9 August 2021
qXR is an automated chest X-ray interpretation tool that detects 30 different findings, including radiological signs of tuberculosis (TB). The artificial intelligence algorithm underlying qXR is trained to detect not only classical primary pulmonary TB, but also atypical manifestations. It can be used to simultaneously screen for radiological signs of lung malignancies, chronic obstructive pulmonary disease, and COVID-19.
qXR can automatically fetch a chest X-ray from the X-ray machine or picture archiving and communication system (PACS) and process it offline on a customized solution or online on a HIPAA compliant cloud service. It scans the chest X-ray for the presence of 30 different findings and generates a report indicating the presence of each of the findings. It also marks the scan with the location of the abnormalities. It produces a yes/no output as well as a probability score of the confidence of the deep learning model for the particular abnormality.
qXR comes bundled with a complete case finding web-based software called qTrack that helps track clients, monitor presumptives and log microbiological confirmatory results across different sites. qTrack can be accessed with any browser and includes a free radiology viewer that can be used to see the chest X-ray scans and qXR results. qTrack is a workflow software for disease management that embeds qXR as a part of the radiology workflow with a built in viewer.
On the market
Online & offline
Intended Age Group
Primary health centres, general hospital (above primary level), teleradiology companies, government/public sector, e.g. national TB programme, private sector
Global, qXR is being used in over 20 countries
»» Can be used to read images from any kind of chest X-ray machine (vendor neutral)
»» Chest X-ray image format: DICOM for digital chest X-rays, PNG and JPEG formats using qTrack for acquiring pictures of analog chest X-ray films
»» Chest X-ray type: posterior-anterior chest X-ray, anterior-posterior chest X-ray, portable
»» Other requirements: none
Outputs include: visual outputs – updated secondary capture, text outputs – structured report, tabulated pdf report.
Structured report including:
»»Probability score as well as dichotomous output indicating whether each abnormality is present or absent
»»Probability score for TB as well as dichotomous output indicating whether TB is likely present or likely absent
»»A box indicating the location of the abnormalities
Abnormalities detected by the product for which a separate abnormality score is given include: abnormal, TB, opacities (atelectasis, cavities, calcification, consolidation, fibrosis, nodules, reticulonodular pattern), emphysema/hyperinflation, pleural effusion, blunted costophrenic angle, pneumothorax, cardiomegaly, tracheal shift, degenerative spine changes, scoliosis, hilar prominence, rib fractures, COVID-19, pneumoperitoneum, mediastinal widening, elevated hemidiaphragm, abnormal diaphragm shape, lines and tubes (4 types) and lung nodule malignancy risk for nodules.
qXR can generalize to any X-ray manufacturer (CR/DR) and has been tested with over 20 leading X-ray manufacturers globally.
It is possible to integrate qXR with the client’s legacy PACS or directly to the X-ray device. Additionally, each deployment of qXR can be coupled with access to qTrack to facilitate viewing of results and managing the screening workflow.
Ubuntu 18.04 is preferred
Less than a minute
Data Sharing & Privacy
»» Server location (for online product): the server can be set up in any location (preferred location is Amazon Web Services, but the software can be deployed on any cloud provider). Set-up of a local or national server is possible
»» Online deployment can be done using a recommended cloud service in-country or with a qure.ai preferred partner. Offline deployment can be done using a small footprint portable computer
»» No data are shared with the developer
»» All data are anonymized before analysis by qXR
»» Routine product upgrades for improvements and periodic updates for improved accuracy (annual)
Product Development Method
Deep learning to analyse chest X-ray scans. To find out more about the product's algorithm development, see the blog at blog.qure.ai or contact firstname.lastname@example.org.
The current commercial version of qXR v3 has been trained and tested on 3.7 million chest X-rays acquired globally.
Human readers, culture, sputum smear microscopy, and GeneXpert results
»» Tavaziva, G., Harris, M., Abidi, S.K. et al. Chest X-ray analysis with deep learning-based software as a triage test for pulmonary tuberculosis: an individual patient data meta-analysis of diagnostic accuracy. Clin Infect Dis. 2021:ciab639. https://doi.org/10.1093/cid/ciab639
»» Ebrahimian, S., Homayounieh, F., Rockenbach, M.A.B.C. et al. Artificial intelligence matches subjective severity assessment of pneumonia for prediction of patient outcome and need for mechanical ventilation: a cohort study. Sci Rep. 2021;11(1):858. https://doi.org/10.1038/s41598-020-79470-0
»» Khan, F.A., Majidulla, A., Tavaziva, G. et al. Chest x-ray analysis with deep learning-based software as a triage test for pulmonary tuberculosis: a prospective study of diagnostic accuracy for culture-confirmed disease. Lancet Digit Health. 2020;2(11):e573-e581. https://doi.org/10.1016/S2589-7500(20)30221-1
»» Mushtaq, J., Pennella, R., Lavalle, S. et al. Initial chest radiographs and artificial intelligence (AI) predict clinical outcomes in COVID-19 patients: analysis of 697 Italian patients. Eur Radiol. 2021;31(3):1770-1779. https://doi.org/10.1007/s00330-020-07269-8
»» Qin, Z.Z., Ahmed, S., Sarker, M.S. et al. Tuberculosis detection from chest x-rays for triaging in a high tuberculosis-burden setting: an evaluation of five artificial intelligence algorithms. Lancet Digital Health. 2021;3:e543–5437: 2113–31. https://doi.org/10.1016/S2589-7500(21)00116-3
»» Engle, E., Gabrielian, A., Long, A., Hurt, D.E., Rosenthal, A. Performance of Qure.ai automatic classifiers against a large annotated database of patients with diverse forms of tuberculosis. PLoS One. 2020;15(1):e0224445. https://doi.org/10.1371/journal.pone.0224445
»» Nash, M., Kadavigere, R., Andrade, J. et al. Deep learning, computer-aided radiography reading for tuberculosis: a diagnostic accuracy study from a tertiary hospital in India. Sci Rep. 2020;10(1):210. https://doi.org/10.1038/s41598-019-56589-3
»» Qin, Z.Z., Sander, M.S., Rai, B. et al. Using artificial intelligence to read chest radiographs for tuberculosis detection: A multi-site evaluation of the diagnostic accuracy of three deep learning systems. Sci Rep. 2019;9(1):15000. https://doi.org/10.1038/s41598-019-51503-3
»» Putha, P., Tadepalli, M., Reddy, B. Can Artificial Intelligence Reliably Report Chest X-Rays?: Radiologist Validation of an Algorithm trained on 2.3 Million X-Rays. Available from: https://arxiv.org/abs/1807.07455
»» Singh, R., Kalra, M.K., Nitiwarangkul, C et al. Deep learning in chest radiography: Detection of findings and presence of change. PLoS One. 2018;13(10):e0204155. https://doi.org/10.1371/journal.pone.0204155
This website works best with browsers other than Internet Explorer.