Evaluation of Decision Compatibility Between a Multidisciplinary Cancer Board and ChatGPT-4O

NCT06986564 | Completed DMC

• 1 other identifier: 2024-050695
• Study Type: observational
• Target: 100
• Locations: 1 country
• Sites: 1

Brief Summary

This study aims to compare the decisions of a multidisciplinary cancer board at Van research and educational Hospital with those generated by ChatGPT-4O. Patient data, including age, gender, comorbidities, clinical findings, laboratory results, radiology, and pathology reports, will be collected and recorded in patient files. These records will then be converted into a textual format appropriate for input into the ChatGPT-4O model. Statistical analyses will be conducted to assess the level of agreement between the board's decisions and those of ChatGPT-4O, providing insights into their compatibility and potential utility in clinical practice.

Conditions

Cancer; Artificial Intelligence

Keywords

Clinical decision agreement study; Cancer; Artificial intelligence

Outcome Measures

Primary Outcomes (1)

• to evaluate the agreement between decisions made by a multidisciplinary cancer board and those generated by the ChatGPT-4O model using kappa statistics.

  Within the scope of this research, we aimed to elucidate the compatibility of artificial intelligence with the decisions made in patients brought to the multidisciplinary tumor council. We sought to answer whether we can accept artificial intelligence as a new member, other than humans, as a decision-support tool in the council.

  4 month

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

A total of 100 cancer patients presented to the multidisciplinary cancer board will be included in the study. Patient data, including age, gender, comorbidities, clinical findings, laboratory results, radiology, and pathology reports, will be systematically recorded and reformatted into text suitable for the ChatGPT-4O model. The AI model's decisions will then be compared to those made by the cancer board. Statistical analyses will be performed to quantify the level of agreement between the two decision sources.

You may qualify if:

• above 18 y/o
• patients diagnosed with malignancy

You may not qualify if:

• patients who refused to consent

Sponsors & Collaborators

• Sebahattin Celik MD: lead

Study Sites (1)

Van state hospital, general surgery department

Van, 6500, Turkey (Türkiye)

Location

MeSH Terms

Conditions

Neoplasms

Study Officials

• Sebahattin Çelik, professor

  General surgery department chief

  STUDY DIRECTOR

Study Design

• Study Type: observational
• Observational Model: OTHER
• Time Perspective: PROSPECTIVE
• Target Duration: 3 Months
• Sponsor Type: OTHER
• Responsible Party: SPONSOR INVESTIGATOR
• PI Title: Professor

Study Record Dates

• First Submitted: December 20, 2024
• First Posted: May 23, 2025
• Study Start: November 1, 2024
• Primary Completion: January 30, 2025
• Study Completion: May 1, 2025
• Last Updated: May 23, 2025

Record last verified: 2025-05

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, SAP

Locations

TU (1)