Feasibility study of unsupervised anomaly detection using Wasserstein GAN in SPECT image
Ryosuke Kasai, Hideki Otsuka · 2025
Read Feasibility study of unsupervised anomaly detection using Wasserstein GAN in SPECT image by Ryosuke Kasai, Hideki Otsuka (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-12864569.txt.
Description
Ryosuke Kasai and Hideki Otsuka test a computer system that flags odd spots in brain scans, in a 2025 paper. The scans, called SPECT images, help doctors spot brain disease, but reading them takes time and skill. Their system learns only from healthy scans, so anything that strays from that pattern stands out. It builds on a generative model grounded in optimal transport theory.
A fake test image with known noise and signal levels checked how the system behaves. It sorted odd from normal with near-perfect scores on that test. Real brain scans from clinic use came next. The oddness score matched the odd regions, and a side-by-side image made those regions plain to see. The authors say the system could ease the load on human readers once it is proven further.
Catalog record
| CREATOR | Ryosuke Kasai, Hideki Otsuka |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |