AI-based System for Lung Tuberculosis Screening: Diagnostic Accuracy Evaluation

Testing of AI solutions to assess diagnostic accuracy for tuberculosis detection.

Study Overview

• Status: Active, not recruiting
• Conditions: Tuberculosis, Pulmonary
• Intervention / Treatment: Diagnostic test: AI-based x-ray analysis and triage ("normal/tuberculosis suspected")

Detailed Description

Tuberculosis remains a key problem of modern medicine. New approaches for burden overcoming should be proposed. New screening strategies may include artificial intelligence (AI). An AI-based system for chest x-ray analysis and triage ("normal/tuberculosis suspected") have been developed and trained. A special data-set was prepared. There are 238 normal x-rays and 70 x-rays with lung tuberculosis in data-set. The data-set was randomly divided into 2 samples:

• sample N1 (n=140) with ratio "normal: tuberculosis" 50:50,
• sample N1 (n=150) with ratio "normal: tuberculosis" 95:5. Both samples will be analysed by AI-based system. Results will be quantified using diagnostic accuracy metrics: sensitivity and specificity, positive and negative predictor values, likelihood ratio, and area under the ROC (receiver operating characteristic) curve.

• Study Type: Observational
• Enrollment (Actual): 308

Contacts and Locations

Study Locations

• Russian Federation
  • Moscow, Russian Federation, 109029
    • Research and Practical Center of Medical Radiology, Department of Health Care of Moscow

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: N/A

• Sampling Method: Non-Probability Sample

Study Population

308 chest x-ray images selected from the database of Unified Radiological Information Service (URIS) of Moscow

Description

Inclusion Criteria:

• no pathology in a lung on chest x-ray
• signs of lung tuberculosis on chest x-ray

Exclusion Criteria:

• any pathology in the lungs (except tuberculosis)

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Sample N1; (n=140) with ratio "normal: tuberculosis" 50:50	Diagnostic test: AI-based x-ray analysis and triage ("normal/tuberculosis suspected"); All included x-rays will be analysed by the AI-based system. Then results will be compared with opinions of 2 experienced radiologists (they make peer-review of all included images independently of each other).; Other Names: artificial intelligence analysis of medical images
Sample N2; (n=150) with ratio "normal: tuberculosis" 95:5	Diagnostic test: AI-based x-ray analysis and triage ("normal/tuberculosis suspected"); All included x-rays will be analysed by the AI-based system. Then results will be compared with opinions of 2 experienced radiologists (they make peer-review of all included images independently of each other).; Other Names: artificial intelligence analysis of medical images

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic accuracy metric 1	Sensitivity	Day 1 upon receipt of data
Diagnostic accuracy metric 2	Specificity	Day 2 upon receipt of data
Diagnostic accuracy metric 3	Positive predictor values	Day 3 upon receipt of data
Diagnostic accuracy metric 4	Negative predictor values	Day 4 upon receipt of data
Diagnostic accuracy metric 5	Likelihood ratio	Day 5 upon receipt of data
Diagnostic accuracy metric 6	Area under the ROC curve	Day 6 upon receipt of data

Collaborators and Investigators

• Sponsor: Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department
• Investigators: Principal Investigator: Anton Vladzymyrskyy, Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department

Study record dates

Study Major Dates

• Study Start (Actual): February 1, 2018
• Primary Completion (Actual): March 15, 2018
• Study Completion (Estimated): December 30, 2023

Study Registration Dates

• First Submitted: March 28, 2018
• First Submitted That Met QC Criteria: May 25, 2023
• First Posted (Actual): June 5, 2023

Study Record Updates

• Last Update Posted (Actual): June 5, 2023
• Last Update Submitted That Met QC Criteria: May 25, 2023
• Last Verified: May 1, 2023

More Information

Terms related to this study

• Keywords: artificial intelligence; screening; tuberculosis; x-ray; AI; e-health; chest; deep learning; radiology
• Additional Relevant MeSH Terms: Infections; Respiratory Tract Infections; Respiratory Tract Diseases; Lung Diseases; Bacterial Infections; Bacterial Infections and Mycoses; Gram-Positive Bacterial Infections; Actinomycetales Infections; Mycobacterium Infections; Tuberculosis; Tuberculosis, Pulmonary
• Other Study ID Numbers: 2018-1

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