AI Assisted Reader Evaluation in Acute Computed Tomography (CT) Head Interpretation

NCT06018545 | Completed

• 1 other identifier: 310995 - A
• Study Type: observational
• Target: 33
• Locations: 1 country
• Sites: 4

Brief Summary

This study has been added as a sub study to the Simulation Training for Emergency Department Imaging 2 study (ClinicalTrials.gov ID NCT05427838). The purpose of the study is to assess the impact of an Artificial Intelligence (AI) tool called qER 2.0 EU on the performance of readers, including general radiologists, emergency medicine clinicians, and radiographers, in interpreting non-contrast CT head scans. The study aims to evaluate the changes in accuracy, review time, and diagnostic confidence when using the AI tool. It also seeks to provide evidence on the diagnostic performance of the AI tool and its potential to improve efficiency and patient care in the context of the National Health Service (NHS). The study will use a dataset of 150 CT head scans, including both control cases and abnormal cases with specific abnormalities. The results of this study will inform larger follow-up studies in real-life Emergency Department (ED) settings.

Conditions

Intracranial Hemorrhages; Acute Ischemic Stroke; Hydrocephalus; Cerebral Infarction; Cerebral Edema; Cerebral Injury

Keywords

Radiology; Head tomography; Emergency medicine; Radiographer; Artificial intelligence

Outcome Measures

Primary Outcomes (8)

• Reader performance: Sensitivity, specificity, comparative between with and without AI assistance.

  Reader performance will be evaluated as sensitivity, specificity, with and without AI assistance.

  During 6 weeks, which is the period for reading or reviewing the cases/scans.

• Reader performance: Positive and negative predictive value, comparative between with and without AI assistance.

  Reader performance will be evaluated as Positive Predictive Value (PPV) and negative predictive value (NPV), with and without AI assistance.

  During 6 weeks, which is the period for reading or reviewing the cases/scans.

• Reader performance: Area Under Receiver Operating Characteristic Curve (AUROC), comparative between with and without AI assistance.

  Reader performance will be evaluated as Area Under Receiver Operating Characteristic Curve (AUROC), with and without AI assistance.

  During 6 weeks, which is the period for reading or reviewing the cases/scans.

• Reader speed: Mean time taken to review a scan, with versus without AI assistance.

  Reader speed will be evaluated as the man time taken to review a scan, using time unite of seconds.

  During 6 weeks, which is the period for reading or reviewing the cases/scans.

• Reader confidence: Self-reported diagnostic confidence on a 10 point visual analogue scale, with vs without AI assistance.

  On the reading platform (RAIQC), one of the questions asks the level of confidence that the participant has in their diagnostic opinion. The question offers a scale of 1 to 10, where 1 is not confident, and 10 is highly confident.

  During 6 weeks, which is the period for reading or reviewing the cases/scans.

• qER (AI algorithm) performance: Sensitivity and specificity

  qER performance will be evaluated as sensitivity, specificity.

  During 6 weeks, which is the period for reading or reviewing the cases/scans.

• qER (AI algorithm) performance: Positive and negative predictive value.

  qER performance will be evaluated as Positive Predictive Value (PPV) and negative predictive value (NPV).

  During 6 weeks, which is the period for reading or reviewing the cases/scans.

• qER (AI algorithm) performance: Area Under Receiver Operating Characteristic Curve (AUROC).

  qER performance will be evaluated as Area Under Receiver Operating Characteristic Curve (AUROC)

  During 6 weeks, which is the period for reading or reviewing the cases/scans.

Study Arms (2)

Readers

30 readers will be recruited across four NHS trusts including ten general radiologists, fifteen emergency medicine clinicians, and five CT radiographers of varying seniority. Readers will interpret each scan first without, then with, the assistance of the AI tool, with an intervening 4-week washout period. Using a panel of neuroradiologists as ground truth, the stand-alone performance of qER will be assessed, and its impact on the readers' performance will be analysed as change in accuracy, mean review time per scan, and self-reported diagnostic confidence. Subgroup analyses will be performed by reader professional group, reader seniority, pathological finding, and neuroradiologist-rated difficulty.

Other: Reading

Ground truthers

Two Consultant neuroradiologists will independently review the images to establish the 'ground truth' findings on the CT scans which will be used as the reference standard. In the case of disagreement, a third senior neuroradiologist's opinion will be sought for arbitration. A difficulty score will be assigned to each scan by the ground truthers using a 5-point Likert scale.

