Polyps ColorectalFAIRNCT07158203
Optical Diagnosis of Neoplasia Using Artificial Intelligence
A study testing CADx simultaneously and CADx delayed in people with Polyps Colorectal and Colonoscopy, run by Fundacin Biomedica Galicia Sur at 2 sites in 2 countries.
Trial summary
Computer-aided diagnosis (CADx) for colonoscopy aims to enhance optical diagnosis but often underperforms when used alongside humans due to under-reliance on AI. Psychological interventions like cognitive forcing, such as delaying CADx suggestions, may improve human-AI interaction by fostering critical assessment. Read more about this study
- Who is it for?
- Ages 18 years and olderHealthy volunteers can take part
- Conditions the study is for
- Polyps Colorectal, Colonoscopy, Optical Biopsy
- What is being tested?
- CADx simultaneously and CADx delayed
- Type of treatment
- Medical device, Lifestyle or therapy programme
- Could I receive a placebo?
- No, everyone receives an active treatment
- How many people
- 70 participants
- Where
- 2 sitesin 2 countries
- Study dates
- Oct 2026to Dec 2026
- Phase
- Interventional
- Are there results?
- Not yet: the study is under way
Key requirements
Who can take part, as listed by the sponsor. Only the study team can confirm whether you are eligible, so take this list to your doctor.
- Ages 18 years and older
- All sexes
- Healthy volunteers accepted
- Adult, Older adult
Endoscopists who meet the following inclusion/exclusion criteria Inclusion criteria
- Endoscopists experienced with more than 100 colonoscopies. Exclusion criteria
- Endoscopists who are involved in the development of the protocol of the present study.
Inclusion and exclusion criteria for the study videos. Inclusion criteria.
- Videos with a duration of 15 seconds including both WL and NBI.
- Videos with diminutive polyps with confirmed pathology. Exclusion criteria.
- Videos with no clear image of the polyps.
- Videos with more than one polyp on-screen.
- Inflammatory bowel disease
- Polyposis
- Hereditary colorectal disease
- Videos which CADx cannot provide sufficient number of outputs.
Trial locations and contacts
2 sites in 2 countries. Contact the study team or a site near you to ask about taking part.
Pedro Davila Piñón, Master degree in biotechnology contact
2 sites in 2 countries
Norway1 site
Clinical Effectiveness Research Group
Oslo, Oslo, 0424
Yuichi Mori, MD, PhD, (+47) 934 81 380, yuichi.mori@medisin.uio.no
Spain1 site
Research Group in Gastrointestinal Oncology Ourense
Ourense, Ourense, 32005
Pedro Davila, Master Degree in biotechnology, +34 988385047 ext. 285047, pedro.davila@iisgaliciasur.es
Joaquin Cubiella, MD, PhD, +34 988385047 ext. 285047, joaquin.cubiella.fernandez@sergas.es
About this study
Official title: Assistance for Optical Diagnosis of Neoplasia Using Artificial Intelligence (FAIR Study)
Computer-aided diagnosis (CADx) for colonoscopy aims to enhance optical diagnosis but often underperforms when used alongside humans due to under-reliance on AI. Psychological interventions like cognitive forcing, such as delaying CADx suggestions, may improve human-AI interaction by fostering critical assessment. However, their impact on patient-important outcomes remains unexplored.
The investigators will conduct an ex-vivo randomized study with 70 endoscopists assessing 100 polyp videos (≤5 mm) using a CADx tool (GI Genius, Medtronic). Participants will be randomized to either:
Read more
- Intervention group: CADx suggestions will be shown in the last 3 seconds of the 15 second polyp video.
- Control group: CADx suggestions will be shown in real-time throughout the playback of the 15 second polyp video.
The primary endpoint is sensitivity for high-confidence neoplasia detection, with secondary endpoints assessing endoscopists' reliance on AI.
CADx systems on the market function in various ways, such as real-time, delayed, or on-demand diagnosis. Our study aims to inform users and manufacturers whether cognitive forcing through delayed CADx suggestions enhances human-AI interaction, leading to improved clinical outcomes.
Read the detailed description
Computer-Aided Diagnosis and cognitive forcing Computer-aided diagnosis (CADx) for colonoscopy is expected to improve physicians' ability to predict colorectal polyp pathologies (optical diagnosis). However, recent randomized trials indicate that collaboration between humans and CADx yields lower performance than CADx alone.
This suggests suboptimal human-AI interaction due to users' under-reliance on CADx advice, favoring their inherent biases over critically assessing AI suggestions. Psychological interventions, such as cognitive forcing, aim to address this by encouraging crucial assessment of AI suggestions. One example, delayed display of CADx suggestions may promote proactive thinking by giving endoscopists enough time to consider polyp pathologies before receiving CADx suggestions, potentially leading to critical and optimal human-AI interaction. The effectiveness of such cognitive forcing was observed in experimental studies of AI for mammography reading.
However, despite its potential, no studies have evaluated the impact of such interventions on optical diagnosis accuracy in colonoscopy. To investigate the value of psychological intervention in optical diagnosis, the investigators will conduct an ex-vivo randomised controlled study.
