Skip to content

Multimodal Radiology Report to Improve Patient-centered Radiology

Multimodal Radiology Report to Improve Patient-centered Radiology

Status
Not yet recruiting
Phases
NA
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT07159438
Enrollment
200
Registered
2025-09-08
Start date
2026-12-31
Completion date
2028-12-31
Last updated
2025-09-08

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Artificial Intelligence, Diagnostic Imaging, Health Literacy, Imaging Results

Keywords

Patient-centered Understanding, Radiology Reports, Artificial Intelligence

Brief summary

The goal of this study is to learn if AI-generated video explanations help people better understand their radiology reports. The main question it aims to answer is: Do AI-generated videos help participants understand their medical imaging results better than written reports alone? Participants will send their own radiology images and written reports to the research team; receive a personalized AI-generated video that explains their results in easy-to-understand language; watch their video explanation (about 1-5 minutes long); and complete 15-minute online survey about how well the video helped them understand their results.

Interventions

OTHERAI-Generated Video Radiology Report Explanation

Participants receive a personalized video explanation of their own radiology imaging results, generated using artificial intelligence technology. The AI converts the participant's existing written radiology report into an easy-to-understand video format featuring an avatar narrator. Each video is approximately 1-5 minutes long and explains the participant's specific imaging findings in plain language. All AI-generated videos are reviewed by qualified medical personnel before delivery to ensure accuracy and prevent misleading information. Participants watch their personalized video and then complete survey to evaluate how well the video helped them understand their radiology results compared to the original written report.

Sponsors

Harvard Medical School (HMS and HSDM)
Lead SponsorOTHER

Study design

Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE

Eligibility

Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes

Inclusion criteria

* Age 18+ adults who have taken a radiology examination.

Exclusion criteria

* N/A

Design outcomes

Primary

MeasureTime frameDescription
Change in Patient Understanding of Radiology Report After AI-Generated Video ExplanationBaseline (before viewing video) and immediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Change in patient-reported understanding score measured by the question How would you rank your understanding of the radiology report? assessed before and after viewing the AI-generated video explanation. Scores range from 1 (Not at all) to 5 (Very well), where higher scores indicate better understanding. The primary outcome is calculated as the difference between post-video and pre-video understanding scores, with positive values indicating improved understanding.
Change in Patient Understanding of Radiology Images After AI-Generated Video ExplanationBaseline (before viewing video) and immediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Change in patient-reported understanding score measured by the question How would you rank your understanding of the radiology image? assessed before and after viewing the AI-generated video explanation. Scores range from 1 (Not at all) to 5 (Very well), where higher scores indicate better understanding. The outcome is calculated as the difference between post-video and pre-video understanding scores, with positive values indicating improved understanding.

