Diagnostic performance of AI-powered prostate MRI against biopsy ground truth: quasi-continuous risk scoring versus fixed PI-RADS thresholds
Nadine Bayerl, Joel Appel, Maximilian Schmidt, Alexander Cavallaro, Robert Grimm, Heinrich von Busch, Bernd Wullich, Arndt Hartmann, Michael Schlicht, Michael Uder, Matthias S. May, Stephan Ellmann · 2026
Read Diagnostic performance of AI-powered prostate MRI against biopsy ground truth: quasi-continuous risk scoring versus fixed PI-RADS thresholds by Nadine Bayerl, Joel Appel, Maximilian Schmidt, Alexander Cavallaro, Robert Grimm, Heinrich von Busch, Bernd Wullich, Arndt Hartmann, Michael Schlicht, Michael Uder, Matthias S. May, Stephan Ellmann (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-13580003.txt.
Description
Nadine Bayerl and co-authors test a shop-bought AI tool that reads prostate scans for signs of cancer. Their 2026 study looks back at just over one hundred scans, each followed by a tissue sample that gave the true answer. The tool gives two kinds of output. One is a fixed grade scale used by doctors, and one is a smooth score of how suspect a scan looks.
For the more serious cancers, the smooth score did a bit better than the fixed scale. At its best cut point, the score kept the same catch rate but cut false alarms. Small checks for bias in the cut point left the gains much the same.
Catalog record
| CREATOR | Nadine Bayerl, Joel Appel, Maximilian Schmidt, Alexander Cavallaro, Robert Grimm, Heinrich von Busch, Bernd Wullich, Arndt Hartmann, Michael Schlicht, Michael Uder, Matthias S. May, Stephan Ellmann |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |