AI Timing in Chest X-ray Interpretation Using Eye-Tracking
Trial statusRecruiting
Trial phaseNot applicable
Trial typeInterventional
Biological sexAll
Age18+
SponsorUniversity Hospitals, Leicester
About this trial
Chest X-rays are commonly used to help diagnose and manage chest conditions. Artificial intelligence (AI) tools are increasingly being used to support chest X-ray interpretation. However, it is not yet clear whether the timing of AI information affects how clinicians review images, make decisions, and use AI support.
This study will look at whether showing AI information before or after a clinician first reviews a chest X-ray changes how they look at the image, how long they take, their interpretation decisions, their confidence, and their trust in AI support.
Healthcare professional participants will complete two chest X-ray interpretation sessions in a controlled NHS research setting. During each session, participants will review de-identified chest X-ray images while wearing eye-tracking equipment. Eye-tracking will record where a participant looks on the image and how long they spend looking at different areas.
In one session, AI information will be shown before the participant reviews the chest X-ray. In the other session, AI information will be shown after the participant has first reviewed the chest X-ray. The order of these two sessions will be balanced across participants.
The study uses de-identified chest X-ray images from existing examinations. It does not involve patients directly, does not change clinical care, and no clinical decisions will be made from the study readings. Participants will also complete a short questionnaire about their experience of using AI support. A separate anonymous survey will collect wider views from clinicians, patients, members of the public, and healthcare staff about the use of AI in chest X-ray interpretation.
Eligibility criteria
This trial accepts healthy volunteersQualifiers
Healthcare professionals aged 18 years or over
Registered to practise in the UK
Employed by the NHS or another healthcare service operating in the UK
Current or recent, within the last 3 years, clinical experience involving chest X-ray interpretation, review, or use in clinical practice
Disqualifiers
Inability to attend both onsite study sessions at University Hospitals of Leicester NHS Trust
Eye-tracking incompatibility, such as visual, neurological, or physical conditions preventing adequate gaze tracking or participant comfort
Direct involvement in selection, adjudication, or preparation of the chest X-rays used in the study
Conflicts of interest, including direct involvement in development of the AI system under evaluation
Trial design
Design model
Crossover
Treatments tested in this trial
Original CXR First Timing
BehavioralParticipants first review the original chest X-ray without AI output and complete an initial structured interpretation response. AI output is then introduced, and participants may review or revise their response before submitting the final response for that case.
AI Output First Timing
BehavioralParticipants first view AI output before reviewing the original chest X-ray. The original chest X-ray is then introduced, and participants complete or revise their structured interpretation response before submitting the final response for that case.
Treatment groups
24 Participants
are divided into 2 treatment groupsGroup A: Original CXR First SequenceExperimental treatment 2 interventions
Group B: AI Output First SequenceExperimental treatment 2 interventions
Trial outcomes
Primary outcomes
1
Change in Chest X-ray Case Interpretation Time
Case interpretation time is the duration, in seconds, from initial case display to final structured report submission. Change in case interpretation time will be measured between Session 1 and Session 2, displaying either original chest X-ray first or AI output first.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeks2
Change in Chest X-ray Case Diagnostic Accuracy
Diagnostic accuracy will be assessed using structured report outcomes compared with clinician-verified chest X-ray ground truth. Each participant's final structured interpretation for each chest X-ray case will be compared with the verified study ground truth for that case. Change in diagnostic accuracy will be measured between Session 1 and Session 2, displaying either original chest X-ray first or AI output first.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeks3
Change in Chest X-ray Case Visual Search Behaviour
Visual search behaviour will be assessed using predefined eye-tracking metrics that capture gaze location and search behaviour during chest X-ray interpretation. Metrics include time to first fixation, hit time, dwell time within abnormal case areas of interest, fixation count, mean fixation duration, revisit count, and fixation dispersion. Change in visual search behaviour will be measured between Session 1 and Session 2, displaying either original chest X-ray first or AI output first.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeks4
Trust in AI Support After Reader-Study Completion
Trust in AI support will be assessed using structured questionnaire ratings after participants have completed both Session 1 and Session 2. This measure captures participants' overall perceived usefulness, reliance on AI support, and willingness to revise decisions after reviewing AI information.
Time frame
Immediately after Session 2, anticipated average 5 weeks after enrolmentSecondary outcomes
1
Change in Chest X-ray Case Clinical Management Recommendations
Clinical management recommendations will be recorded from participants' structured report responses for each chest X-ray case. Change in clinical management recommendations will be measured between Session 1 and Session 2, displaying either original chest X-ray first or AI output first.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeks2
Change in Chet X-ray Case Decision Confidence
Decision confidence will be recorded as the participant's confidence rating in their final diagnostic judgement for each chest X-ray case. Change in decision confidence will be measured between Session 1 and Session 2, displaying either original chest X-ray first or AI output first.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeks3
Overall Reflective Confidence After Reader-Study Completion
Overall reflective confidence will be assessed using structured questionnaire ratings after participants completed both Session 1 and Session 2. This measure captures participants' overall perceived confidence in their interpretive decisions across the reader study.
Time frame
Immediately after Session 2, anticipated average 5 weeks after enrolment4
Participant Perceptions of AI Support After Reader-Study Completion
Participant perceptions of AI support will be assessed using a post-study questionnaire after participants completed both Session 1 and Session 2. This measure captures participants' overall perceptions of usability, clarity, workflow implications, communication preferences, and responsibility.
Time frame
Immediately after Session 2, anticipated average 5 weeks after enrolmentOther outcomes
1
Differences in Chest X-ray Case Diagnostic Accuracy, Visual Search Behaviour, and Trust in AI Support by Clinical Background and Experience
Subgroup differences between participants with different clinical background and years of experience. Differences between participant subgroups will be measured for diagnostic accuracy, visual search behaviour, and trust in AI support, obtained during Session 1 and Session 2. Diagnostic accuracy uses structured report outcomes. Visual search behaviour metrics include time to first fixation, hit time, dwell time within abnormal case areas of interest, fixation count, mean fixation duration, revisit count, and fixation dispersion. Trust in AI support uses structured questionnaire ratings.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeks2
Association Between Chest X-ray Case Visual Search Behaviour and Diagnostic Accuracy
The association between visual search behaviour and diagnostic accuracy will be explored using outputs obtained during Session 1 and Session 2. Diagnostic accuracy will be assessed using structured report outcomes. Visual search behaviour metrics include time to first fixation, hit time, dwell time within abnormal case areas of interest, fixation count, mean fixation duration, revisit count, and fixation dispersion.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeks3
Association Between Chest X-ray Case Visual Search Behaviour and Case Complexity
The association between visual search behaviour and chest X-ray case complexity will be explored using outputs obtained during Session 1 and Session 2. Visual search behaviour metrics include time to first fixation, hit time, dwell time within abnormal case areas of interest, fixation count, mean fixation duration, revisit count, and fixation dispersion. Chest X-ray case complexity is predefined as low, moderate, or high.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeks4
Association Between Chest X-ray Case Visual Search Behaviour and AI Error Type
The association between visual search behaviour and chest X-ray case AI error type will be explored using outputs obtained during Session 1 and Session 2. Visual search behaviour metrics include time to first fixation, hit time, dwell time within abnormal case areas of interest, fixation count, mean fixation duration, revisit count, and fixation dispersion. Chest X-ray case AI error type is predefined as false-positive or false-negative.
Time frame
Session 1 at enrolment and Session 2 at least 4 weeks later, anticipated average 5 weeksSponsors and contacts
Click on the lead sponsor to view all of their trials.
Source ClinicalTrials.gov