Study aim:
This is an ex-vivo randomised controlled study. The hypothesis of our study is that cognitive forcing by delaying display of CADx suggestions facilitates physicians' critical thinking, leading to better clinical outcomes in optical diagnosis in colonoscopy.
Comercially CADx device and video details:
In this study, the investigators are going to use the comercially available CADx system (GI Genius CADx, manufactured by Cosmo Intelligent Medical Devices and distributed by Medtronic Corp). The GI Genius CADx highlights the suspicious area for polyps on-screen with bounding boxes and provides optical diagnosis prediction (i.e. adenoma, non-adenoma). All the endoscopists will be given the information that the CADx tool used in the study is GI Genius with a link to their product overview.
The investigators will collect 100 colonoscopy videos of 100 different diminutive polyps from the Polyp Image BAnk database (PIBAdb). In accordance with the real-world prevalence, 65 polyps will be neoplastic while the remaining 35 will be non-neoplastic. The duration of each video will be adjusted to contain 15 second appearance of the lesion including wite light (WL) and narrow band imaging (NBI). All the polyps should be < = 5 mm. The original videos were recorded without having any CADx interaction.
PIBAdb contains 507 videos of diminutive colorectal polyps with WL and NBI, of which 231 have a duration of 15 seconds or more. All the videos contain polyps with their histopathology available (i.e. adenoma, sessile serrated lesions, traditional serrated adenomas, invasive, hyperplastic and non-neoplastic). The polyps that have no histology category are going to be excluded from the study.
The investigators will use this database as a pool to select the 100 study videos. First, the investigators are going to split the pool of data into two groups according to polyp histology: one pool containing neoplasia (i.e. adenoma, sessile serrated lesions and traditional serrated adenoma) and the other containing non-neoplasia (i.e. hyperplastic and non-epithelial neoplastic). In each pool, 20 videos will be randomly selected as a first step. These 20 videos will be assessed if they meet our inclusion and exclusion criteria (see below), leaving only eligible videos. This selection process will be repeatedly done until the investigators collect 65 videos of polyps with neoplasia and 35 videos of polyps with non-neoplasia. The figure below shows how the video selection process takes place.
GI Genius´s specification The GI Genius system processes the videos at 50-60 frames per second and in high-definition format to have a good quality of the videos. To process the videos, the GI Genius splits them into two different streams by dedicated video card. One stream is transmitted to the first path to the output without any processing of the AI algorithm. The second stream is sent to the AI algorithm (i.e. second path) and after appropriate computation, if there is a polyp present on-screen appears an overlay in the output highlighting the polyp. The system was designed to work on unaltered WL video streams, but it works under both WL and NBI similarly.The transmission of the video stream to GI Genius is through a Serial Digital Interface (SDI) cable, acquiring the output stream from the video displaying in a computer. Finally, the SDI output stream is transmitted to the monitor containing the original video stream with additional markers superimposed on it.
How to make CADx overlaid videos In the present study, the selected videos of colorectal lesions will be stored in our high-spec computer system first. Then, these videos will be transmitted to GI Genius with an SDI cable. A capture card (DeckLink 8K Pro Mini, Blackmagic) is integrated into the computer system that converts the recorded video output to SDI signal, which allows GI Genius to process the transmitted videos.
All 100 selected videos will be processed by GI Genius in two different ways. In the first set, only the frames from the last 3 seconds of each 15-second video will be processed, and CADx suggestions will appear only in that final 3-second segment. This first set will be shown to endoscopists who are allocated to the intervention arm. In the second set, all frames from the entire 15-second duration will be processed, and CADx suggestions will be overlaid on every frame. This first set will be shown to endoscopists who are allocated to the control arm.
- Conditions
- Polyps Colorectal, Colonoscopy, Optical Biopsy, Colorectal Cancer Control and Prevention, Colorectal Cancer Screening, Behavior Change
- Being tested
- CADx simultaneously, CADx delayed
- Keywords
- behaviour, optical biopsy, colonoscopy, CADx system, colorectal polyps, psychological intervention
- Medical subject headings
- Behavior
What participants receive
What participants receive. In a randomised study each person is assigned to one group by chance.
CADx simultaneouslyDevice
The investigators showed the CADx suggestion during the 15-second playback of the video
CADx delayedBehavioural
During the 15-seconds polyp video the CADx suggestion appear only in the last 3 seconds of the video
| Study group | Role | What they receive |
|---|---|---|
| CADx suggestions will be shown in 15 second polyp video. | Active comparator | CADx suggestions will be shown in real-time throughout the playback of the 15 second polyp video. Receives: Device: CADx simultaneously |
| CADx suggestions will be shown in the last 3 seconds of the 15 second polyp video. | Experimental | CADx suggestions will be shown in the last 3 seconds of the 15 second polyp video. Receives: Behavioral: CADx delayed |
What is being measured
The primary outcome is the main question the study is designed to answer.
Sensitivity of the optical diagnosis of neoplastic lesions.
- Sensitivity of each endoscopist and in each arm in the optical diagnosis of neoplastic or non-neoplastic lesions with high confidence level
Measured: Through study completion, an average of 1 year
Secondary outcomes (12)
The reliance level on artificial intelligence, measured using the C value of the signal detection theory
Measured by the C value of the signal detection theory
Measured: Through study completion, an average of 1 year
Discrimination (d´) level of neoplastic lesions based on the signal detection theory
Measured by the d' value of the signal detection theory
Measured: Through study completion, an average of 1 year
Receiver Operating characteristic (ROC) curve to determine overall discrimination in the signal detection theory
Overall discrimination assessment by the d' measured by the Receiver Operating characteristic Curve
Measured: Through study completion, an average of 1 year
Proportion of high confidence diagnosis
Proportion of lesions classified as neoplastic or non-neoplastic with high confidence diagnosis
Measured: Through study completion, an average of 1 year
Association between the reliance level (C value) on AI and endoscopists age,sex, level of expertise in colonoscopy or CADx, confidence level and area of procedence
Association between the reliance level (i.e. the response bias (c) that measures the shift regards to the ideal observer or criteria of the responses) on AI and endoscopists' age, sex, level of expertise in colonoscopy, CADx, confidence level and area of procedence in optical diagnosis.
Measured: Through study completion, an average of 1 year
Specificity of the optical diagnosis for neoplastic lesions.
We are going to calculate the specificity of each endoscopist in each arm for the optical diagnosis of neoplastic o non-neoplastic lesions with high confidence level
Measured: Through study completion, an average of 1 year
Positive and negative predictive values and accuracy of the optical diagnosis for neoplastic lesions.
We are going to calculate the positive, negative and accuracy of the optical diagnosis of neoplastic and non-neoplastic lesions for each endoscopist and in each arm
Measured: Through study completion, an average of 1 year
Sensitivity of the optical diagnosis for neoplastic lesions by endoscopists' age, sex, level of expertise in colonoscopy and CADx and area of procedence.
Sensitivity sub analysis by endoscopist age, sex, level of expertise in colonoscopy and CADx,and area of procedence
Measured: Through study completion, an average of 1 year
Specificity of the optical diagnosis for neoplastic lesions by endoscopists' age, sex, level of expertise in colonoscopy and CADx and area of procedence.
Specificity sub-analysis by the endoscopist age, sex, level of expertise in colonoscopy and CADx and area of procedence
Measured: Through study completion, an average of 1 year
Positive and negative predictive values and accuracy of the optical diagnosis for neoplastic lesions by endoscopists' age, sex, level of expertise in colonoscopy and CADx and area of procedence.
Sub-analysis by positive predictive value, negative predictive value and accuracy for the optical diagnosis of neoplasia or non-neoplasia by endoscopists age, sex, level of expertise in colonoscopy and CADx and area of procedence
Measured: Through study completion, an average of 1 year
Association between endoscopists' sensitivity in optical diagnosis of neoplasia and their reliance level on CADx.suggestions
Measured: Through study completion, an average of 1 year
Survey responses.
The endoscopists of both arms will answer the following three questions in the RedCap application with yes / no response to measure how psychological interventions affect their behaviors:
- Did you have enough time to critically assess CADx suggestions?
- Did you get exhausted by this task?
- Do you think the timing of presenting CADx you experienced today is optimal in real clinical practice?
Measured: Through study completion, an average of 1 year
Timeline
From the start of the study to its completion; the mark is today. Estimated dates are the sponsor’s plan.
- First posted5 Sep 2025
- Study startOct 2026 (estimated)
- Primary completionNov 2026 (estimated)
- Study completionDec 2026 (estimated)
Resources
- Official record on ClinicalTrials.govThe registry entry this page is built from, with its full change history.
- Publications (13)Papers the sponsor has linked to this study.
- Other trials for Polyps ColorectalStudies of the same condition, open and under way.
- What the trial phases meanHow a treatment moves from first tests in people to approval.
More details
For doctors and researchers: how the study is designed, who runs it, and its registry IDs.
Study design
- Study type
- Interventional
- Phase
- Phase N/A
- Primary purpose
- Diagnosis
- Allocation
- Randomised
- Intervention model
- Parallel groups
- Model description
- The investigators will conduct an ex vivo image-interpretation study where the endoscopists are going to watch 100 short colonoscopy videos of 100 diminutive (\<=5mm) colonic lesions and make optical diagnoses of them. The endoscopists will be given support of computer-aided diagnosis (CADx) before making their decision. The endoscopists will be randomized to the following two arms with a 1:1 ratio, and they are given CADx support accordingly: * Intervention group: CADx suggestions will be shown in the last 3 seconds of the 15 second polyp video. * Control group: CADx suggestions will be shown in real-time throughout the playback of the 15 second polyp video. The endoscopists are going to classify the lesions in the videos as neoplastic (i.e. cancer, adenomas, serrated lesions) or non-neoplastic (i.e. hyperplastic lesions and non-epithelial neoplastic). In each diagnosis, the endoscopists give their confidence level (i.e. high or low) in optical diagnosis.
