AI-Assisted Colonoscopy for Colorectal Cancer Detection

Overseen By Alex Rodriguez, BS

Age: Any Age

Sex: Any

Trial Phase: Academic

Sponsor: University of Southern California

Disqualifiers: Refusal of consent, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Approved in 4 Jurisdictions This treatment is already approved in other countries

Compensation: Varies

Number of Visits: 1

Duration: 2 years

What You Need to Know Before You Apply

What is the purpose of this trial?

This trial tests whether artificial intelligence (AI) during colonoscopies helps doctors-in-training find more polyps, potentially improving colorectal cancer detection outcomes. Polyps, small growths in the colon, are crucial to find early. The trial compares two groups: one using AI technology in the endoscopy room to spot polyps and another following the standard procedure without AI. It specifically involves gastroenterology fellows at USC performing colonoscopies. As an unphased trial, this study offers a unique opportunity to contribute to innovative research that could enhance future colonoscopy procedures.

Do I need to stop my current medications for this trial?

The trial protocol does not specify whether participants need to stop taking their current medications.

What prior data suggests that AI use in endoscopy is safe?

Research shows that using artificial intelligence (AI) in colonoscopy is generally safe for patients. Studies have found that AI-assisted colonoscopy helps doctors find adenomas (growths that might turn into cancer) without adding any risk to patients. For instance, AI has been used in many colonoscopies and has demonstrated high accuracy, even when bowel preparation wasn't perfect. This indicates that AI can still perform well in different situations.

No reports have indicated serious problems or safety issues directly caused by using AI in these procedures. The main purpose of using AI is to help doctors find more polyps (small growths) during colonoscopies, which can improve the chances of preventing colorectal cancer.

In summary, AI-assisted colonoscopy appears safe and well-tolerated, enhancing the detection of potentially harmful growths. ^{1 2 3 4 5}

Why are researchers excited about this trial?

Researchers are excited about using AI in colonoscopy because it has the potential to enhance the detection of colorectal cancer. Unlike traditional colonoscopy procedures, which rely solely on the expertise of the medical team, AI-assisted colonoscopy employs advanced algorithms to help identify suspicious areas in the colon more accurately and quickly. This could lead to earlier detection of cancerous or pre-cancerous polyps, ultimately improving patient outcomes and potentially reducing the number of missed diagnoses. By integrating AI into the endoscopy room, there's an opportunity to set a new standard in how we screen for colorectal cancer, making the process more efficient and reliable.

What evidence suggests that AI use in endoscopy rooms is effective for improving adenoma detection rates?

Research shows that using AI during colonoscopies can greatly improve the detection of adenomas, which are polyps that can turn into cancer. In this trial, participants will receive colonoscopies in either an Artificial Intelligence Endoscopy Room or a Non-Artificial Intelligence Endoscopy Room. Recent studies have found that AI increases the rate of polyp detection, regardless of the doctor's experience. This is important because finding more adenomas can help prevent colorectal cancer. AI tools also accurately identify tumors in the digestive system compared to traditional methods. However, while AI aids in finding polyps, its overall effect on preventing colorectal cancer remains under investigation. ^{4 6 7 8 9}

Who Is on the Research Team?

James Buxbaum, MD

Principal Investigator

University of Southern California

Are You a Good Fit for This Trial?

This trial is for Gastroenterology fellows at USC who perform endoscopies. They must agree to participate and give informed consent. Procedures in intensive care or operating rooms, or those done solely by faculty without the fellow as primary operator, are excluded.

Inclusion Criteria

All Gastroenterology fellows at USC performing Endoscopies will be included in the study.

Exclusion Criteria

Procedures performed only by faculty, in which the fellow is not the primary operator, will not be used for study metrics.

Fellows who refuse informed consent will be excluded

My procedure will be in the endoscopy unit, not in intensive care or the operating room.

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Educational Session

Fellows undergo an educational session on quality metrics and AI software usage

1 week

1 visit (in-person)

Treatment

Fellows perform colonoscopies with and without AI to measure adenoma detection rate

2 years

Follow-up

Participants are monitored for safety and effectiveness after treatment

4 weeks

What Are the Treatments Tested in This Trial?

Interventions

• AI use in Endoscopy Room
• Non-AI use Standard of Care endoscopy room

Trial Overview The study aims to see if using AI during colonoscopy helps trainees find more polyps compared to standard methods without AI. It's a randomized test where some will use AI assistance and others won't, measuring the adenoma detection rate (ADR).

How Is the Trial Designed?

Group I: Non-Artificial Intelligence Endoscopy Room Active Control 1 Intervention

The fellows will be randomized on a daily basis to perform colonoscopies in a non-AI endoscopy room (standard of care).

Group II: Artificial Intelligence Endoscopy Room Active Control 1 Intervention

The fellows will be randomized on a daily basis to perform colonoscopies in a room with AI (intervention)

AI use in Endoscopy Room is already approved in European Union, United States, Japan, Canada for the following indications:

Approved in European Union as AIAC for:

• Colorectal cancer screening
• Polyp detection

Approved in United States as AIAC for:

• Colorectal cancer screening
• Polyp detection

Approved in Japan as AIAC for:

• Colorectal cancer screening
• Polyp detection

Approved in Canada as AIAC for:

• Colorectal cancer screening
• Polyp detection

Who Is Running the Clinical Trial?

University of Southern California

Lead Sponsor

Trials

956

Recruited

1,609,000+

Published Research Related to This Trial

Artificial intelligence-assisted colonoscopy (AIAC) significantly reduces the adenoma missed detection rate (AMR), with a decrease from 28.8% in the control group to 13.2% in the AIAC group, indicating improved diagnostic accuracy.

AIAC also enhances the adenoma detection rate (ADR), increasing from 28.7% in the control group to 36.4% in the AIAC group, suggesting that this technology can help detect more adenomas during colonoscopy.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Effects of ai-assisted colonoscopy on adenoma miss rate/adenoma detection rate: A protocol for systematic review and meta-analysis. Shao, L., Yan, X., Liu, C., et al.[2023]

The use of computer-aided detection (CAD) significantly improved the sensitivity of polyp detection in CT colonography by less-experienced radiologists, increasing average sensitivity from 81.0% to 90.8%.

While CAD increased the number of false-positive results from 0.70 to 0.96 per patient, the overall improvement in detection rates suggests that CAD can enhance diagnostic accuracy for polyp identification.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Computer-aided detection of colorectal polyps: can it improve sensitivity of less-experienced readers? Preliminary findings. Baker, ME., Bogoni, L., Obuchowski, NA., et al.[2016]

In a study comparing 213 AI-assisted colonoscopies (AIAC) to 213 conventional colonoscopies (CC), the AIAC group had a significantly higher adenoma detection rate (ADR) of 47.9% compared to 38.5% in the CC group, indicating improved efficacy in detecting precancerous polyps.

The AIAC group also had a longer withdrawal time (15 minutes) compared to the CC group (13 minutes), suggesting that the AI technology may encourage more thorough examinations, although the overall polyp detection rate was similar between both groups.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Artificial intelligence improves adenoma detection rate during colonoscopy. Schauer, C., Chieng, M., Wang, M., et al.[2022]

AI-assisted endoscopy can significantly improve the detection rates of lesions, helping both experienced and trainee endoscopists by identifying abnormalities that may be missed during procedures.

The effectiveness of AI in this context has been evaluated in three key areas: detecting polyps, characterizing them as adenomatous or non-adenomatous, and predicting the likelihood of invasive cancer in polypoid lesions, enhancing diagnostic accuracy without extending procedure times.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Inteligencia artificial en la colonoscopia de tamizaje y la disminución del error. Galvis-García, E., Vega-González, FJ., Emura, F., et al.[2023]

A systematic review of six randomized controlled trials involving 5058 individuals showed that AI-assisted colonoscopy (AIAC) significantly increases adenoma detection rates (ADRs) to 33.7% compared to 22.9% with conventional colonoscopy (CC), indicating a strong efficacy of AI in improving detection of potential cancerous growths.

