Pneumothorax detection and segmentation from chest X-ray radiographs using a patch-based fully convolutional encoder-decoder network
Jakov Ivan S. Dumbrique, Reynan B. Hernandez, Juan Miguel L. Cruz, Ryan M. Pagdanganan, Prospero C. Naval · 2024
Read Pneumothorax detection and segmentation from chest X-ray radiographs using a patch-based fully convolutional encoder-decoder network by Jakov Ivan S. Dumbrique, Reynan B. Hernandez, Juan Miguel L. Cruz, Ryan M. Pagdanganan, Prospero C. Naval (2024) 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-11668597.txt.
Description
Dumbrique, Hernandez, Cruz, Pagdanganan, and Naval build a model to find collapsed lung on chest X-rays in twenty twenty-four. The condition, pneumothorax, traps air near the lung and can be life threatening. X-rays are fast and cheap, but the only sign is often a thin pleural line. That makes the condition easy for human readers to miss.
Their model blends a fully convolutional network with ideas from vision transformers, without costly attention. It works on patches and uses skip links to mix fine and coarse features. On two data sets, it beat prior models on detection and outline accuracy while staying efficient. One data set was newly built from a hospital in the Philippines. A mixed loss function gave a further clear gain in tests.
Catalog record
| CREATOR | Jakov Ivan S. Dumbrique, Reynan B. Hernandez, Juan Miguel L. Cruz, Ryan M. Pagdanganan, Prospero C. Naval |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 70 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |