EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A CT-based multi-scale fusion model with SHAP interpretation for preoperative differentiation between lung adenocarcinoma in situ/minimally invasive adenocarcinoma and invasive adenocarcinoma: a multicenter study

Ning Dong, Shihang Sun, Zhongwei Li, Zhenjie Cong, Yi Lin, Jingxian Chen, Hu Zhang, Yuxin Liu, Shuxia Li, Jing Xu, Demin Kong, Qingfeng Yin

Frontiers in Oncology · 12.05.2026 · 16

DOI: 10.3389/fonc.2026.1824735 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Oncology, online erschienen am 12.05.2026. Verfasst von Ning Dong und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Yantaishan Hospital, Yantai, China. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

To develop a multi-scale model integrating radiomics, habitat, deep learning, and clinical information for preoperative precise differentiation between lung adenocarcinoma in situ/minimally invasive adenocarcinoma and invasive adenocarcinoma in ground-glass nodules (GGNs).

Methods

A total of 621 patients with GGNs from two centers were retrospectively enrolled. Radiomics, habitat, deep learning (2D, 2.5D, and 3D), and clinic models were constructed separately. Early fusion (feature-level fusion) and late fusion (decision-level fusion) strategies were systematically compared. The SHapley Additive exPlanations method was applied for interpretability analysis of the optimal model.

Results

The early fusion model demonstrated superior performance, achieving areas under the receiver operating characteristic curves of 0.988, 0.903, and 0.872 in the training, internal validation, and external testing cohorts, respectively. Among the deep learning architectures, the 2.5D ResNet50 achieved the best performance.

Conclusions

This multi-scale early fusion model provides robust support for preoperative precise diagnosis and individualized treatment decision-making for GGNs.

Schlagwörter

Stichwörter der Autor:innen

Deep learningSHAPground-glass noduleshabitatlung adenocarcinoma

Autor:innen und Institutionen

  • Department of Radiology, Yantaishan Hospital, Yantai, China.
  • Department of Medical Imaging, Weifang Yidu Central Hospital, Qingzhou, China.
  • Department of Radiology, Affiliated Hospital of Binzhou Medical University, Binzhou, China.
  • Department of Radiology, Zibo Central Hospital, Zibo, China.

Zitieren

APA

Dong, N., Sun, S., Li, Z., Cong, Z., Lin, Y., Chen, J., Zhang, H., Liu, Y., Li, S., Xu, J., Kong, D., & Yin, Q. (2026). A CT-based multi-scale fusion model with SHAP interpretation for preoperative differentiation between lung adenocarcinoma in situ/minimally invasive adenocarcinoma and invasive adenocarcinoma: a multicenter study. Frontiers in Oncology, 16. https://doi.org/10.3389/fonc.2026.1824735

Vancouver

Dong N, Sun S, Li Z, Cong Z, Lin Y, Chen J, et al. A CT-based multi-scale fusion model with SHAP interpretation for preoperative differentiation between lung adenocarcinoma in situ/minimally invasive adenocarcinoma and invasive adenocarcinoma: a multicenter study. Frontiers in Oncology. 2026;16. doi:10.3389/fonc.2026.1824735

BibTeX
@article{Dong2026,
  author = {Dong, Ning and Sun, Shihang and Li, Zhongwei and Cong, Zhenjie and Lin, Yi and Chen, Jingxian and Zhang, Hu and Liu, Yuxin and Li, Shuxia and Xu, Jing and Kong, Demin and Yin, Qingfeng},
  title = {A CT-based multi-scale fusion model with SHAP interpretation for preoperative differentiation between lung adenocarcinoma in situ/minimally invasive adenocarcinoma and invasive adenocarcinoma: a multicenter study},
  journal = {Frontiers in Oncology},
  year = {2026},
  volume = {16},
  doi = {10.3389/fonc.2026.1824735}
}
Kostenlos starten →