range. Threshold for interpretation: median of 10 minutes or less.

    Time frame: During the index examination, up to 1 hour

Secondary outcomes

  1. Chance-corrected agreement on free fluid

    Gwet's AC1 and Cohen's kappa (unweighted) between nurse and reference standard, for the abdomen and per abdominal window, with bootstrap confidence intervals. Percent agreement per abdominal window is also reported.

    Time frame: Within 24 hours after the index examination

  2. Diagnostic accuracy of AI-guided nurse eFAST

    Sensitivity, specificity, positive and negative predictive value, and positive and negative percent agreement against the reference standard, with Wilson 95 percent confidence intervals.

    Time frame: Within 24 hours after the index examination

  3. Proportion of diagnostic windows

    Proportion of windows where the nurse obtained the key anatomical structure, reported per eFAST window.

    Time frame: During the index examination, up to 1 hour

  4. Cognitive workload (NASA-TLX)

    NASA Task Load Index, 0 to 20 scale, completed by the nurse.

    Time frame: Immediately after each index examination

  5. User experience

    Operator confidence questionnaire (10 items, adapted from Hurd et al. 2025) and System Usability Scale (0 to 100), completed by the nurse.

    Time frame: Immediately after each index examination

  6. Change in examination duration over successive examinations

    Learning curve. Change in examination duration (minutes:seconds) with examination number per nurse, analysed with a linear mixed-effects model.

    Time frame: Through study completion, an average of 3 months

  7. Change in image quality over successive examinations

    Learning curve. Change in image quality score with examination number per nurse, analysed with a linear mixed-effects model. Image quality is scored on a Likert scale from 1 to 5, where higher scores mean better image quality.

    Time frame: Through study completion, an average of 3 months

  8. Change in NASA Task Load Index score over successive examinations

    Learning curve. Change in NASA Task Load Index (NASA-TLX) score with examination number per nurse, analysed with a linear mixed-effects model. The score ranges from 0 to 20, where higher scores mean higher workload.

    Time frame: Through study completion, an average of 3 months

Other outcomes

  1. Image quality, subxiphoid pericardial window (exploratory)

    Likert scale 1 to 5, scored as for the primary image quality outcome. Threshold for interpretation: median of 3 or higher.

    Time frame: Within 1 hour after the index examination

07

Study locations

1 site
08

References and documents

Publications

  • Hartline CPTAD, Hartvickson MAJS, Perdue CPTMJ, Sandoval CPTC, Walker LTCJD, Soules CPTA, Mitchell COLCA. Army Medic Performance in Trauma Sonography: The Impact of Artificial Intelligence Assistance in Focused Assessments With Sonography in Trauma-A Prospective Randomized Controlled Trial. Mil Med. 2025 Nov 1;190(11-12):e2376-e2381. doi: 10.1093/milmed/usaf215. PubMed 40886703 ↗
  • Sounderajah V, Guni A, Liu X, Collins GS, Karthikesalingam A, Markar SR, Golub RM, Denniston AK, Shetty S, Moher D, Bossuyt PM, Darzi A, Ashrafian H; STARD-AI Steering Committee. The STARD-AI reporting guideline for diagnostic accuracy studies using artificial intelligence. Nat Med. 2025 Oct;31(10):3283-3289. doi: 10.1038/s41591-025-03953-8. Epub 2025 Sep 15. PubMed 40954311 ↗

Individual participant data

Plan to share: No — Individual participant data are stored in Oslo University Hospital's approved research storage. Public sharing is not covered by the current data protection approvals.

09

Timeline

1 registry update since Sep 25, 2026
ElapsedEstimatedEstimated, not yet reachedSimilar studies: half by the dot; the band is the middle halfHalf of similar studies reached it by the dot; the band is the middle half, from a quarter to three quarters

Start is estimated for Dec 2026; primary completion for Mar 2027, 3 months later.

Of 33 interventional studies with no phase under Ascites, half reached primary completion within 32 months of starting (the median), counting those still running; a quarter within 21 and three quarters within 62. This study's planned 3 months is below that.

How these figures are worked out, and which studies are left out of them.

Registered
Nov 4, 2026
First posted
Nov 17, 2026
StartEst.
Dec 2026
Primary completionEst.
Mar 2027
CompletionEst.
May 2027
Last update
Nov 17, 2026
What these dates mean
Start.
When the first participant enrolled; while estimated, when the study expects to open to recruitment.
Primary completion.
When the last participant was examined or treated to collect data for the study's main question, its primary outcome. A study that stopped early gives the date its data collection stopped.
Completion.
When the last participant's last visit took place, for every outcome, secondary ones and side effects included: the end of the study. Often later than primary completion, sometimes the same day.
Actual and Est.
Actual means the registry says the date has happened. Est. (estimated) is a date the sponsor plans or expects. It can still move, and the sponsor is expected to change it to actual once it has passed.
Registered
First appeared on the registry. No changes since
Nov 17, 2026
Show all 1 update
  1. Nov 17, 2026
    First appeared on the registry

From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗

10

Registry details

Key details

Study ID
NCT07941700
Collaborators
SINTEF Digital
Responsible party
Henrik Brun (Professor, Oslo University Hospital) — Principal investigator
First posted
Nov 17, 2026
Start date
Dec 2026 (estimated)
Primary completion
Mar 2027 (estimated)
Completion
May 2027 (estimated)
Last update
Nov 17, 2026

Study contacts

Peder C Engelsen, MD
Contact
peder_engelsen@hotmail.com
+47 417 34 374
Henrik Brun, MD, PhD
Contact
hbrun@ous-hf.no
+47 470 11 012
Peder C Engelsen, MD
study director · Oslo University Hospital

Oversight

Data monitoring committee
No
FDA-regulated drug
No
FDA-regulated device
No
View the source record on ClinicalTrials.gov ↗

Not currently enrolling

This study is not yet recruiting, as verified in Nov 2026. You cannot join it, but the record below documents what was studied.

Follow this study

Get an email when the registry record changes — status, dates, results — or when someone posts here.

Sign in to follow

Discussion

Questions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.

Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.

Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.

Start the discussion