AIAC also resulted in a higher polyp detection rate (PDR) of 45.6% versus 30.6% for CC, although it did lead to a slightly longer total withdrawal time during the procedure.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Impact of real-time use of artificial intelligence in improving adenoma detection during colonoscopy: A systematic review and meta-analysis. Ashat, M., Klair, JS., Singh, D., et al.[2021]

In a multicenter randomized controlled trial involving 3059 asymptomatic participants, AI-assisted colonoscopy significantly improved the overall adenoma detection rate (ADR) to 39.9% compared to 32.4% with conventional colonoscopy, indicating enhanced efficacy in detecting polyps.

The AI-assisted method also increased the detection of advanced adenomas and showed better performance for both expert and non-expert endoscopists, while slightly increasing the median withdrawal time during the procedure.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Artificial Intelligence-Assisted Colonoscopy for Colorectal Cancer Screening: A Multicenter Randomized Controlled Trial. Xu, H., Tang, RSY., Lam, TYT., et al.[2023]

In a large-volume medical center, the introduction of artificial intelligence-aided colonoscopy (AIAC) did not improve adenoma and polyp detection rates, with rates of 30.3% for ADR and 36.5% for PDR compared to 35.2% and 40.9% in the pre-AIAC period.

Despite the lower detection rates, the AIAC group experienced significantly shorter procedure times, suggesting that while AIAC may streamline the process, it did not enhance the effectiveness of polyp detection.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Artificial Intelligence-Aided Colonoscopy Does Not Increase Adenoma Detection Rate in Routine Clinical Practice. Levy, I., Bruckmayer, L., Klang, E., et al.[2023]

Citations

1. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC10047293/

Artificial Intelligence-Aided Endoscopy and Colorectal ... The aim of this review is to summarize the current knowledge on AI-aided endoscopy, and to outline its potential role in colorectal cancer prevention.

2. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC8704267/

Artificial intelligence-assisted colonoscopy: A review of ... Several recently developed applications for AI-assisted colonoscopy have shown promising results for the detection and classification of colorectal polyps and ...

3. giejournal.org giejournal.org/article/S0016-5107(24)03471-0/fulltext

Use of artificial intelligence improves colonoscopy ... AI-assisted colonoscopy significantly improved adenoma detection but not sessile serrated lesion detection irrespective of endoscopist experience, system type, ...

4. ghadvances.org ghadvances.org/article/S2772-5723(25)00141-4/fulltext

Artificial Intelligence–Assisted Endoscopy in Diagnosis of ... AI-assisted endoscopy demonstrated a significantly higher detection rate for GI tumors compared to conventional endoscopy, alongside high diagnostic accuracy ...

5. gastro.org gastro.org/news/new-guideline-ai-technology-in-colonoscopy-not-yet-certain-to-further-prevent-colorectal-cancer-beyond-the-human-eye/

New guideline: AI technology in colonoscopy not yet ... We found that artificial intelligence-assisted technology helps identify colorectal polyps; however, its impact on preventing colorectal cancer remains unclear.

6. cancer.gov cancer.gov/news-events/cancer-currents-blog/2023/colonoscopy-cad-artificial-intelligence

CAD-Aided Colonoscopy and Advanced Adenomas - NCI Overall, the addition of AI increased the rate of polyp detection, but again, this was driven by an increase in detection of small polyps, not ...

7. wjgnet.com wjgnet.com/1007-9327/full/v31/i40/111499.htm

Enhancing quality indicators for optimal patient outcomes Accuracy remained > 95% across varying bowel preparation quality. Another AI cecal intubation recognition system compared clinical metrics ...

8. karger.com karger.com/dig/article/106/2/138/918599/Current-Status-of-Artificial-Intelligence-Use-in

Current Status of Artificial Intelligence Use in Colonoscopy AI-assisted systems employing deep learning offer improved detection and differentiation of colorectal polyps, potentially increasing adenoma ...

9. sciencedirect.com sciencedirect.com/science/article/pii/S2772572325001414

Artificial Intelligence–Assisted Endoscopy in Diagnosis of ... Colonoscopy using AI significantly improved adenoma detection rates (ADRs) and polyp detection rates (PDRs) compared to non-AI colonoscopy. ADR with AI was 29.6 ...