EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Deep learning with attention modules and residual transformations improves hepatocellular carcinoma (HCC) differentiation using multiphase CT

Yuenan Wang iD, Wanwei Jian, Zhidong Yuan, Fada Guan iD, David Carlson

Precision Radiation Oncology · 2025-03 · 9 (1) · 13-22

DOI: 10.1002/pro6.70003 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Precision Radiation Oncology, online erschienen am 22.03.2025. Verfasst von Yuenan Wang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Therapeutic Radiology Yale University School of Medicine New Haven Connecticut USA. Die Arbeit wurde bisher 30-mal zitiert (Crossref) und nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

We hypothesize generative adversarial networks (GAN) combined with self‐attention (SA) and aggregated residual transformations (ResNeXt) perform better than conventional deep learning models in differentiating hepatocellular carcinoma (HCC). Attention modules facilitate concentrating on salient features and suppressing redundancies, while residual transformations can reuse relevant features. Therefore, we aim to propose a GAN+SA+ResNeXt deep learning model to improve HCC prediction accuracy.

Methods

228 multiphase CTs from 57 patients were retrospectively analyzed with local IRB's approval, where 30 patients were pathologically confirmed with HCC and the rest 27 were non‐HCC. Pre‐processing of automatic liver segmentation and Hounsfield unit (HU) normalization was performed, followed by deep learning training with five‐fold cross validation in a conventional 3D GAN, a 3D GAN+A, and a 3D GAN+A+ ResNeXt, respectively (training: testing ∼ 4:1). Area under receiver operating characteristics curves (AUROC), accuracy, sensitivity and specificity of HCC prediction were evaluated.

Results

Results showed the proposed method had larger AUROC (95%), better accuracy (91%) and sensitivity (93%) with acceptable specificity (88%) and prediction time (0.04s). Deep GAN with attentions and residual transformations for HCC diagnosis using multiphase CT is feasible and favorable with improved accuracy and efficiency, which harbors clinical potentials in differentiating HCC from other benign or malignant liver lesions.

Schlagwörter

Stichwörter der Autor:innen

dynamic contrast enhanced (DCE) CTgenerative adversarial networks (GAN)hepatocellular carcinoma (HCC)

Autor:innen und Institutionen

  • Department of Therapeutic Radiology Yale University School of Medicine New Haven Connecticut USA.
  • Department of Radiation Oncology Hartford HealthCare - St Vincent Medical Center Bridgeport Connecticut USA.
  • Department of Radiotherapy the Second Affiliated Hospital of Guangzhou University of Chinese Medicine Guangzhou China.
  • Department of Diagnostic Radiology Peking University Shenzhen Hospital Shenzhen China.

ORCID-Profile: Yuenan Wang, Fada Guan

Zitieren

APA

Wang, Y., Jian, W., Yuan, Z., Guan, F., & Carlson, D. (2025). Deep learning with attention modules and residual transformations improves hepatocellular carcinoma (HCC) differentiation using multiphase CT. Precision Radiation Oncology, 9(1), 13-22. https://doi.org/10.1002/pro6.70003

Vancouver

Wang Y, Jian W, Yuan Z, Guan F, Carlson D. Deep learning with attention modules and residual transformations improves hepatocellular carcinoma (HCC) differentiation using multiphase CT. Precision Radiation Oncology. 2025;9(1):13-22. doi:10.1002/pro6.70003

BibTeX
@article{Wang2025,
  author = {Wang, Yuenan and Jian, Wanwei and Yuan, Zhidong and Guan, Fada and Carlson, David},
  title = {Deep learning with attention modules and residual transformations improves hepatocellular carcinoma (HCC) differentiation using multiphase CT},
  journal = {Precision Radiation Oncology},
  year = {2025},
  volume = {9},
  number = {1},
  pages = {13--22},
  doi = {10.1002/pro6.70003}
}
Kostenlos starten →