- Masking
- Single blind
- Who is masked
- Outcomes assessor
- Enrolment
- 70 participants (estimated)
Sponsor and investigators
- Lead sponsor
- Fundacin Biomedica Galicia Sur
- Sponsor type
- University, hospital or other
- Responsible party
- Sponsor
- Investigators
- Yuichi Mori, MD, PhD, Principal investigator, Clinical Effectiveness Research groupPedro Davila Piñón, Masters degree biotechnology, Principal investigator, Research Group in Gastrointestinal Oncology Ourense / Galicia-Sur Public Galician FoundationJoaquin Cubiella, MD, PhD, Principal investigator, University Hospital of Ourense
IDs, dates and publications
- ClinicalTrials.gov ID
- NCT07158203
- Sponsor's protocol ID
- 2025/055
- Acronym
- FAIR
- Status last verified
- Sep 2026
- First submitted
- 29 Apr 2025
- Last updated
- 3 Sep 2026
- Results posted
- No
- Data monitoring committee
- No
- FDA-regulated drug
- No
- FDA-regulated device
- Yes
- Individual data sharing
- No
Publications
Zana Buçinca, Maja Barbara Malaya, and Krzysztof Z. Gajos. 2021. To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on AI in AI-assisted Decision-making. Proc. ACM Hum.-Comput. Interact. 5, CSCW1, Article 188 (April 2021), 21 pages. https://doi.org/10.1145/3449287
Kunar MA, Watson DG. Framing the fallibility of Computer-Aided Detection aids cancer detection. Cogn Res Princ Implic. 2023 May 24;8(1):30. doi: 10.1186/s41235-023-00485-y.
PubMed 37222932Kaminski MF, Anderson J, Valori R, Kraszewska E, Rupinski M, Pachlewski J, Wronska E, Bretthauer M, Thomas-Gibson S, Kuipers EJ, Regula J. Leadership training to improve adenoma detection rate in screening colonoscopy: a randomised trial. Gut. 2016 Apr;65(4):616-24. doi: 10.1136/gutjnl-2014-307503. Epub 2015 Feb 10.
PubMed 25670810Mori Y, Kudo SE, Chiu PW, Singh R, Misawa M, Wakamura K, Kudo T, Hayashi T, Katagiri A, Miyachi H, Ishida F, Maeda Y, Inoue H, Nimura Y, Oda M, Mori K. Impact of an automated system for endocytoscopic diagnosis of small colorectal lesions: an international web-based study. Endoscopy. 2016 Dec;48(12):1110-1118. doi: 10.1055/s-0042-113609. Epub 2016 Aug 5.
PubMed 27494455Mori Y, Jin EH, Lee D. Enhancing artificial intelligence-doctor collaboration for computer-aided diagnosis in colonoscopy through improved digital literacy. Dig Liver Dis. 2024 Jul;56(7):1140-1143. doi: 10.1016/j.dld.2023.11.033. Epub 2023 Dec 16.
PubMed 38105144Meinikheim M, Mendel R, Palm C, Probst A, Muzalyova A, Scheppach MW, Nagl S, Schnoy E, Rommele C, Schulz DAH, Schlottmann J, Prinz F, Rauber D, Ruckert T, Matsumura T, Fernandez-Esparrach G, Parsa N, Byrne MF, Messmann H, Ebigbo A. Influence of artificial intelligence on the diagnostic performance of endoscopists in the assessment of Barrett's esophagus: a tandem randomized and video trial. Endoscopy. 2024 Sep;56(9):641-649. doi: 10.1055/a-2296-5696. Epub 2024 Mar 28.
PubMed 38547927Stanislaw H, Todorov N. Calculation of signal detection theory measures. Behav Res Methods Instrum Comput. 1999 Feb;31(1):137-49. doi: 10.3758/bf03207704.
PubMed 10495845Kim J, Lim SH, Kang HY, Song JH, Yang SY, Chung GE, Jin EH, Choi JM, Bae JH. Impact of 3-second rule for high confidence assignment on the performance of endoscopists for the real-time optical diagnosis of colorectal polyps. Endoscopy. 2023 Oct;55(10):945-951. doi: 10.1055/a-2073-3411. Epub 2023 May 12.
PubMed 37172938Jin EH, Lee D, Bae JH, Kang HY, Kwak MS, Seo JY, Yang JI, Yang SY, Lim SH, Yim JY, Lim JH, Chung GE, Chung SJ, Choi JM, Han YM, Kang SJ, Lee J, Chan Kim H, Kim JS. Improved Accuracy in Optical Diagnosis of Colorectal Polyps Using Convolutional Neural Networks with Visual Explanations. Gastroenterology. 2020 Jun;158(8):2169-2179.e8. doi: 10.1053/j.gastro.2020.02.036. Epub 2020 Feb 29.
PubMed 32119927Cherubini A, Dinh NN. A Review of the Technology, Training, and Assessment Methods for the First Real-Time AI-Enhanced Medical Device for Endoscopy. Bioengineering (Basel). 2023 Mar 24;10(4):404. doi: 10.3390/bioengineering10040404.
PubMed 37106592Suna N, Koksal AS, Yildiz H, Parlak E, Kuzu UB, Yuksel M, Aydinli O, Turhan N, Sakaogullari SZ, Yalinkilic ZM, Ozin Y, Sasmaz N. Prevalence of advanced histologic features in diminutive colon polyps. Acta Gastroenterol Belg. 2015 Jul-Sep;78(3):287-91.
