Feasibility of deep learning-based image reconstruction using TrueFidelity for coronary calcium scoring
Ji Won Kim, Tae Hoon Kim, Seung Hyun Lee, Ji Eun Nam, Chul Hwan Park · 2026
Read Feasibility of deep learning-based image reconstruction using TrueFidelity for coronary calcium scoring by Ji Won Kim, Tae Hoon Kim, Seung Hyun Lee, Ji Eun Nam, Chul Hwan Park (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/plos-pone-0358342.txt.
Description
Ji Won Kim and four colleagues test whether an artificial intelligence reconstruction changes heart calcium scores, in PLOS ONE in 2026. The score, built from CT images, sorts patients into risk categories that drive statin decisions. A deep learning method promises cleaner images from the same scan. The team scored a cardiac phantom and one hundred thirteen patients three ways: the classic reconstruction, an iterative method, and the deep learning one at high strength.
In the phantom, all three agreed. In patients, the deep learning scores ran significantly lower than the classic method's, while matching the iterative method. Category agreement stayed high overall. Individual patients, though, could shift substantially between methods, with differences large enough to matter near the cutoffs where risk labels change. The authors' caution is practical: a score is only comparable when the reconstruction behind it is held constant.
Catalog record
| CREATOR | Ji Won Kim, Tae Hoon Kim, Seung Hyun Lee, Ji Eun Nam, Chul Hwan Park |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41576 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |