Diagnostic performance of machine learning models based on dual-phase 99mTc-MIBI SPECT/CT semiquantitative parameters for differentiating benign and malignant pulmonary nodules
Kun Zhang, Xin Zhou, Yuhang Zhang, Gang Jin, Ping Li, Yuzhuo Xing · 2026
Read Diagnostic performance of machine learning models based on dual-phase 99mTc-MIBI SPECT/CT semiquantitative parameters for differentiating benign and malignant pulmonary nodules by Kun Zhang, Xin Zhou, Yuhang Zhang, Gang Jin, Ping Li, Yuzhuo Xing (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-0353271.txt.
Description
Kun Zhang and five colleagues present this 2026 PLOS One imaging study on telling harmless lung nodules from cancerous ones without surgery. Their method uses a tracer scan taken at two time points, producing simple uptake ratios for each nodule. Records from more than one hundred thirty patients, most with malignant nodules, were used to build and test machine learning classifiers on those ratios plus clinical variables.
Cancerous nodules took up more tracer early and held onto less of it over time. A blood tumor marker and one retention measure independently predicted malignancy. The best machine learning model separated the groups almost perfectly on its training data, with others close behind. The authors pitch the approach as an interpretable option for hospitals where advanced image analysis is unavailable, though outside validation is still needed.
Catalog record
| CREATOR | Kun Zhang, Xin Zhou, Yuhang Zhang, Gang Jin, Ping Li, Yuzhuo Xing |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25153 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |