This prospective, single-center observational study will evaluate the concordance between a multimodal artificial intelligence system and multidisciplinary team decisions in patients with gastrointestinal tumors. For each enrolled case, the AI system and the MDT will independently review the same available clinical information, including medical history, laboratory findings, imaging, pathology, and other relevant diagnostic data, and will generate recommendations regarding diagnosis, staging, treatment planning, and further examinations. An independent expert panel will assess the agreement, clinical appropriateness, and potential major errors of the two decision pathways. AI-generated recommendations will be used solely for research evaluation and will not directly influence patient care. The study aims to determine the feasibility, reliability, and safety of multimodal AI-assisted decision-making in real-world gastrointestinal oncology workflows.
Aged 18 years or older, with no restriction based on sex.
Histologically confirmed gastric, colon, or rectal cancer; or highly suspected gastric, colon, or rectal malignancy based on clinical, endoscopic, or imaging findings, with further diagnostic, staging, or treatment planning required.
Scheduled for multidisciplinary team evaluation at the study center because of an initial diagnosis, clinical staging, perioperative treatment, surgical planning, recurrence or metastasis, conversion therapy, treatment response assessment, or another complex clinical management question.
A clearly identifiable clinical decision time point can be established, and the clinical information available before that time point can be defined.
Further examination confirms a condition other than gastric, colon, or rectal cancer, and the case is not applicable to the gastrointestinal oncology pathways evaluated in this study.
The participant is undergoing routine follow-up or continuation of a previously established treatment plan and has no clinical question requiring additional diagnostic tasks, modification of the principal treatment pathway, or multidisciplinary team decision-making.
A clear clinical decision time point cannot be established, or information available before and after the decision cannot be distinguished, preventing comparison of the artificial intelligence system and the multidisciplinary team under the same information conditions.
Core clinical information is severely incomplete, such that the current stage of care and the clinical question cannot be identified and a clinically meaningful artificial intelligence output cannot be generated.
Consecutive adult patients with suspected or confirmed gastric, colon, or rectal cancer who are scheduled for multidisciplinary team evaluation at Ruijin Hospital because of a clinical question related to diagnosis, staging, perioperative treatment, surgical planning, recurrence or metastasis, conversion therapy, response assessment, or another complex management decision. Eligible participants must have an identifiable clinical decision time point and sufficient clinical information to permit paired evaluation by the multimodal artificial intelligence system and the multidisciplinary team.
Cohort
Prospective
Multidisciplinary Team Clinical Decision Assessment
Other interventionUsing the same clinical information available at the predefined decision time point, the institutional multidisciplinary team will independently formulate recommendations regarding diagnosis, staging, additional examinations, and treatment planning according to routine clinical practice. Clinical management will remain under the responsibility of the treating physicians and the multidisciplinary team. The MDT decision and the independently generated AI output will subsequently be compared and evaluated by an independent expert panel.
Multimodal Artificial Intelligence-Based Clinical Decision Assessment
Other interventionAvailable clinical information for each enrolled participant, including medical history, laboratory findings, endoscopic findings, imaging, pathology, and molecular testing results when available, will be entered into a multimodal artificial intelligence system. The system will independently generate a structured assessment of diagnosis, staging, additional diagnostic tasks, and treatment planning. The AI-generated output will be recorded solely for research comparison, will not be disclosed to the treating multidisciplinary team before its decision is finalized, and will not directly influence patient care.
Clinical Acceptability Concordance Rate Between the Multimodal Artificial Intelligence System and the Multidisciplinary Team
The proportion of evaluable participants whose AI-generated diagnostic task plan and principal treatment recommendation are judged by an independent expert panel to be fully concordant with the MDT decision or a clinically acceptable alternative. Results will be reported as a percentage with a 95% confidence interval.
Full Concordance Rate
The proportion of evaluable participants whose AI-generated diagnostic task plan and principal treatment recommendation are judged to be fully concordant with the MDT decision.
Major Discordance Rate
The proportion of evaluable participants whose AI-generated recommendation is judged to contain a major discordance with potential for clinically significant harm.
Successful AI Output Rate
The proportion of enrolled participants for whom the AI system successfully generates an evaluable diagnostic task plan and principal treatment recommendation.
Distribution of AI-MDT Discordance Types
The number and percentage of discordant cases classified by type, including diagnostic or staging errors, inappropriate additional testing, treatment selection errors, treatment sequencing errors, and potential safety risks.
Click on the lead sponsor to view all of their trials.
This trial is not recruiting at the moment. You can still explore other options:
