FindTrial
Sign In

Use of artificial intelligence to interpret chest X-rays

Can Artificial Intelligence reliably report chest X-Rays? Radiologist validation of an algorithm trained on 2.3 million chest X-rays.

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
CTRI
Registry ID
CTRI/2020/08/027488
Enrollment
2000
Registered
2020-08-31
Start date
Unknown
Completion date
Unknown
Last updated
2021-11-24

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Health Condition 1: J989- Respiratory disorder, unspecified

Interventions

Control Intervention1: Nil: Nil

Sponsors

Qureai
Lead Sponsor

Eligibility

Inclusion criteria

Inclusion criteria: Anonymized Chest X-rays with View: PA/AP Patient Position: Erect File format: Valid DICOM

Exclusion criteria

Exclusion criteria: Non Chest X-ray scans Patient Position: Supine File format: Invalid DICOM

Design outcomes

Primary

MeasureTime frame
Retrospective assessment of 1. A three- radiologist majority on an independent, retrospectively collected set of 2000 anonymized chest X-rays(CQ2000) 2. The radiologist report on a separate validation set of anonymized 100,000 scans(CQ100k). The primary accuracy measure was area under the ROC curve (AUC), estimated separately for each abnormality as well as for normal versus abnormal scans.Timepoint: 1 year - retrospective study—

Secondary

MeasureTime frame
Validation of algorithms for detecting abnormal chest X-rays from normal onesTimepoint: 1 year - retrospective study—

Countries

India

Contacts

Public ContactAmmar Jagirdar

Qure.ai

ammar.jagirdar@qure.ai

Outcome results

None listed

Source: CTRI (via WHO ICTRP) · Data processed: Feb 4, 2026