Development and validation of a multimodal MRI habitat-based deep learning fusion model for predicting 252Cf neutron therapy response in cervical cancer
Shu-Peng Wang, Zhi-Guo Chen, Xi-Jing Shan, Miao Jin, Wei Zhao, Xu Tong, Xin Meng · 2026
Read Development and validation of a multimodal MRI habitat-based deep learning fusion model for predicting 252Cf neutron therapy response in cervical cancer by Shu-Peng Wang, Zhi-Guo Chen, Xi-Jing Shan, Miao Jin, Wei Zhao, Xu Tong, Xin Meng (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-13621295.txt.
Description
Californium-252 neutron brachytherapy works well for locally advanced cervical cancer, but patients respond unevenly, and a non-invasive forecast would help plan care. The seven authors built a prediction model from pre-treatment MRI in 100 patients treated with the neutron therapy plus external beam radiation. Clustering on contrast and diffusion maps split each tumor into three habitat subregions by vascularity and cellularity.
Radiomics features from the whole tumor and each habitat, deep learning features from a pretrained ResNet50, and clinical-dose data were fused, selected, and fed to a support vector machine. In the 30-patient validation set the fusion model reached an AUC of 0.892 and accuracy of 0.867, beating clinical, radiomics-only, and combined rivals. Decision curve analysis gave it the highest net benefit.
Catalog record
| CREATOR | Shu-Peng Wang, Zhi-Guo Chen, Xi-Jing Shan, Miao Jin, Wei Zhao, Xu Tong, Xin Meng |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |