Screening and Identifying Hepatobiliary Diseases Via Deep Learning Using Ocular Images

Artificial Intelligence may provide insight into exploring the potential covert association behind and reveal some early ocular architecture changes in individuals with hepatobiliary disorders. We conducted a pioneer work to explore the association between the eye and liver via deep learning, to develop and evaluate different deep learning models to predict the hepatobiliary disease by using ocular images.

Study Overview

• Status: Completed
• Conditions: Artificial Intelligence; Hepatobiliary Disease; Ophthalmology
• Intervention / Treatment: Diagnostic test: Hepatobiliary Disorders

• Study Type: Observational
• Enrollment (Actual): 1789

Contacts and Locations

Study Locations

• China
  • Guangdong
    • Guangzhou, Guangdong, China, 510000
      • Zhongshan Ophthalmic Center, Sun Yat-sen Univerisity

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: ADULT; OLDER_ADULT; CHILD
• Accepts Healthy Volunteers: Yes
• Genders Eligible for Study: All

• Sampling Method: Probability Sample

Study Population

Slit-lamp and retinal fundus images collected from Department of Infectious Diseases, Third Affiliated Hospital of Sun Yat-sen University, Department of Infectious Diseases, Third Affiliated Hospital of Sun Yat-sen University, Affiliated Huadu Hospital of Southern Medical University, Nantian Medical Centre of Aikang Health Care, and Huanshidong Medical Centre of Aikang Health Care.

Description

Inclusion Criteria:

• The quality of fundus and slit-lamp images should clinical acceptable.
• More than 90% of the fundus image area including four main regions (optic disk, macular, upper and lower retinal vessel archs) are easy to read and discriminate.
• More than 90% of the slit-lamp image area including three main regions (sclera, pupil, and lens) are easy to read and discriminate.

Exclusion Criteria:

• Images with light leakage (>10% of the area), spots from lens flares or stains, and overexposure were excluded from further analysis.

Study Plan

How is the study designed?

Number of groups / cohorts

5

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
development dataset 01; Slit-lamp and retinal fundus images collected from Department of Hepatobiliary Surgery of the Third Affiliated Hospital of Sun Yat-sen University.	Diagnostic test: Hepatobiliary Disorders; The training dataset was used to train the deep learning model, which was validated and tested by the other two datasets.
development dataset 02; Slit-lamp and retinal fundus images collected from Affiliated Huadu Hospital of Southern Medical University.	Diagnostic test: Hepatobiliary Disorders; The training dataset was used to train the deep learning model, which was validated and tested by the other two datasets.
development dataset 03; Slit-lamp and retinal fundus images collected from Nantian Medical Centre of Aikang Health Care.	Diagnostic test: Hepatobiliary Disorders; The training dataset was used to train the deep learning model, which was validated and tested by the other two datasets.
test dataset 01; Slit-lamp and retinal fundus images collected from Department of Infectious Diseases, Third Affiliated Hospital of Sun Yat-sen University.	Diagnostic test: Hepatobiliary Disorders; The training dataset was used to train the deep learning model, which was validated and tested by the other two datasets.
test dataset 02; Slit-lamp and retinal fundus images collected from Huanshidong Medical Centre of Aikang Health Care.	Diagnostic test: Hepatobiliary Disorders; The training dataset was used to train the deep learning model, which was validated and tested by the other two datasets.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
area under the receiver operating characteristic curve of the deep learning system	The investigators will calculate the area under the receiver operating characteristic curve of deep learning system and compare this index between deep learning system and human doctors	baseline

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
sensitivity and specificity of the deep learning system	The investigators will calculate the sensitivity and specifity of deep learning system and compare this index between deep learning system and human doctors	baseline

Collaborators and Investigators

• Sponsor: Sun Yat-sen University
• Collaborators: Third Affiliated Hospital, Sun Yat-Sen University; Affiliated Huadu Hospital of Southern Medical University; Aikang Health Care

Study record dates

Study Major Dates

• Study Start (ACTUAL): December 1, 2018
• Primary Completion (ACTUAL): January 31, 2020
• Study Completion (ACTUAL): January 31, 2020

Study Registration Dates

• First Submitted: December 25, 2019
• First Submitted That Met QC Criteria: December 25, 2019
• First Posted (ACTUAL): December 30, 2019

Study Record Updates

• Last Update Posted (ACTUAL): August 18, 2020
• Last Update Submitted That Met QC Criteria: August 16, 2020
• Last Verified: August 1, 2020

More Information

Terms related to this study

• Keywords: Artificial Intelligence; Hepatobiliary Disease; Eye Images
• Additional Relevant MeSH Terms: Liver Diseases; Digestive System Diseases
• Other Study ID Numbers: AEHD-2019

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No