PubMed 26448409Vleugels JLA, Hassan C, Senore C, Cassoni P, Baron JA, Rex DK, Ponugoti PL, Pellise M, Parejo S, Bessa X, Arnau-Collell C, Kaminski MF, Bugajski M, Wieszczy P, Kuipers EJ, Melson J, Ma KH, Holman R, Dekker E, Pohl H. Diminutive Polyps With Advanced Histologic Features Do Not Increase Risk for Metachronous Advanced Colon Neoplasia. Gastroenterology. 2019 Feb;156(3):623-634.e3. doi: 10.1053/j.gastro.2018.10.050. Epub 2018 Nov 2.
PubMed 30395813Djinbachian R, Haumesser C, Taghiakbari M, Pohl H, Barkun A, Sidani S, Liu Chen Kiow J, Panzini B, Bouchard S, Deslandres E, Alj A, von Renteln D. Autonomous Artificial Intelligence vs Artificial Intelligence-Assisted Human Optical Diagnosis of Colorectal Polyps: A Randomized Controlled Trial. Gastroenterology. 2024 Jul;167(2):392-399.e2. doi: 10.1053/j.gastro.2024.01.044. Epub 2024 Feb 7.
PubMed 38331204
Study at a glance
From the registry record, last updated by the sponsor on 3 Sep 2026. Recruitment status is as the sponsor reported it: confirm with the study team.
- Official title
- Assistance for Optical Diagnosis of Neoplasia Using Artificial Intelligence (FAIR Study)
- ClinicalTrials.gov ID
- NCT07158203
- Acronym / sponsor ID
- FAIR / 2025/055
- Conditions
- Polyps Colorectal, Colonoscopy, Optical Biopsy, Colorectal Cancer Control and Prevention, Colorectal Cancer Screening, Behavior Change
- Keywords
- behaviour, optical biopsy, colonoscopy, CADx system, colorectal polyps, psychological intervention
- Study type
- Interventional
- Recruitment status
- Not yet recruiting
- Enrolment
- 70 participants (estimated)
- Lead sponsor
- Fundacin Biomedica Galicia SurUniversity, hospital or other
- Study start
- Oct 2026 (estimated)
- Primary completion
- Nov 2026 (estimated)
- Study completion
- Dec 2026 (estimated)
- First posted
- 5 Sep 2025 (actual)
- Results
- Not posted in the source record
Clinical summary
The research question as the sponsor registered it. This describes what the study plans to test, not what it found.
Computer-aided diagnosis (CADx) for colonoscopy aims to enhance optical diagnosis but often underperforms when used alongside humans due to under-reliance on AI. Psychological interventions like cognitive forcing, such as delaying CADx suggestions, may improve human-AI interaction by fostering critical assessment. However, their impact on patient-important outcomes remains unexplored.
The investigators will conduct an ex-vivo randomized study with 70 endoscopists assessing 100 polyp videos (≤5 mm) using a CADx tool (GI Genius, Medtronic). Participants will be randomized to either:
- Intervention group: CADx suggestions will be shown in the last 3 seconds of the 15 second polyp video.
- Control group: CADx suggestions will be shown in real-time throughout the playback of the 15 second polyp video.
The primary endpoint is sensitivity for high-confidence neoplasia detection, with secondary endpoints assessing endoscopists' reliance on AI.
CADx systems on the market function in various ways, such as real-time, delayed, or on-demand diagnosis. Our study aims to inform users and manufacturers whether cognitive forcing through delayed CADx suggestions enhances human-AI interaction, leading to improved clinical outcomes.
Detailed description
Computer-Aided Diagnosis and cognitive forcing Computer-aided diagnosis (CADx) for colonoscopy is expected to improve physicians' ability to predict colorectal polyp pathologies (optical diagnosis). However, recent randomized trials indicate that collaboration between humans and CADx yields lower performance than CADx alone.
This suggests suboptimal human-AI interaction due to users' under-reliance on CADx advice, favoring their inherent biases over critically assessing AI suggestions. Psychological interventions, such as cognitive forcing, aim to address this by encouraging crucial assessment of AI suggestions. One example, delayed display of CADx suggestions may promote proactive thinking by giving endoscopists enough time to consider polyp pathologies before receiving CADx suggestions, potentially leading to critical and optimal human-AI interaction. The effectiveness of such cognitive forcing was observed in experimental studies of AI for mammography reading.
However, despite its potential, no studies have evaluated the impact of such interventions on optical diagnosis accuracy in colonoscopy. To investigate the value of psychological intervention in optical diagnosis, the investigators will conduct an ex-vivo randomised controlled study.
Study aim:
This is an ex-vivo randomised controlled study. The hypothesis of our study is that cognitive forcing by delaying display of CADx suggestions facilitates physicians' critical thinking, leading to better clinical outcomes in optical diagnosis in colonoscopy.
Comercially CADx device and video details:
In this study, the investigators are going to use the comercially available CADx system (GI Genius CADx, manufactured by Cosmo Intelligent Medical Devices and distributed by Medtronic Corp). The GI Genius CADx highlights the suspicious area for polyps on-screen with bounding boxes and provides optical diagnosis prediction (i.e. adenoma, non-adenoma). All the endoscopists will be given the information that the CADx tool used in the study is GI Genius with a link to their product overview.
