AI-assisted Rare Disease Diagnosis

NCT07650799 | Not Yet Recruiting DMC

• 1 other identifier: PUMCH I-26PJ0002
• Study Type: interventional
• Target: 1,056
• Locations: 1 country
• Sites: 13

Brief Summary

A multicentre randomised controlled trial evaluating whether a rare-disease diagnostic large language model can improve diagnostic quality, efficiency, and health-economic outcomes for physicians managing patients with suspected rare or diagnostically unresolved disease.

Conditions

Rare Disorders; Rare Diseases

Keywords

rare diseases; AI; LLM; diagnosis; cost-effectiveness

Outcome Measures

Primary Outcomes (1)

• Overall Correct Diagnostic Yield

  The proportion of all randomised patients whose clinical diagnosis by the end of follow-up is concordant with the blinded-adjudicated final reference diagnosis determined by an independent committee.

  From the first visit to final reference diagnosis adjudication, an average of 8 weeks.

Secondary Outcomes (7)

• Candidate Diagnostic Accuracy

  From the first visit to final reference diagnosis adjudication, an average of 8 weeks.

• Molecular Diagnostic Yield

  From the first visit to final reference diagnosis adjudication, an average of 8 weeks.

• Time to a Correct Diagnosis

  From enrollment to the end of follow-up, up to 8 weeks.

• Appropriate Genetic Testing Recommendation Rate

  From the initial consultation to genetic testing indication adjudication, approximately 8 weeks

• Duration of the Initial Physician Consultation

  Assessed at each consultation (day 1), within 1 day.

Study Arms (2)

AI system

EXPERIMENTAL

AI system will be used to provide diagnostic support during the encounter in addition to conventional clinical workflow. Use of other generative AI tools is prohibited.

Other: AI system

Standard of care

NO INTERVENTION

The physician conducts the encounter per standard hospital workflow using conventional clinical resources only. Use of any generative AI tool is prohibited.

Interventions

AI system OTHER

The study AI system will be used to provide diagnostic support during the clinical encounter, including structuring relevant clinical information, generating a clinical analysis, and suggesting candidate diagnoses for review by the treating physician.

AI system

Eligibility Criteria

• Age: 0 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)

You may qualify if:

• Any age. Legal guardian co-signs consent for minors or individuals lacking legal capacity.
• Diagnostically unresolved or suspected rare disease, with at least one prior complete clinical evaluation at a secondary-level or higher institution yielding no confirmed explanatory diagnosis.
• First presentation to the enrolling institution for the current condition, with no prior records in the institutional HIS or outpatient system.
• No prior genetic testing related to the current condition; no results or reports available.
• Written informed consent provided voluntarily by patient or legal guardian, with commitment and ability to complete structured follow-up.

You may not qualify if:

• Confirmed diagnosis (clinical, pathological, or molecular) explaining the primary symptoms.
• Emergency presentation, critical illness, or any condition incompatible with trial participation.
• Neither patient nor legally authorised proxy able to complete follow-up.
• Concurrent enrollment in another interventional study with diagnostic accuracy or genetic testing yield as a primary endpoint.
• Prior use of another AI system has already yielded a confirmed diagnosis for the current condition.
• Licensed physician in internal medicine, neurology, pediatrics, general medicine, rare disease, or a related specialty.
• ≥2 years of clinical practice; competent to manage rare disease patients; stratified into junior or senior tier.
• Voluntary participation with written informed consent.
• No longer in clinical practice, or unable to fulfill required outpatient duties during the study period.
• Unwilling to provide informed consent or to permit protocol-required collection of consultation and questionnaire data.
• Currently enrolled in another AI-assisted clinical workflow, or expected to be unable to comply with the procedures.

Sponsors & Collaborators

• Peking Union Medical College Hospital: lead
• Cangzhou Central Hospital: collaborator
• Zhangzhou Municipal Hospital: collaborator
• Dongguan People's Hospital: collaborator
• First People's Hospital of Foshan: collaborator
• Guizhou Provincial People's Hospital: collaborator
• Tianjin Children's Hospital: collaborator
• The First People's Hospital of Yunnan: collaborator
• Qinghai People's Hospital: collaborator

Study Sites (13)

Peking Union Medical College Hospital

Beijing, China

Location

Cangzhou Central Hospital

Cangzhou, China

Location

Changchun Sacred Heart Hospital

Changchun, China

Location

Dongguan People's Hospital

Dongguan, China

Location

First People's Hospital of Foshan

Foshan, China

Location

Guizhou Provincial People's Hospital

Guiyang, China

Location

Jilin Central General Hospital

Jilin City, China

Location

The First People's Hospital of Yunnan Province

Kunming, China

Location

Tianjin Children's Hospital

Tianjin, China

Location

Tianshui 407 Hospital

Tianshui, China

Location

Wuhai People's Hospital

Wuhai, China

Location

Qinghai Provincial People's Hospital

Xining, China

Location

Zhangzhou Municipal Hospital of Fujian Province

Zhangzhou, China

Location

MeSH Terms

Conditions

Rare Diseases; Disease

Interventions

Intelligent Systems

Condition Hierarchy (Ancestors)

Disease Attributes; Pathologic Processes; Pathological Conditions, Signs and Symptoms

Intervention Hierarchy (Ancestors)

Artificial Intelligence; Algorithms; Mathematical Concepts

Study Officials

• Shuyang Zhang, MD, PhD

  Peking Union Medical College Hospital

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Shuyang Zhang, MD, PhD

CONTACT

+86-13911667211; shuyangzhang103@163.com

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: OUTCOMES ASSESSOR
• Masking Details: Patient level: open label. Treating physicians: open label. Outcome adjudicators (independent expert committee) and statisticians: blinded. Adjudicators do not know arm assignment and are not shown any AI-generated or AI-attributed material; statistical analysis coded A/B until unblinding.
• Purpose: DIAGNOSTIC
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: President of PUMCH

Model Details: This is a multicentre, prospective, parallel controlled study. Eligible patients will be assigned to either an AI-assisted diagnostic workflow or a standard diagnostic workflow. Participating physicians will provide care under both study conditions.

Study Record Dates

• First Submitted: June 7, 2026
• First Posted: June 16, 2026
• Study Start: August 1, 2026
• Primary Completion (Estimated): July 1, 2027
• Study Completion (Estimated): December 1, 2027
• Last Updated: July 31, 2026

Record last verified: 2026-07

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, SAP, ANALYTIC CODE
• Time Frame: Beginning 6 months after publication of the primary study results and ending 5 years after publication.
• Access Criteria: Requests must include a scientifically sound research proposal and analysis plan, and evidence of ethics approval or exemption where applicable. Access will be subject to approval by the study steering committee and participating institutions, execution of a data use agreement, and compliance with applicable ethical, legal, and data-protection requirements. Data will be made available through a secure access environment or other controlled transfer mechanism. No directly identifiable participant information will be shared.

Locations

CN (13)