Multimodal Artificial Intelligence for Non-Invasive Colorectal Cancer Screening Based on Holographic Biology

NCT07815093 — Not yet recruiting, 11000 participants, Xiuli Zuo

ClinicalTrials Daily · Areas · Conditions · Drugs · Sponsors ·

Multimodal Artificial Intelligence for Non-Invasive Colorectal Cancer Screening Based on Holographic Biology.

Registry	NCT07815093
Status	Not yet recruiting
Study type	Observational
Enrollment	11000 (Estimated)
Sponsor	Xiuli Zuo (Other)
Conditions	Colorectal Cancer
Interventions	Multimodal non-invasive detection for colorectal cancer (Other)
Dates	start 2026-10-01 (estimated), primary completion 2028-04-01 (estimated), completion 2028-07-01 (estimated), first posted 2026-09-11, last updated 2026-09-11

-Summary

The purpose of this study is to propose a non-invasive screening method by integrating holographic biological theory and artificial intelligence technology. A colorectal cancer risk assessment model will be constructed by analyzing multimodal data including facial features, tongue image characteristics and exhaled gas. The hypothesis of this study is that the model can attain both sensitivity and specificity of 80%.This study will enroll patients aged 18-80 years scheduled to undergo colonoscopy. All participants shall provide informed consent and sign the informed consent form. Patients will be excluded if they have severe cardiac, cerebral, pulmonary or renal dysfunction, or psychiatric disorders precluding colonoscopy, have a history of gastrointestinal surgery, or have taken bismuth agents or other staining medications. Based on estimates from existing literature, the training set will require tongue images, facial images, exhaled gas analysis results and colonoscopic diagnoses from 5000 patients, and the test set will require such data from 6000 patients.

Expand all sections · Collapse all

+Arms and interventions

Multimodal non-invasive detection for colorectal cancer (Other): Collection of facial images, tongue images and exhaled gas data from participants. No therapeutic intervention is performed. The collected multimodal data will be used to build an artificial intelligence-based colorectal cancer risk assessment model.

+Outcome measures

Primary outcomes

Presence of colonic polyps and colorectal cancer. [At the time of enrollment and colonoscopy examination]

+Eligibility

Sex: All, ages: 18 Years to 80 Years

Inclusion Criteria:

1. Patients aged 18-80 years scheduled to undergo colonoscopy.
2. All patients provide informed consent and sign the informed consent form.

Exclusion Criteria:

1. Patients with severe cardiac, cerebral, pulmonary or renal dysfunction, or psychiatric disorders that prevent colonoscopy;
2. Patients with a history of gastrointestinal surgery;
3. Patients taking bismuth agents or other staining medications.