The investigators will collect 100 colonoscopy videos of 100 different diminutive polyps from the Polyp Image BAnk database (PIBAdb). In accordance with the real-world prevalence, 65 polyps will be neoplastic while the remaining 35 will be non-neoplastic. The duration of each video will be adjusted to contain 15 second appearance of the lesion including wite light (WL) and narrow band imaging (NBI). All the polyps should be < = 5 mm. The original videos were recorded without having any CADx interaction.
PIBAdb contains 507 videos of diminutive colorectal polyps with WL and NBI, of which 231 have a duration of 15 seconds or more. All the videos contain polyps with their histopathology available (i.e. adenoma, sessile serrated lesions, traditional serrated adenomas, invasive, hyperplastic and non-neoplastic). The polyps that have no histology category are going to be excluded from the study.
The investigators will use this database as a pool to select the 100 study videos. First, the investigators are going to split the pool of data into two groups according to polyp histology: one pool containing neoplasia (i.e. adenoma, sessile serrated lesions and traditional serrated adenoma) and the other containing non-neoplasia (i.e. hyperplastic and non-epithelial neoplastic). In each pool, 20 videos will be randomly selected as a first step. These 20 videos will be assessed if they meet our inclusion and exclusion criteria (see below), leaving only eligible videos. This selection process will be repeatedly done until the investigators collect 65 videos of polyps with neoplasia and 35 videos of polyps with non-neoplasia. The figure below shows how the video selection process takes place.
GI Genius´s specification The GI Genius system processes the videos at 50-60 frames per second and in high-definition format to have a good quality of the videos. To process the videos, the GI Genius splits them into two different streams by dedicated video card. One stream is transmitted to the first path to the output without any processing of the AI algorithm. The second stream is sent to the AI algorithm (i.e. second path) and after appropriate computation, if there is a polyp present on-screen appears an overlay in the output highlighting the polyp. The system was designed to work on unaltered WL video streams, but it works under both WL and NBI similarly.The transmission of the video stream to GI Genius is through a Serial Digital Interface (SDI) cable, acquiring the output stream from the video displaying in a computer. Finally, the SDI output stream is transmitted to the monitor containing the original video stream with additional markers superimposed on it.
How to make CADx overlaid videos In the present study, the selected videos of colorectal lesions will be stored in our high-spec computer system first. Then, these videos will be transmitted to GI Genius with an SDI cable. A capture card (DeckLink 8K Pro Mini, Blackmagic) is integrated into the computer system that converts the recorded video output to SDI signal, which allows GI Genius to process the transmitted videos.
All 100 selected videos will be processed by GI Genius in two different ways. In the first set, only the frames from the last 3 seconds of each 15-second video will be processed, and CADx suggestions will appear only in that final 3-second segment. This first set will be shown to endoscopists who are allocated to the intervention arm. In the second set, all frames from the entire 15-second duration will be processed, and CADx suggestions will be overlaid on every frame. This first set will be shown to endoscopists who are allocated to the control arm.
Study design
- Primary purpose
- Diagnosis
- Allocation
- Randomised
- Intervention model
- Parallel groups
- Masking
- Single blind (outcomes assessor)
- Number of arms
- 2
- Model description
- The investigators will conduct an ex vivo image-interpretation study where the endoscopists are going to watch 100 short colonoscopy videos of 100 diminutive (<=5mm) colonic lesions and make optical diagnoses of them. The endoscopists will be given support of computer-aided diagnosis (CADx) before making their decision. The endoscopists will be randomized to the following two arms with a 1:1 ratio, and they are given CADx support accordingly: * Intervention group: CADx suggestions will be shown in the last 3 seconds of the 15 second polyp video. * Control group: CADx suggestions will be shown in real-time throughout the playback of the 15 second polyp video. The endoscopists are going to classify the lesions in the videos as neoplastic (i.e. cancer, adenomas, serrated lesions) or non-neoplastic (i.e. hyperplastic lesions and non-epithelial neoplastic). In each diagnosis, the endoscopists give their confidence level (i.e. high or low) in optical diagnosis.
| Arm | Type | Receives |
|---|---|---|
| CADx suggestions will be shown in 15 second polyp video. | Active comparator | Device: CADx simultaneouslyCADx suggestions will be shown in real-time throughout the playback of the 15 second polyp video. |
| CADx suggestions will be shown in the last 3 seconds of the 15 second polyp video. | Experimental | Behavioral: CADx delayedCADx suggestions will be shown in the last 3 seconds of the 15 second polyp video. |
Who may qualify
- Ages
- 18 years and older
- Sex
- All
- Healthy volunteers
- Accepted
- Age groups
- Adult, Older adult
Eligibility is decided by the study team after screening; this overview does not establish that a patient qualifies. Full inclusion and exclusion criteria.
Interventions
| Intervention | Type | Arms |
|---|---|---|
| CADx simultaneouslyThe investigators showed the CADx suggestion during the 15-second playback of the video | Device | CADx suggestions will be shown in 15 second polyp video. |
| CADx delayedDuring the 15-seconds polyp video the CADx suggestion appear only in the last 3 seconds of the video | Behavioural | CADx suggestions will be shown in the last 3 seconds of the 15 second polyp video. |
Primary outcome
Sensitivity of the optical diagnosis of neoplastic lesions.
