[공지사항]을 빙자한 안부와 근황 
Show more

Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

A Deep Learning-Based Ensemble System for Automated Shoulder Fracture Detection in Clinical Radiographs

Created by
  • Haebom

Author

Hemanth Kumar M, Karthika M, Saianiruth M, Vasanthakumar Venugopal, Anandakumar D, Revathi Ezhumalai, Charulatha K, Kishore Kumar J, Dayana G, Kalyan Sivasailam, Bargava Subramanian

Outline

In this paper, we developed a multi-model deep learning system to aid in the early diagnosis of shoulder fractures, which are often missed in emergency and high-volume settings. Using 10,000 annotated shoulder X-ray images, we developed models based on architectures such as Faster R-CNN, EfficientDet, and RF-DETR, and applied bounding box and classification-level ensemble techniques such as Soft-NMS, WBF, and NMW fusion. As a result, the NMW ensemble achieved an accuracy of 95.5% and an F1 score of 0.9610, outperforming the individual models in all key metrics. This confirms its effectiveness for clinical fracture detection on shoulder X-ray images. The model in this study is limited to binary fracture detection to support rapid triage and classification.

Takeaways, Limitations

•
Takeaways:
◦
We demonstrate that ensemble-based AI can reliably detect fractures in shoulder X radiographs with high accuracy.
◦
The model's accuracy and deployment readiness suggest its potential for integration into real-time diagnostic workflows.
•
Limitations:
◦
It is only capable of detecting binary fractures and cannot be applied to detailed orthopedic classification.
◦
Further research may be needed on the generalization performance of the model.
👍
Made with Slashpage