EUVIMEDEUVIMED entdecken
Crossref

Preoperative prediction of perineural invasion in intrahepatic cholangiocarcinoma with interpretable machine learning based on MRI

Xiaoqi Zhou, Meicheng Chen, Danyang Xu, Jing Hu, Ziwei Liu, Chenyu Song, Mimi Tang, Jifei Wang, Yuying Chen, Yanji Luo, Zhenpeng Peng, Shi-Ting Feng

European Journal of Surgical Oncology · 2026-03 · 52 (3) · 111450

DOI: 10.1016/j.ejso.2026.111450 ↗

Kurz gefasst

Fachartikel in der Zeitschrift European Journal of Surgical Oncology, 2026-03. Verfasst von Xiaoqi Zhou und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, 510080, Province Guangdong, PR China. Hauptthemen laut MEDLINE: Cholangiocarcinoma, Bile Duct Neoplasms, Magnetic Resonance Imaging, Machine Learning. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 39 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Guangdong Provincial Natural Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

INTRODUCTION: To develop radiomics and deep learning (DL) based interpretable models using MRI for preoperative prediction of perineural invasion (PNI) in intrahepatic cholangiocarcinoma (ICC). MATERIALS AND METHODS: A total of 165 pathologically confirmed ICC patients with preoperative MRI were retrospectively enrolled from two centers (center1, training set, n = 115; validation set, n = 14; internal test set, n = 15; center 2, external test set, n = 21). Radiomics and DL models were constructed for single-phase (pre-contrast, arterial phase, portal venous phase, hepatobiliary phase [HBP]) and multi-phase MRI using the Shukun AI platform and PNI-MambaNet. Model performance was evaluated with the area under the receiver operating characteristic curve (AUC). Gradient-weighted class activation mapping (Grad-CAM) heatmaps visualized the regions prioritized by the DL models. RESULTS: The PNI positive rate was 42.4 % (61/144) and 28.6 % (6/21) in the two centers. Radiomics HBP models achieved the highest AUC in the internal test set, while multi-phase model performed best in the external test set (AUC: HBP, 0.778 and 0.733 for the internal and external test sets, respectively; multi-phase, 0.759 and 0.778). For DL models, multi-phase model achieved the highest AUC in the internal test set, while HBP model performed best in the external test set (AUC: HBP, 0.926 and 0.856; multi-phase, 0.944 and 0.844). DL models outperformed radiomics models in the external test set, with Grad-CAM visualizing tumor margin regions as the interest area. CONCLUSIONS: DL models based on MRI effectively predict PNI in ICC, with visualizations enhancing clinical interpretability and potential application.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

CholangiocarcinomaBile Duct NeoplasmsMagnetic Resonance ImagingMachine LearningDeep LearningPeripheral NervesHumansFemaleNeoplasm InvasivenessRadiomicsRetrospective StudiesMaleMiddle AgedAgedPredictive Learning ModelsROC Curve

Stichwörter der Autor:innen

Deep learningIntrahepatic cholangiocarcinomaMulti-phase MRIPerineural invasionRadiomics

Autor:innen und Institutionen

  • Department of Radiology, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, 510080, Province Guangdong, PR China.
  • Shukun Technology Co., Ltd, Jinhui Building, Qiyang Road, Beijing, 100102, PR China.
  • Department of Radiology, Southern Medical University (The First People's Hospital of Shunde), Foshan, 528000, Province Guangdong, PR China.
  • Department of Radiology, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, 510080, Province Guangdong, PR China. Electronic address: luoyj26@mail.sysu.edu.cn.
  • Department of Radiology, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, 510080, Province Guangdong, PR China. Electronic address: pengzhp@mail.sysu.edu.cn.
  • Department of Radiology, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, 510080, Province Guangdong, PR China. Electronic address: fengsht@mail.sysu.edu.cn.

Förderung

  • National Natural Science Foundation of China
  • Guangdong Provincial Natural Science Foundation

Zitieren

APA

Zhou, X., Chen, M., Xu, D., Hu, J., Liu, Z., Song, C., Tang, M., Wang, J., Chen, Y., Luo, Y., Peng, Z., & Feng, S. T. (2026). Preoperative prediction of perineural invasion in intrahepatic cholangiocarcinoma with interpretable machine learning based on MRI. European Journal of Surgical Oncology, 52(3), 111450. https://doi.org/10.1016/j.ejso.2026.111450

Vancouver

Zhou X, Chen M, Xu D, Hu J, Liu Z, Song C, et al. Preoperative prediction of perineural invasion in intrahepatic cholangiocarcinoma with interpretable machine learning based on MRI. European Journal of Surgical Oncology. 2026;52(3):111450. doi:10.1016/j.ejso.2026.111450

BibTeX
@article{Zhou2026,
  author = {Zhou, Xiaoqi and Chen, Meicheng and Xu, Danyang and Hu, Jing and Liu, Ziwei and Song, Chenyu and Tang, Mimi and Wang, Jifei and Chen, Yuying and Luo, Yanji and Peng, Zhenpeng and Feng, Shi-Ting},
  title = {Preoperative prediction of perineural invasion in intrahepatic cholangiocarcinoma with interpretable machine learning based on MRI},
  journal = {European Journal of Surgical Oncology},
  year = {2026},
  volume = {52},
  number = {3},
  pages = {111450},
  doi = {10.1016/j.ejso.2026.111450}
}
Kostenlos starten →