Time frame: Through study completion, an average of 1 year
- Sensitivity of each endoscopist and in each arm in the optical diagnosis of neoplastic or non-neoplastic lesions with high confidence level
12 secondary outcomes
The reliance level on artificial intelligence, measured using the C value of the signal detection theory
Time frame: Through study completion, an average of 1 year
Discrimination (d´) level of neoplastic lesions based on the signal detection theory
Time frame: Through study completion, an average of 1 year
Receiver Operating characteristic (ROC) curve to determine overall discrimination in the signal detection theory
Time frame: Through study completion, an average of 1 year
Proportion of high confidence diagnosis
Time frame: Through study completion, an average of 1 year
Association between the reliance level (C value) on AI and endoscopists age,sex, level of expertise in colonoscopy or CADx, confidence level and area of procedence
Time frame: Through study completion, an average of 1 year
Specificity of the optical diagnosis for neoplastic lesions.
Time frame: Through study completion, an average of 1 year
Positive and negative predictive values and accuracy of the optical diagnosis for neoplastic lesions.
Time frame: Through study completion, an average of 1 year
Sensitivity of the optical diagnosis for neoplastic lesions by endoscopists' age, sex, level of expertise in colonoscopy and CADx and area of procedence.
Time frame: Through study completion, an average of 1 year
Specificity of the optical diagnosis for neoplastic lesions by endoscopists' age, sex, level of expertise in colonoscopy and CADx and area of procedence.
Time frame: Through study completion, an average of 1 year
Positive and negative predictive values and accuracy of the optical diagnosis for neoplastic lesions by endoscopists' age, sex, level of expertise in colonoscopy and CADx and area of procedence.
Time frame: Through study completion, an average of 1 year
Association between endoscopists' sensitivity in optical diagnosis of neoplasia and their reliance level on CADx.suggestions
Time frame: Through study completion, an average of 1 year
Survey responses.
Time frame: Through study completion, an average of 1 year
Results and safety
This study is still under way. Results are not available in the source record yet.
Research links
- Official record on ClinicalTrials.govLast updated by the sponsor 3 Sep 2026.
- BackgroundZana Buçinca, Maja Barbara Malaya, and Krzysztof Z. Gajos. 2021. To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on AI in AI-assisted Decision-making. Proc. ACM Hum.-Comput. Interact. 5, CSCW1, Article 188 (April 2021), 21 pages. https://doi.org/10.1145/3449287
- Background: PubMed 37222932Kunar MA, Watson DG. Framing the fallibility of Computer-Aided Detection aids cancer detection. Cogn Res Princ Implic. 2023 May 24;8(1):30. doi: 10.1186/s41235-023-00485-y.
- Background: PubMed 25670810Kaminski MF, Anderson J, Valori R, Kraszewska E, Rupinski M, Pachlewski J, Wronska E, Bretthauer M, Thomas-Gibson S, Kuipers EJ, Regula J. Leadership training to improve adenoma detection rate in screening colonoscopy: a randomised trial. Gut. 2016 Apr;65(4):616-24. doi: 10.1136/gutjnl-2014-307503. Epub 2015 Feb 10.
- Background: PubMed 27494455Mori Y, Kudo SE, Chiu PW, Singh R, Misawa M, Wakamura K, Kudo T, Hayashi T, Katagiri A, Miyachi H, Ishida F, Maeda Y, Inoue H, Nimura Y, Oda M, Mori K. Impact of an automated system for endocytoscopic diagnosis of small colorectal lesions: an international web-based study. Endoscopy. 2016 Dec;48(12):1110-1118. doi: 10.1055/s-0042-113609. Epub 2016 Aug 5.
- Background: PubMed 38105144Mori Y, Jin EH, Lee D. Enhancing artificial intelligence-doctor collaboration for computer-aided diagnosis in colonoscopy through improved digital literacy. Dig Liver Dis. 2024 Jul;56(7):1140-1143. doi: 10.1016/j.dld.2023.11.033. Epub 2023 Dec 16.
- Background: PubMed 38547927Meinikheim M, Mendel R, Palm C, Probst A, Muzalyova A, Scheppach MW, Nagl S, Schnoy E, Rommele C, Schulz DAH, Schlottmann J, Prinz F, Rauber D, Ruckert T, Matsumura T, Fernandez-Esparrach G, Parsa N, Byrne MF, Messmann H, Ebigbo A. Influence of artificial intelligence on the diagnostic performance of endoscopists in the assessment of Barrett's esophagus: a tandem randomized and video trial. Endoscopy. 2024 Sep;56(9):641-649. doi: 10.1055/a-2296-5696. Epub 2024 Mar 28.
- Background: PubMed 10495845Stanislaw H, Todorov N. Calculation of signal detection theory measures. Behav Res Methods Instrum Comput. 1999 Feb;31(1):137-49. doi: 10.3758/bf03207704.
- Background: PubMed 37172938Kim J, Lim SH, Kang HY, Song JH, Yang SY, Chung GE, Jin EH, Choi JM, Bae JH. Impact of 3-second rule for high confidence assignment on the performance of endoscopists for the real-time optical diagnosis of colorectal polyps. Endoscopy. 2023 Oct;55(10):945-951. doi: 10.1055/a-2073-3411. Epub 2023 May 12.