Other: Ground truthing

Interventions

Ground truthing OTHER

Two Consultant neuroradiologists will independently review the images to establish the 'ground truth' findings on the CT scans which will be used as the reference standard. In the case of disagreement, a third senior neuroradiologist's opinion will be sought for arbitration.

Ground truthers

Reading OTHER

All 30 readers will review all 150 cases, in each of two study phases. The readers will provide their opinion on the presence or absence of some acute abnormalities, including intracranial haemorrhage, infarct, midline shift and fracture. They will provide a confidence in their diagnosis (10-point visual analogue scale), and a single click point to mark the location of each abnormality that they consider as being present. The time taken for each scan will be automatically recorded.

Readers

Eligibility Criteria

• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Setting: Readers will be recruited from the following four hospital Trusts (secondary and tertiary level: * Guy's \& St Thomas NHS Foundation Trust * Northumbria Healthcare NHS Foundation Trust * NHS Greater Glasgow and Clyde * Oxford University Hospitals NHS Foundation Trust Participants: 30 volunteer participant readers will be selected from the following groups: * Emergency Medicine Consultants and Registrars (5 Consultant, 5 Registrar (ST3-6), 5 junior (F1-ST2) * General Radiologist Consultants and Registrars (5 Consultant, 5 Registrar) * 5 CT Radiographers

You may qualify if:

• Radiologists/Radiographers/ED clinicians who review CT head scans as part of their clinical practice

You may not qualify if:

• Neuroradiologists.
• Non-radiologist groups: Clinicians with previous formal postgraduate CT reporting training
• Emergency Medicine group: Clinicians with previous career in radiology/neurosurgery to registrar level

Sponsors & Collaborators

• Oxford University Hospitals NHS Trust: lead

Study Sites (4)

Guy's & St Thomas NHS Foundation Trust

London, London, SE1 7EH, United Kingdom

Location

Oxford University Hospitals NHS Foundation Trust

Oxford, Oxfordshire, OX3 9DU, United Kingdom

Location

NHS Greater Glasgow and Clyde

Glasgow, G12 0XH, United Kingdom

Location

Northumbria Healthcare NHS Foundation Trust

Newcastle upon Tyne, NE27 0QJ, United Kingdom

Location

Related Publications (22)

• Juszczyk K, Ireland K, Thomas B, Kroon HM, Hollington P. Reduction in hospital admissions with an early computed tomography scan: results of an outpatient management protocol for uncomplicated acute diverticulitis. ANZ J Surg. 2019 Sep;89(9):1085-1090. doi: 10.1111/ans.15285. Epub 2019 Jun 17.

  PMID: 31206250 BACKGROUND

• Chan J, Fan KS, Mak TLA, Loh SY, Ng SWY, Adapala R. Pre-Operative Imaging can Reduce Negative Appendectomy Rate in Acute Appendicitis. Ulster Med J. 2020 Jan;89(1):25-28. Epub 2020 Feb 18.

  PMID: 32218624 BACKGROUND

• Greenhalgh R, Howlett DC, Drinkwater KJ. Royal College of Radiologists national audit evaluating the provision of imaging in the severely injured patient and compliance with national guidelines. Clin Radiol. 2020 Mar;75(3):224-231. doi: 10.1016/j.crad.2019.10.025. Epub 2019 Dec 19.

  PMID: 31864722 BACKGROUND

• Lin E, Yuh EL. Computational Approaches for Acute Traumatic Brain Injury Image Recognition. Front Neurol. 2022 Mar 9;13:791816. doi: 10.3389/fneur.2022.791816. eCollection 2022.

  PMID: 35370919 BACKGROUND

• Sheth SA, Giancardo L, Colasurdo M, Srinivasan VM, Niktabe A, Kan P. Machine learning and acute stroke imaging. J Neurointerv Surg. 2023 Feb;15(2):195-199. doi: 10.1136/neurintsurg-2021-018142. Epub 2022 May 25.

