NLP to Aid in the Evaluation and Diagnosis of FGIDs

• Conditions: Functional Gastrointestinal Disorders

• Registration Number: NCT05501028
• Lead Sponsor: Massachusetts General Hospital

Brief Summary

The study has two arms, where the same natural language processing (NLP) and probabilistic graphical modeling technology will be utilized on patients' report of symptoms in both arms. The clinical arm is focused on patients presenting for consultation with a gastroenterologist. The endoscopy arm is focused generally on patients presenting for a diagnostic endoscopy, with the goal of capturing Functional Gastrointestinal Disorder (FGID) patients prior to diagnosis.

Detailed Description

Not available

Study Timeline

• Study Start: 2018-08-09
• Study First Submitted: 2022-03-01
• Status Verified: 2022-08
• Study First Submitted QC: 2022-08-12
• Last Update Submitted: 2022-08-12
• Study First Posted: 2022-08-15
• Last Update Posted: 2022-08-15
• Primary Completion: 2023-06-01
• Study Completion: 2023-06-01

Recruitment & Eligibility

• Status: UNKNOWN
• Sex: All
• Target Recruitment: 700

Inclusion Criteria

• Scheduled for consult in the Motility Clinic in the MGH Gastroenterology Unit or for diagnostic endoscopy in the MGH Gastroenterology Unit
• Patient must agree to have their interactions audio-recorded
• Informed consent form signed by the subjects

Exclusion Criteria

• Non-native English speaker
• Patients unable to communicate their own symptoms

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Latent themes present in patient descriptions of FGIDs symptoms	08/09/2018-08/09/2023	Latent themes present in patient descriptions of FGIDs symptoms as generated by machine learning as well as quantitative comparisons to traditional metrics of patient descriptions including Rome IV criteria and patient descriptions of severity.

Trial Locations

Locations (1)

Massachusetts General Hospital; 🇺🇸 Boston, Massachusetts, United States