- Background: PubMed 32119927Jin EH, Lee D, Bae JH, Kang HY, Kwak MS, Seo JY, Yang JI, Yang SY, Lim SH, Yim JY, Lim JH, Chung GE, Chung SJ, Choi JM, Han YM, Kang SJ, Lee J, Chan Kim H, Kim JS. Improved Accuracy in Optical Diagnosis of Colorectal Polyps Using Convolutional Neural Networks with Visual Explanations. Gastroenterology. 2020 Jun;158(8):2169-2179.e8. doi: 10.1053/j.gastro.2020.02.036. Epub 2020 Feb 29.
- Background: PubMed 37106592Cherubini A, Dinh NN. A Review of the Technology, Training, and Assessment Methods for the First Real-Time AI-Enhanced Medical Device for Endoscopy. Bioengineering (Basel). 2023 Mar 24;10(4):404. doi: 10.3390/bioengineering10040404.
- Background: PubMed 26448409Suna N, Koksal AS, Yildiz H, Parlak E, Kuzu UB, Yuksel M, Aydinli O, Turhan N, Sakaogullari SZ, Yalinkilic ZM, Ozin Y, Sasmaz N. Prevalence of advanced histologic features in diminutive colon polyps. Acta Gastroenterol Belg. 2015 Jul-Sep;78(3):287-91.
- Background: PubMed 30395813Vleugels JLA, Hassan C, Senore C, Cassoni P, Baron JA, Rex DK, Ponugoti PL, Pellise M, Parejo S, Bessa X, Arnau-Collell C, Kaminski MF, Bugajski M, Wieszczy P, Kuipers EJ, Melson J, Ma KH, Holman R, Dekker E, Pohl H. Diminutive Polyps With Advanced Histologic Features Do Not Increase Risk for Metachronous Advanced Colon Neoplasia. Gastroenterology. 2019 Feb;156(3):623-634.e3. doi: 10.1053/j.gastro.2018.10.050. Epub 2018 Nov 2.
- Individual participant dataSharing planned: No
Related trials for Polyps Colorectal
Recruiting
Five or Ten Year Colonoscopy for 1-2 Non-Advanced Adenomatous Polyps
Adenocarcinoma of the Colon, Adenocarcinoma of the Rectum
Recruiting
Artificial Intelligence to Implement Cost-saving Strategies for Colonoscopy Screening Based on in Vivo Prediction of Polyp Histology
Colonic Neoplasms
RecruitingObservational study
KLF4 Epigenetic Modulation
KLF4 Methylation, Epigenetic
RecruitingObservational study
Malignant Polyp Registry
Malignant Colorectal Polyp, Colorectal Cancer
Not yet recruiting
Social Software-Based Intelligent Management System for Bowel Preparation
Colorectal Polyp, Colorectal Adenoma
Referral and contact
AzTrials does not enrol participants or take referrals. Contact the study team or a site listed below; they decide who can take part.
Pedro Davila Piñón, Master degree in biotechnology contact
Yuichi Mori, MD, PhD principal investigator
Clinical Effectiveness Research group
Pedro Davila Piñón, Masters degree biotechnology principal investigator
Research Group in Gastrointestinal Oncology Ourense / Galicia-Sur Public Galician Foundation
Joaquin Cubiella, MD, PhD principal investigator
University Hospital of Ourense
2 sites in 2 countries: see every site and its status.
Questions about this trial
What is this clinical trial studying?
A study testing CADx simultaneously and CADx delayed in people with Polyps Colorectal and Colonoscopy, run by Fundacin Biomedica Galicia Sur at 2 sites in 2 countries.
The full registered description is in the About this study section and the official record.
Who may be eligible to take part?
Eligibility is decided by the study team after screening, so read the full criteria and contact a listed site.
Is this clinical trial currently recruiting?
The record says: opening soon, as of the sponsor's last update on 3 Sep 2026. Status can change between updates, so confirm with the study contact before making plans.
Where is the study taking place?
The record lists 2 sites in Norway, Spain. Whether a site is taking part right now should be confirmed with that site. See every site.
What are the possible benefits and risks?
They depend on the study and the treatment, and no study can promise a benefit. The study team explains the known risks during informed consent, before anyone joins; the record itself does not list them.
Are results available for this study?
This study is still under way. Results are not available in the source record yet.
How can a doctor refer a patient or contact the research team?
Contact Pedro Davila Piñón, Master degree in biotechnology (+34 988385047 ext. 285047, pedro.davila@iisgaliciasur.es), or a site near the patient. AzTrials does not enrol participants or take referrals.
What does the primary outcome measure?
Sensitivity of the optical diagnosis of neoplastic lesions. (time frame: Through study completion, an average of 1 year). It is what the study was designed to measure, not a finding.
Where can I read the original study information?
On the ClinicalTrials.gov record for NCT07158203, which also lists 13 linked publications.
Does a listing mean the treatment is approved or proven to work?
No. A listing means the study is registered. It does not show that the treatment is safe, works, or is approved by a regulator.
How often is this information updated?
AzTrials copies changed records from ClinicalTrials.gov every night and reloads the whole registry every week. The sponsor last updated this record on 3 Sep 2026.