Skip to content

Construction and Evaluation of Evaluation Dataset for Medical Large Model of Difficult and Severe Cases in Clinical Nutrition Department

Construction and Evaluation of Evaluation Dataset for Medical Large Model of Difficult and Severe Cases in Clinical Nutrition Department

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600129656
Enrollment
Unknown
Registered
2026-08-07
Start date
2026-08-10
Completion date
Unknown
Last updated
2026-08-10

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Malnutrition

Interventions

Wasting and Nutrient Deficiency Disorders: Protein-energy malnutrition (PEM), sarcopenia, cancer cachexia, short bowel syndrome, malabsorption secondary to inflammatory bowel disease. Excess and Metab

Sponsors

the first hospital of hebei medical university
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1.Definite diagnosis: Cases comply with internationally accepted classification criteria for diseases managed by the Department of Clinical Nutrition, and the diagnoses are verified through expert consensus within the department or long-term follow-up. 2. Complex clinical profile: Cases satisfy at least two of the following conditions: Difficult diagnostic workup: Differential diagnosis involves infections and multiple comorbidities; Highly challenging therapeutic decision-making: Refractory cases with poor response to conventional treatment; Cases requiring selection and sequential arrangement of multiple medications; Cases complicated by severe adverse drug reactions or other intractable complications during treatment; Conflicting multisystem management issues: Atypical or contradictory autoantibody test results that require in-depth interpretation combined with clinical phenotypes.

Exclusion criteria

Exclusion criteria: 1. Cases with indefinite diagnosis that have not been validated by expert consensus or long-term follow-up; 2. Cases with incomplete clinical data missing key medical history, laboratory tests, imaging records or diagnosis and treatment records; 3.Routine cases with simple clinical scenarios without complicated diagnostic or therapeutic decision-making demands; 4.Cases without complete desensitization processing that carry risks of privacy leakage.

Design outcomes

Primary

MeasureTime frame
Kappa value;Differences in Model Capabilities;Exam Difficulty and Discrimination;

Countries

China

Contacts

Public ContactLi Zengning

the first hospital of hebei medical university

lizengning@126.com+86 311 87156512

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Aug 25, 2026