  PMID: 35613840 BACKGROUND

• Yeo M, Tahayori B, Kok HK, Maingard J, Kutaiba N, Russell J, Thijs V, Jhamb A, Chandra RV, Brooks M, Barras CD, Asadi H. Review of deep learning algorithms for the automatic detection of intracranial hemorrhages on computed tomography head imaging. J Neurointerv Surg. 2021 Apr;13(4):369-378. doi: 10.1136/neurintsurg-2020-017099. Epub 2021 Jan 21.

  PMID: 33479036 BACKGROUND

• Chilamkurthy S, Ghosh R, Tanamala S, Biviji M, Campeau NG, Venugopal VK, Mahajan V, Rao P, Warier P. Deep learning algorithms for detection of critical findings in head CT scans: a retrospective study. Lancet. 2018 Dec 1;392(10162):2388-2396. doi: 10.1016/S0140-6736(18)31645-3. Epub 2018 Oct 11.

  PMID: 30318264 BACKGROUND

• Guo Y, He Y, Lyu J, Zhou Z, Yang D, Ma L, Tan HT, Chen C, Zhang W, Hu J, Han D, Ding G, Liu S, Qiao H, Xu F, Lou X, Dai Q. Deep learning with weak annotation from diagnosis reports for detection of multiple head disorders: a prospective, multicentre study. Lancet Digit Health. 2022 Aug;4(8):e584-e593. doi: 10.1016/S2589-7500(22)00090-5. Epub 2022 Jun 17.

  PMID: 35725824 BACKGROUND

• Lee JY, Kim JS, Kim TY, Kim YS. Detection and classification of intracranial haemorrhage on CT images using a novel deep-learning algorithm. Sci Rep. 2020 Nov 25;10(1):20546. doi: 10.1038/s41598-020-77441-z.

  PMID: 33239711 BACKGROUND

• Arbabshirani MR, Fornwalt BK, Mongelluzzo GJ, Suever JD, Geise BD, Patel AA, Moore GJ. Advanced machine learning in action: identification of intracranial hemorrhage on computed tomography scans of the head with clinical workflow integration. NPJ Digit Med. 2018 Apr 4;1:9. doi: 10.1038/s41746-017-0015-z. eCollection 2018.

  PMID: 31304294 BACKGROUND

• Davis MA, Rao B, Cedeno PA, Saha A, Zohrabian VM. Machine Learning and Improved Quality Metrics in Acute Intracranial Hemorrhage by Noncontrast Computed Tomography. Curr Probl Diagn Radiol. 2022 Jul-Aug;51(4):556-561. doi: 10.1067/j.cpradiol.2020.10.007. Epub 2020 Nov 15.

  PMID: 33243455 BACKGROUND

• Wardlaw JM, Mair G, von Kummer R, Williams MC, Li W, Storkey AJ, Trucco E, Liebeskind DS, Farrall A, Bath PM, White P. Accuracy of Automated Computer-Aided Diagnosis for Stroke Imaging: A Critical Evaluation of Current Evidence. Stroke. 2022 Jul;53(7):2393-2403. doi: 10.1161/STROKEAHA.121.036204. Epub 2022 Apr 20.

  PMID: 35440170 BACKGROUND

• Finck T, Moosbauer J, Probst M, Schlaeger S, Schuberth M, Schinz D, Yigitsoy M, Byas S, Zimmer C, Pfister F, Wiestler B. Faster and Better: How Anomaly Detection Can Accelerate and Improve Reporting of Head Computed Tomography. Diagnostics (Basel). 2022 Feb 10;12(2):452. doi: 10.3390/diagnostics12020452.

  PMID: 35204543 BACKGROUND

• Warman R, Warman A, Warman P, Degnan A, Blickman J, Chowdhary V, Dash D, Sangal R, Vadhan J, Bueso T, Windisch T, Neves G. Deep Learning System Boosts Radiologist Detection of Intracranial Hemorrhage. Cureus. 2022 Oct 13;14(10):e30264. doi: 10.7759/cureus.30264. eCollection 2022 Oct.

  PMID: 36381767 BACKGROUND

• Dyer T, Chawda S, Alkilani R, Morgan TN, Hughes M, Rasalingham S. Validation of an artificial intelligence solution for acute triage and rule-out normal of non-contrast CT head scans. Neuroradiology. 2022 Apr;64(4):735-743. doi: 10.1007/s00234-021-02826-4. Epub 2021 Oct 8.

