Lunit INSIGHT CXR is an AI solution that detects 10 different radiologic findings on chest X-rays including tuberculosis (TB). It is designed to provide a quick analysis in hard-to-reach areas regardless of internet connections, with proven versatility among different X-ray devices.

The AI solution generates (1) location information for detected lesions in color or outline, (2) abnormality scores reflecting the probability that the detected lesion is abnormal, and (3) an AI “case report” that summarizes the analysis result by each finding.

Target Setting

Primary health centers, general hospital (above primary level), teleradiology companies, government/public sector, e.g. national TB program, private sector

Current Market

Asia-Pacific, Europe, Middle East/North Africa, North America, South/Central America

Output

Structured report including: 

» Heat map, 

» Probability score as well as dichotomous output indicating whether each abnormality is present or absent, 

» Probability score as well as dichotomous output indicating whether TB is likely present or likely absent, 

» Specification of the location of each abnormality. 

Default threshold probability score is set at 15% for all abnormalities including TB. 

Chest abnormalities detected by the product for which a separate abnormality score is given include: atelectasis, calcification, cardiomegaly, consolidation, fibrosis, mediastinal widening, nodule, pleural effusion, pneumoperitoneum, pneumothorax and TB.

Data Sharing & Privacy

»» Server Location: Amazon Web Services is used. Local or national servers can be set up if required

»» Data is not shared with the developer

»» There is an option to de-identify data. The Lunit DICOM Gateway anonymizes all personal information before transferring original images to the analysis server

Price

»» LUNIT INSIGHT offers volume-based pricing for the use of the license. When using the online solution, the volume-based price includes maintenance

»» When using the offline solution, maintenance is optional. Fees for infrastructure, customization and consulting are subject to change depending on the operating environment

Publications

1. Nam JG, Park S, Hwang EJ, Lee JH, Jin K-N, Lim KY, et al. Development and validation of deep learning–based automatic detection algorithm for malignant pulmonary nodules on chest radiographs. Radiology. 2019;290(1):218–28. doi: 10.1148/radiol.2018180237 Available from: http://pubs.rsna.org/doi/10.1148/radiol.2018180237 

2. Eui Jin Hwang, Sunggyun Park, et al. Development and validation of a deep learning–based automatic detection algorithm for active pulmonary tuberculosis on chest radiographs. Clin Infect Dis. 2019;69(5):739-747. doi: 10.1093/cid/ciy967. Available from: https://academic.oup.com/cid/article/69/5/739/5174137 

3. Hwang EJ, Park S, Jin KN, Kim JI, Choi SY, Lee JH, et al. Development and validation of a deep learning-based automated detection algorithm for major thoracic diseases on chest radiographs. JAMA Netw Open. 2019;2(3):e191095. doi: 10.1001/jamanetworkopen.2019.1095. Available from: https://doi.org/10.1001/jamanetworkopen.2019.1095 

4. Qin ZZ, Sander MS, Rai B, Titahong CN, Sudrungrot S, Laah SN, 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. doi: 10.1038/s41598-019-51503-3. Available from: https://doi.org/10.1038/s41598-019-51503-3 

5. Hwang EJ, Nam JG, Lim WH, Park SJ, Jeong YS, Kang JH, et al. Deep learning for chest radiograph diagnosis in the emergency department. Radiology. 2019;293(3):573–80. doi: 10.1148/radiol.2019191225. Available from: http://pubs.rsna.org/doi/10.1148/radiol.2019191225 

6. Kim H, Park CM, Goo JM. Test-retest reproducibility of a deep learning–based automatic detection algorithm for the chest radiograph. Eur Radiol. 2020;30(4):2346–55. doi: 10.1007/s00330-019-06589-8. Available from: https://doi.org/10.1007/s00330-019-06589-8