EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Analysis on artificial intelligence-based chest computed tomography in multidisciplinary treatment models for discriminating benign and malignant pulmonary nodules

Xian-Yan Liu, Fa-Cheng Shan, Hui Li, Jian-Bo Zhu

Clinics · 2025

DOI: 10.1016/j.clinsp.2025.100734 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Clinics, 2025. Verfasst von Xian-Yan Liu und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Respiratory Medicine, Binzhou People's Hospital, Binzhou, Shandong, China. Hauptthemen laut MEDLINE: Artificial Intelligence, Tomography, X-Ray Computed, Lung Neoplasms, Multiple Pulmonary Nodules. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Wu Jieping Medical Foundation, binzhou yixueyuan.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

To evaluate the effectiveness of AI-based chest Computed Tomography (CT) in a Multidisciplinary Diagnosis and Treatment (MDT) model for differentiating benign and malignant pulmonary nodules.

Methods

This retrospective study screened a total of 87 patients with pulmonary nodules who were treated between January 2019 and December 2020 at Binzhou People's Hospital, Qingdao Municipal Hospital, and Laiwu People's Hospital. AI analysis, MDT consultation, and a combined diagnostic approach were assessed using postoperative pathology as the reference standard.

Results

Among 87 nodules, 69 (79.31 %) were malignant, and 18 (20.69 %) were benign. AI analysis showed moderate agreement with pathology (κ = 0.637, p < 0.05), while MDT and the combined approach demonstrated higher consistency (κ = 0.847, 0.888, p < 0.05). Sensitivity and specificity were as follows: AI (89.86 %, 77.78 %, AUC = 0.838), MDT (100 %, 77.78 %, AUC = 0.889), and the combined approach (100 %, 83.33 %, AUC = 0.917). The accuracy of the combined method (96.55 %) was superior to MDT (95.40 %) and AI alone (87.36 %) (p < 0.05).

Conclusion

AI-based chest CT combined with MDT may improve diagnostic accuracy and shows potential for broader clinical application.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceTomography, X-Ray ComputedLung NeoplasmsMultiple Pulmonary NodulesSolitary Pulmonary NoduleHumansRetrospective StudiesMaleFemaleMiddle AgedAgedSensitivity and SpecificityDiagnosis, DifferentialAdultReproducibility of Results

Stichwörter der Autor:innen

Artificial intelligenceChest scanCombined diagnosticMDTPulmonary nodules

Autor:innen und Institutionen

  • Department of Respiratory Medicine, Binzhou People's Hospital, Binzhou, Shandong, China.
  • Department of Radiotherapy, Binzhou People's Hospital, Binzhou, Shandong, China.
  • Department of Respiratory Medicine, Binzhou People's Hospital, Binzhou, Shandong, China. Electronic address: zhujianbo@bzmc.edu.cn.

Förderung

  • Wu Jieping Medical Foundation
  • binzhou yixueyuan

Zitieren

APA

Liu, X. Y., Shan, F. C., Li, H., & Zhu, J. B. (2025). Analysis on artificial intelligence-based chest computed tomography in multidisciplinary treatment models for discriminating benign and malignant pulmonary nodules. Clinics. https://doi.org/10.1016/j.clinsp.2025.100734

Vancouver

Liu XY, Shan FC, Li H, Zhu JB. Analysis on artificial intelligence-based chest computed tomography in multidisciplinary treatment models for discriminating benign and malignant pulmonary nodules. Clinics. 2025. doi:10.1016/j.clinsp.2025.100734

BibTeX
@article{Liu2025,
  author = {Liu, Xian-Yan and Shan, Fa-Cheng and Li, Hui and Zhu, Jian-Bo},
  title = {Analysis on artificial intelligence-based chest computed tomography in multidisciplinary treatment models for discriminating benign and malignant pulmonary nodules},
  journal = {Clinics},
  year = {2025},
  doi = {10.1016/j.clinsp.2025.100734}
}
Kostenlos starten →