Testing the LIRA AI tool for screening TB from chest X-rays in general hospital settings

Not yet recruiting

Conditions: Tuberculosis of lung,

Registration Number: CTRI/2025/06/088028

Lead Sponsor: Indian Council of Medical Research

Brief Summary: The purpose of this study is to evaluate the diagnostic accuracy and validate the performance ofLIRA AI Tool, in detecting Pulmonary TB and other key clinical findings or abnormalities from adult chestX-rays and to compare its results with those of expert radiologists.

The objectives of this clinical study are divided into two major group studies:

**Retrospective study:**The primary objective is to assess LIRA’s sensitivity, specificity, positive predictive value, andnegative predictive value in detecting tuberculosis (TB) versus the reference standardmicrobiological confirmation using previously collected chest X-ray data.The secondary objective is to compare LIRA’s sensitivity, specificity, positive predictive value,and negative predictive value in detecting tuberculosis (TB) versus a radiologist’s opinion usingpreviously collected chest X-ray data.

**Prospective study**:The primary objective is to evaluate LIRA’s accuracy and reliability in real-time TB detection,aiming to enhance diagnostic speed and efficiency.Page 9 of 74CMT\_LIRA\_001Version 01The secondary objective focuses on LIRA’s impact on radiologist workflows, specifically itsintegration with PACS, its effect on turnaround times, and its capacity to prioritize high-risk casesin clinical and community settings.

This is a case-control design and prospective cohort study designed to evaluate the clinical efficacyand diagnostic accuracy of LIRA, an AI-powered radiology software, in identifying abnormalitiesassociated with tuberculosis (TB) on CXR in AP/PA views.

Detailed Description: Not available

Recruitment & Eligibility

Status: Not Yet Recruiting

Sex: All

Target Recruitment: 5640

Inclusion Criteria

Retrospective Phase: a.
Archived chest X-ray images of adult patients, aged 18 years and older , collected within the last 6 months.
Minimum image resolution of 1440 x 1440 pixels.
Availability of corresponding microbiological test results.
No prior diagnosis or treatment for active TB within the past 6 months.
Prospective Phase: a.
Adults aged 18 years and above presenting for chest X-rays due to suspected TB.

Exclusion Criteria

Retrospective phase: a.
Incomplete or poor-quality CXR images.
Patients under 18 years of age.
Absence of corresponding microbiological test results or clinical data.
Prior diagnosis or treatment for active TB within the past 6 months.
Prospective Phase: a.
Pregnant or lactating individuals (due to radiation exposure).
Patients with severe respiratory distress require immediate clinical intervention.
Inability to comply with the study protocol due to cognitive impairments or severe comorbidities.

Study & Design

Study Type: Observational

Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
This study will evaluate the diagnostic accuracy of LIRA in TB screening and detection of other chest abnormalities by comparing its sensitivity and specificity to the performance with that of expert radiologists	6 months

Secondary Outcome Measures

Name	Time	Method
This study will examine the impact of LIRA on clinical workflow efficiency, diagnostic speed, and the qualitative feedback from radiologists regarding its usability.	6 months

Trial Locations

Locations (1): St. Johns Research Institute
🇮🇳
Bangalore, KARNATAKA, India
St. Johns Research Institute
🇮🇳Bangalore, KARNATAKA, India
Dr Prashanth T
Principal investigator
08049467054
prashanth.t@sjri.res.in