Predicting diffusion-FLAIR mismatch from B1000 and ADC without FLAIR: A deep learning-based approach
Pum Jun Kim, Dongyoung Kim, Joonwon Lee, Hyung Chan Kim, Jung Hwa Seo, Sukyoon Lee, Doo Hyuk Kwon, Hyungjong Park, Chang Hun Kim, Ho-Joon Lee, Yeonah Kang, Jaejun Yoo, Seongho Park · 2026
Read Predicting diffusion-FLAIR mismatch from B1000 and ADC without FLAIR: A deep learning-based approach by Pum Jun Kim, Dongyoung Kim, Joonwon Lee, Hyung Chan Kim, Jung Hwa Seo, Sukyoon Lee, Doo Hyuk Kwon, Hyungjong Park, Chang Hun Kim, Ho-Joon Lee, Yeonah Kang, Jaejun Yoo, Seongho 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/pmc-13462109.txt.
Description
Pum Jun Kim and colleagues build a deep learning model that predicts diffusion-FLAIR mismatch without FLAIR imaging in a 2026 study. The mismatch marks stroke patients with unclear onset who may benefit from recanalisation, yet FLAIR scans are often unavailable in acute settings. A model trained on 2,369 multicentre registry cases used only B1000 diffusion and apparent diffusion coefficient images, validated on 679 cases from two independent stroke centres.
On clearly defined cases, the model reached an area under the curve of 0.92 externally, beating human experts at 0.82, with the gap statistically significant. Performance dipped when ambiguous cases were included, for both model and readers. Stroke teams gain a complementary diagnostic tool for acute settings where FLAIR imaging cannot be obtained.
Catalog record
| CREATOR | Pum Jun Kim, Dongyoung Kim, Joonwon Lee, Hyung Chan Kim, Jung Hwa Seo, Sukyoon Lee, Doo Hyuk Kwon, Hyungjong Park, Chang Hun Kim, Ho-Joon Lee, Yeonah Kang, Jaejun Yoo, Seongho Park |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |