A deep-learning noise reduction algorithm outperforms the spatial filters previously required for bone SPECT on a high-speed whole-body 360° CZT-camera
Achraf Bahloul, Franklin Rajadhas, Matthieu Doyen, Yechiel Lamash, Nathaniel Roth, Véronique Roch, Pierre-Yves Marie, Laetitia Imbert · 2025
Read A deep-learning noise reduction algorithm outperforms the spatial filters previously required for bone SPECT on a high-speed whole-body 360° CZT-camera by Achraf Bahloul, Franklin Rajadhas, Matthieu Doyen, Yechiel Lamash, Nathaniel Roth, Véronique Roch, Pierre-Yves Marie, Laetitia Imbert (2025) 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-12864558.txt.
Description
Achraf Bahloul and seven co-authors test, in a 2025 study, a deep-learning cleaner for bone scans from a fast whole-body camera. The camera rings the patient with detectors and images in minutes, but the raw images are noisy. The usual fix, spatial filters, smooths the image and dulls the very contrast doctors need to judge a lesion.
The team compared the learned cleaner, used alone, against the standard filters and against the two stacked together. Across 114 real bone lesions and test spheres, the learned method alone gave higher peak values and sharper recovery, with lower noise. Its contrast-to-noise ratio nearly doubled for lesions near one cubic centimetre. The authors say the cleaner can replace the filters on this camera.
Catalog record
| CREATOR | Achraf Bahloul, Franklin Rajadhas, Matthieu Doyen, Yechiel Lamash, Nathaniel Roth, Véronique Roch, Pierre-Yves Marie, Laetitia Imbert |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |