EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Hybrid deep feature and machine learning framework for classification of thyroid nodules in ultrasound images

Dingnan Zhang, Bo Li, Hao Ju, Tingxue Li, Yanzhu Zhang

Frontiers in Oncology · 25.05.2026 · 16

DOI: 10.3389/fonc.2026.1784411 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Oncology, online erschienen am 25.05.2026. Verfasst von Dingnan Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Automation and Electrical Engineering, Shenyang Ligong University, Shenyang, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Accurate differentiation between benign and malignant thyroid nodules is essential for reducing unnecessary biopsies and improving early clinical decision-making. This study aims to enhance the reliability of ultrasound-based thyroid nodule assessment by proposing an optimized computer-aided diagnosis (CAD) framework.

Methods

A hybrid CAD framework combining deep transfer learning with gradient-boosted decision tree classification was developed. High-level semantic features were extracted from a pretrained ResNet50 model and encoded as fixed-length representations. These features were then fed into the CatBoost algorithm for downstream classification, enabling the framework to leverage both the representational richness of deep neural networks and the decision efficiency of boosted tree ensembles, especially under the condition of limited annotated medical data. A comprehensive set of complementary evaluation measures was used to assess the diagnostic performance, which reflects overall accuracy, error distribution, and the model’s ability to distinguish clinically meaningful malignant cases.

Results

Experimental results demonstrate that the proposed framework, which couples transfer-learned deep features with CatBoost, achieves superior discrimination between benign and malignant thyroid nodules compared with conventional approaches. It exhibits improved robustness across variations in ultrasound appearance and maintains stable performance without the need for extensive parameter tuning.

Discussion

These findings highlight the potential of the proposed method as an efficient and reliable tool for computer-aided thyroid nodule diagnosis. They also underscore the framework’s suitability for integration into real-world clinical workflows, which could further optimize clinical decision-making and reduce unnecessary medical interventions.

Schlagwörter

Stichwörter der Autor:innen

computer-aided diagnosisdeep transfer learningmachine learningmedical image classificationthyroid nodulesultrasound imaging

Autor:innen und Institutionen

  • School of Automation and Electrical Engineering, Shenyang Ligong University, Shenyang, China.
  • Department of Ultrasound, Shengjing Hospital of China Medical University, Shenyang, China.

Zitieren

APA

Zhang, D., Li, B., Ju, H., Li, T., & Zhang, Y. (2026). Hybrid deep feature and machine learning framework for classification of thyroid nodules in ultrasound images. Frontiers in Oncology, 16. https://doi.org/10.3389/fonc.2026.1784411

Vancouver

Zhang D, Li B, Ju H, Li T, Zhang Y. Hybrid deep feature and machine learning framework for classification of thyroid nodules in ultrasound images. Frontiers in Oncology. 2026;16. doi:10.3389/fonc.2026.1784411

BibTeX
@article{Zhang2026,
  author = {Zhang, Dingnan and Li, Bo and Ju, Hao and Li, Tingxue and Zhang, Yanzhu},
  title = {Hybrid deep feature and machine learning framework for classification of thyroid nodules in ultrasound images},
  journal = {Frontiers in Oncology},
  year = {2026},
  volume = {16},
  doi = {10.3389/fonc.2026.1784411}
}
Kostenlos starten →