  PMID: 34623478 BACKGROUND

• Mallon DH, Taylor EJR, Vittay OI, Sheeka A, Doig D, Lobotesis K. Comparison of automated ASPECTS, large vessel occlusion detection and CTP analysis provided by Brainomix and RapidAI in patients with suspected ischaemic stroke. J Stroke Cerebrovasc Dis. 2022 Oct;31(10):106702. doi: 10.1016/j.jstrokecerebrovasdis.2022.106702. Epub 2022 Aug 19.

  PMID: 35994882 BACKGROUND

• Andralojc LE, Kim DH, Edwards AJ. Diagnostic accuracy of a decision-support software for the detection of intracranial large-vessel occlusion in CT angiography. Clin Radiol. 2023 Apr;78(4):e313-e318. doi: 10.1016/j.crad.2022.10.017. Epub 2023 Jan 11.

  PMID: 36754714 BACKGROUND

• Zech JR, Badgeley MA, Liu M, Costa AB, Titano JJ, Oermann EK. Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study. PLoS Med. 2018 Nov 6;15(11):e1002683. doi: 10.1371/journal.pmed.1002683. eCollection 2018 Nov.

  PMID: 30399157 BACKGROUND

• Huang SC, Pareek A, Jensen M, Lungren MP, Yeung S, Chaudhari AS. Self-supervised learning for medical image classification: a systematic review and implementation guidelines. NPJ Digit Med. 2023 Apr 26;6(1):74. doi: 10.1038/s41746-023-00811-0.

  PMID: 37100953 BACKGROUND

• Hillis SL, Obuchowski NA, Schartz KM, Berbaum KS. A comparison of the Dorfman-Berbaum-Metz and Obuchowski-Rockette methods for receiver operating characteristic (ROC) data. Stat Med. 2005 May 30;24(10):1579-607. doi: 10.1002/sim.2024.

  PMID: 15685718 BACKGROUND

• Obuchowski NA. Sample size tables for receiver operating characteristic studies. AJR Am J Roentgenol. 2000 Sep;175(3):603-8. doi: 10.2214/ajr.175.3.1750603.

  PMID: 10954438 BACKGROUND

• Fu H, Novak A, Robert D, Kumar S, Tanamala S, Oke J, Bhatia K, Shah R, Romsauerova A, Das T, Espinosa A, Grzeda MT, Narbone M, Dharmadhikari R, Harrison M, Vimalesvaran K, Gooch J, Woznitza N, Salik N, Campbell A, Khan F, Lowe DJ, Shuaib H, Ather S. AI assisted reader evaluation in acute CT head interpretation (AI-REACT): protocol for a multireader multicase study. BMJ Open. 2024 Feb 12;14(2):e079824. doi: 10.1136/bmjopen-2023-079824.

  PMID: 38346874 DERIVED

Related Links

• Richards M. Diagnostics: Recovery and Renewal - Report of the Independent Review of Diagnostic Services for NHS England. NHS England 2022.
• Royal College of Radiologists. Clinical radiology UK workforce census 2019 report. Royal College of Radiologists 2020.
• National Institute for Health and Care Excellence. Artificial intelligence for analysing CT brain scans. 2020.

MeSH Terms

Conditions

Intracranial Hemorrhages; Ischemic Stroke; Hydrocephalus; Cerebral Infarction; Brain Edema; Brain Injuries

Condition Hierarchy (Ancestors)

Cerebrovascular Disorders; Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Vascular Diseases; Cardiovascular Diseases; Hemorrhage; Pathologic Processes; Pathological Conditions, Signs and Symptoms; Stroke; Brain Infarction; Brain Ischemia; Infarction; Ischemia; Necrosis; Craniocerebral Trauma; Trauma, Nervous System; Wounds and Injuries

Study Officials

• Alex Novak, MSc

  National Health Services in the United Kingdom (NHS UK)

  PRINCIPAL INVESTIGATOR

• Sarim Ather, PhD

  National Health Services in the United Kingdom (NHS UK)

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Primary Investigator

Study Record Dates

• First Submitted: July 25, 2023
• First Posted: August 30, 2023
• Study Start: June 1, 2023
• Primary Completion: September 1, 2023
• Study Completion: June 1, 2025
• Last Updated: November 24, 2025

Record last verified: 2025-11

Locations

GB (4)