Secondary

MeasureTime frameDescription
Perceived Ease of Use of AI-Generated Video Radiology ExplanationImmediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Patient-reported ease of use score measured by the statement The AI video explanation is easy to follow and interact with assessed after viewing the AI-generated video explanation. Scores range from 1 (Strongly Disagree) to 5 (Strongly Agree), where higher scores indicate greater perceived ease of use of the AI video explanation.
Overall Patient Experience with AI-Generated Video Radiology ExplanationImmediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Patient-reported overall experience score measured by the question Overall, how was your experience with the video report? assessed after viewing the AI-generated video explanation. Scores range from 1 (Poor) to 5 (Excellent), where higher scores indicate better overall experience with the AI video explanation.
Patient Likelihood to Recommend AI-Generated Video Radiology ExplanationsImmediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)RetryClaude can make mistakes. Please double-check responses.Patient-reported recommendation likelihood score measured by the statement I would recommend this type of video explanation to other people to understand their radiology findings assessed after viewing the AI-generated video explanation. Scores range from 1 (Strongly Disagree) to 5 (Strongly Agree), where higher scores indicate greater likelihood to recommend the AI video explanation to others.
Change in Patient Knowledge of Body Part Described in Radiology Report After AI-Generated Video ExplanationBaseline (before viewing video) and immediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Change in patient knowledge measured by the question Do you know which part of your body was described in the report? assessed before and after viewing the AI-generated video explanation. Responses are Yes/No, where the outcome is calculated as the proportion of participants who gained knowledge (changed from No to Yes) after viewing the video.
Understanding of Finding SignificanceBaseline (before viewing video) and immediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Change in participant's understanding of the significance of their radiology findings measured on a 5-point Likert scale (1=Not at all to 5=Very well), comparing pre-video to post-video scores
Change in Ability to Explain Radiology Findings to Family Members After AI-Generated Video ExplanationBaseline (before viewing video) and immediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Change in patient-reported confidence score measured by the statement I can explain the main findings of my radiology report to my family members assessed before and after viewing the AI-generated video explanation. Scores range from 1 (Strongly Disagree) to 5 (Strongly Agree), where higher scores indicate greater confidence in explaining findings. The outcome is calculated as the difference between post-video and pre-video confidence scores, with positive values indicating improved confidence.
Change in Patient Confidence Discussing Radiology Results with Healthcare Provider After AI-Generated Video ExplanationBaseline (before viewing video) and immediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Change in patient-reported confidence score measured by the statement I feel confident discussing my radiology results with my healthcare provider assessed before and after viewing the AI-generated video explanation. Scores range from 1 (Strongly Disagree) to 5 (Strongly Agree), where higher scores indicate greater confidence. The outcome is calculated as the difference between post-video and pre-video confidence scores, with positive values indicating improved confidence.
Change in Patient Worry About Misunderstanding Radiology Findings After AI-Generated Video ExplanationBaseline (before viewing video) and immediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Change in patient-reported worry score measured by the statement I am worried about misunderstanding my radiology findings assessed before and after viewing the AI-generated video explanation. Scores range from 1 (Not worried) to 5 (Extremely worried), where lower scores indicate less worry. The outcome is calculated as the difference between post-video and pre-video worry scores, with negative values indicating reduced worry.
Perceived Usefulness of AI-Generated Video Radiology ExplanationImmediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Patient-reported perceived usefulness score measured by the statement Using the AI video explanation improves my understanding of my radiology findings assessed after viewing the AI-generated video explanation. Scores range from 1 (Strongly Disagree) to 5 (Strongly Agree), where higher scores indicate greater perceived usefulness of the AI video explanation.
Patient Trust in AI-Generated Video Radiology ExplanationImmediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Patient-reported trust score measured by the statement I trust the AI's explanation to accurately explain my radiology findings assessed after viewing the AI-generated video explanation. Scores range from 1 (Strongly Disagree) to 5 (Strongly Agree), where higher scores indicate greater trust in the AI explanation's accuracy.
Patient Intention to Use AI-Generated Video Explanations for Future Radiology ReportsImmediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Patient-reported intention to use score measured by the statement I would like to receive this type of AI-based explanation for future radiology reports assessed after viewing the AI-generated video explanation. Scores range from 1 (Strongly Disagree) to 5 (Strongly Agree), where higher scores indicate greater intention to use AI explanations in the future.

Other

MeasureTime frameDescription
Helpful Components of AI-Generated Video Radiology ExplanationImmediately after viewing the AI-generated video explanation (within 24 hours of receiving the video)Participant-reported helpful aspects measured by the question What did you find helpful about the video report? with multiple response options including: findings highlighted in images, avatar using easy-to-understand language, comparisons to normal studies, comparisons to older studies, use of 3D images, easier to understand than written report, and side-by-side comparisons. Responses are analyzed as frequencies and percentages for each component.
Video Engagement - Number of ReplaysDuring viewing of the AI-generated video explanationNumber of times participants replay sections of the AI-generated radiology explanation video. This metric assesses participant engagement and areas requiring clarification or repeated viewing.
Video Engagement - Number of PausesDuring viewing of the AI-generated video explanationNumber of times participants pause the AI-generated radiology explanation video during viewing. This metric assesses participant engagement patterns and need for additional processing time.
Video Engagement - Total Viewing TimeDuring viewing of the AI-generated video explanationTotal time in minutes spent viewing the AI-generated radiology explanation video. This metric assesses overall participant engagement duration with the video content.

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026