EUVIMEDEUVIMED entdecken
Crossref

Non-invasive classification of non-neoplastic and neoplastic gallbladder polyps based on clinical imaging and ultrasound radiomics features: An interpretable machine learning model

Minghui Dou, Hengchao Liu, Zhenqi Tang, Longxi Quan, Mai Xu, Feiqian Wang, Zhilin Du, Zhimin Geng, Qi Li, Dong Zhang

European Journal of Surgical Oncology · 2025-06 · 51 (6) · 109709

DOI: 10.1016/j.ejso.2025.109709 ↗

Kurz gefasst

Fachartikel in der Zeitschrift European Journal of Surgical Oncology, 2025-06. Verfasst von Minghui Dou und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Hepatobiliary Surgery, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, 710061, Shaanxi, China. Hauptthemen laut MEDLINE: Machine Learning, Gallbladder Neoplasms, Polyps, Adenoma. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 42 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: Gallbladder (GB) adenomas, precancerous lesions for gallbladder carcinoma (GBC), lack reliable non-invasive tools for preoperative differentiation of neoplastic polyps from cholesterol polyps. This study aimed to evaluate an interpretable machine learning (ML) combined model for the precise differentiation of the pathological nature of gallbladder polyps (GPs). METHODS: This study consecutively enrolled 744 patients from Xi'an Jiaotong University First Affiliated Hospital between January 2017 and December 2023 who were pathologically diagnosed postoperatively with cholesterol polyps, adenomas or T1-stage GBC. Radiomics features were extracted and selected, while clinical variables were subjected to univariate and multivariate logistic regression analyses to identify significant predictors of neoplastic polyps. A optimal ML-based radiomics model was developed, and separate clinical, US and combined models were constructed. Finally, SHapley Additive exPlanations (SHAP) was employed to visualize the classification process. RESULTS: The areas under the curves (AUCs) of the CatBoost-based radiomics model were 0.852 (95 % CI: 0.818-0.884) and 0.824 (95 % CI: 0.758-0.881) for the training and test sets, respectively. The combined model demonstrated the best performance with an improved AUC of 0.910 (95 % CI: 0.885-0.934) and 0.869 (95 % CI: 0.812-0.919), outperformed the clinical, radiomics, and US model (all P < 0.05), and reduced the rate of unnecessary cholecystectomies. SHAP analysis revealed that the polyp short diameter is a crucial independent risk factor in predicting the nature of the GPs. CONCLUSION: The ML-based combined model may be an effective non-invasive tool for improving the precision treatment of GPs, utilizing SHAP to visualize the classification process can enhance its clinical application.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningGallbladder NeoplasmsPolypsAdenomaGallbladder DiseasesHumansFemaleMaleMiddle AgedAgedUltrasonographyAdultCholecystectomyDiagnosis, DifferentialRetrospective StudiesRadiomics

Stichwörter der Autor:innen

Classification modelGallbladder carcinomaGallbladder polypsMachine learningRadiomicsSHAP

Autor:innen und Institutionen

  • Department of Hepatobiliary Surgery, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, 710061, Shaanxi, China.
  • Department of Ultrasound, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, 710061, Shaanxi, China.
  • Department of Hepatobiliary Surgery, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, 710061, Shaanxi, China. Electronic address: gengzhimin@mail.xjtu.edu.cn.
  • Department of Hepatobiliary Surgery, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, 710061, Shaanxi, China. Electronic address: docliqi@163.com.
  • Department of Hepatobiliary Surgery, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, 710061, Shaanxi, China. Electronic address: zhangdong811021@126.com.

Zitieren

APA

Dou, M., Liu, H., Tang, Z., Quan, L., Xu, M., Wang, F., Du, Z., Geng, Z., Li, Q., & Zhang, D. (2025). Non-invasive classification of non-neoplastic and neoplastic gallbladder polyps based on clinical imaging and ultrasound radiomics features: An interpretable machine learning model. European Journal of Surgical Oncology, 51(6), 109709. https://doi.org/10.1016/j.ejso.2025.109709

Vancouver

Dou M, Liu H, Tang Z, Quan L, Xu M, Wang F, et al. Non-invasive classification of non-neoplastic and neoplastic gallbladder polyps based on clinical imaging and ultrasound radiomics features: An interpretable machine learning model. European Journal of Surgical Oncology. 2025;51(6):109709. doi:10.1016/j.ejso.2025.109709

BibTeX
@article{Dou2025,
  author = {Dou, Minghui and Liu, Hengchao and Tang, Zhenqi and Quan, Longxi and Xu, Mai and Wang, Feiqian and Du, Zhilin and Geng, Zhimin and Li, Qi and Zhang, Dong},
  title = {Non-invasive classification of non-neoplastic and neoplastic gallbladder polyps based on clinical imaging and ultrasound radiomics features: An interpretable machine learning model},
  journal = {European Journal of Surgical Oncology},
  year = {2025},
  volume = {51},
  number = {6},
  pages = {109709},
  doi = {10.1016/j.ejso.2025.109709}
}
Kostenlos starten →