AI in Lung Nodules on Chest CT: Retrospective, Real-World Assessment of AI in Clinical Practice
Heyel Sodhi-Kalra, Shivangi Jha, Subba Digumarthy, Matthew Balcarras · 2026
Read AI in Lung Nodules on Chest CT: Retrospective, Real-World Assessment of AI in Clinical Practice by Heyel Sodhi-Kalra, Shivangi Jha, Subba Digumarthy, Matthew Balcarras (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13218718.txt.
Description
Heyel Sodhi-Kalra and colleagues audit a cleared AI tool for lung nodules against radiologists' reports in a 2026 real-world study. More than a thousand AI products hold clearance, yet few work in clinics, partly from doubts about performance. Errors in counting or classifying nodules can push patients toward needless follow-up or, worse, miss a cancer.
The team took 650 noncontrast chest CT scans from adults, all processed by an FDA-cleared tool that detects, locates, and measures lung nodules. They set its output beside the reports of eighteen subspecialty radiologists. Their finding is frequent discordance between software and radiologists on nodule assessment. Because the tool was deployed, the study measures performance where it matters, in the flow of routine care rather than in a test set. The authors argue such real-world checks should accompany adoption of AI, since clearance testing alone does not guarantee agreement with radiologists.
Catalog record
| CREATOR | Heyel Sodhi-Kalra, Shivangi Jha, Subba Digumarthy, Matthew Balcarras |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 10 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |