EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An Interpretable Hybrid SFNet Deep Learning Framework for Multi-Site Bone Fracture Detection in Medical Imaging

Wijdan S Aljebreen iD, Da'ad Albahdal iD, Shuaa S Alharbi iD, Naif S Alshammari iD, Haifa F Alhasson iD

Diagnostics · 24.03.2026 · 16 (7) · 966

DOI: 10.3390/diagnostics16070966 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 24.03.2026. Verfasst von Wijdan S Aljebreen und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information Technology, College of Computer, Qassim University, Buraydah 52571, Saudi Arabia. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deanship of Graduate Studies and Scientific Research at Qassim University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: Accurate bone fracture detection is essential for orthopedic diagnosis and trauma management. Manual interpretation of X-ray or CT images can be time-consuming and may lead to inter-observer variability, particularly in subtle or multi-site fracture cases. This study proposes an interpretable Hybrid Selective Feature Network (Hybrid SFNet) to improve multi-site bone fracture detection performance and boundary localization.

Methods

The proposed Hybrid SFNet extends the original SFNet architecture by incorporating multi-scale convolutional feature extraction and a semantic flow mechanism to enhance structural representation and fracture boundary delineation. Preprocessing techniques, including Canny edge detection, normalization, and data augmentation, were applied to improve feature quality. Model interpretability was addressed using Gradient-weighted Class Activation Mapping (Grad-CAM) to visualize regions contributing to predictions. The model was evaluated on publicly available multi-site fracture datasets using both standard and class-weighted loss configurations.

Results

For binary fracture classification, the proposed model achieved 90 accuracy, 94% precision, 77% recall, and an F1-score of 85% for fractured cases. When class-weighted loss was applied, recall improved to 85%, reducing false negatives from 145 to 94 cases (approximately 35%). Under the weighted configuration, Cohen’s Kappa reached 0.79 and the Matthews Correlation Coefficient (MCC) reached 0.76.

Conclusions

The proposed Hybrid SFNet provides an interpretable and effective framework for multi-site bone fracture detection. The integration of multi-scale feature extraction and semantic flow mechanisms enhances detection performance and boundary localization, while Grad-CAM supports clinical interpretability. These results indicate the model’s potential for supporting clinical decision-making in orthopedic imaging.

Schlagwörter

Stichwörter der Autor:innen

artificial intelligencebone fracture detectiondeep learning modelsdiagnosisgrad-CAMhybrid SFNetimage segmentationmachine learningmedical imaging

Autor:innen und Institutionen

  • Department of Information Technology, College of Computer, Qassim University, Buraydah 52571, Saudi Arabia.
  • Department of Computer Sciences, College of Computer and Information Sciences, Majmaah University, Majmaah 11952, Saudi Arabia.

ORCID-Profile: Wijdan S Aljebreen, Da'ad Albahdal, Shuaa S Alharbi, Naif S Alshammari, Haifa F Alhasson

Förderung

  • Deanship of Graduate Studies and Scientific Research at Qassim University

Zitieren

APA

Aljebreen, W. S., Albahdal, D., Alharbi, S. S., Alshammari, N. S., & Alhasson, H. F. (2026). An Interpretable Hybrid SFNet Deep Learning Framework for Multi-Site Bone Fracture Detection in Medical Imaging. Diagnostics, 16(7), 966. https://doi.org/10.3390/diagnostics16070966

Vancouver

Aljebreen WS, Albahdal D, Alharbi SS, Alshammari NS, Alhasson HF. An Interpretable Hybrid SFNet Deep Learning Framework for Multi-Site Bone Fracture Detection in Medical Imaging. Diagnostics. 2026;16(7):966. doi:10.3390/diagnostics16070966

BibTeX
@article{Aljebreen2026,
  author = {Aljebreen, Wijdan S and Albahdal, Da'ad and Alharbi, Shuaa S and Alshammari, Naif S and Alhasson, Haifa F},
  title = {An Interpretable Hybrid SFNet Deep Learning Framework for Multi-Site Bone Fracture Detection in Medical Imaging},
  journal = {Diagnostics},
  year = {2026},
  volume = {16},
  number = {7},
  pages = {966},
  doi = {10.3390/diagnostics16070966}
}
Kostenlos starten →