DETECTX: a novel deep learning approach for detecting pneumonia, tuberculosis, and aspergillosis from chest X-rays
Xian-Hong Wang, Naeem Ahmed, Muhammad Saeed, Hassan Ali, Guang-Wei Chen, Shi-Chang Sun, Wei-Xia Dong, Tie-Shan Teng, Zhong-Xia Liao, Umair Ali khan Saddozai · 2026
Read DETECTX: a novel deep learning approach for detecting pneumonia, tuberculosis, and aspergillosis from chest X-rays by Xian-Hong Wang, Naeem Ahmed, Muhammad Saeed, Hassan Ali, Guang-Wei Chen, Shi-Chang Sun, Wei-Xia Dong, Tie-Shan Teng, Zhong-Xia Liao, Umair Ali khan Saddozai (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-13626911.txt.
Description
Xian-Hong Wang and co-authors train a deep learning system to tell apart three lung diseases on chest X-rays. Pneumonia, tuberculosis, and aspergillosis share cough, fever, and breathlessness, so they are easy to confuse, and late calls on aspergillosis can cost lung tissue. Their 2026 study builds a system, called DETECTX, that sorts X-rays into those three diseases or healthy.
Three model types were trained on 7,200 images split into the four groups. A Siamese network, which learns by comparing image pairs, beat both a plain convolutional network and a standard ResNet-50 model. It reached 98.72 percent accuracy, ahead of published rivals in the team's comparison. The group also built a web page where clinicians can upload a scan and get the system's reading. They offer it as an aid to faster, surer diagnosis.
Catalog record
| CREATOR | Xian-Hong Wang, Naeem Ahmed, Muhammad Saeed, Hassan Ali, Guang-Wei Chen, Shi-Chang Sun, Wei-Xia Dong, Tie-Shan Teng, Zhong-Xia Liao, Umair Ali khan Saddozai |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |