Automatic tool detection in X-ray images for robotic assisted joint fracture surgery

YingLiang Ma, Giulio Dagnino, Ioannis Georgilas, Sanja Dogramadzi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)


Long-bone fractures such as femur fractures are very common in trauma centers. Robotic assisted fracture surgery (RAFS) can facilitate the minimally invasive surgery which reduces scarring, infection risk and long hospital stays. One important step in RAFS is to establish the coordinate system link between patient joint (rigidly connected with the robotic system) and an external tracking system. As X-ray fluoroscopic images are routinely used during the procedure, an automatic method is proposed to detect and localize landmarks on the tracking tool using live X-ray image. The proposed method uses combination of block detection, geometric model matching and principle component analysis. A successful rate of 91.3% is achieved after testing on 650 X-ray images and accuracy is within 0.5 mm.
Original languageEnglish
Title of host publicationProceedings - 2017 IEEE International Conference on Internet of Things, IEEE Green Computing and Communications, IEEE Cyber, Physical and Social Computing, IEEE Smart Data, iThings-GreenCom-CPSCom-SmartData 2017
PublisherThe Institute of Electrical and Electronics Engineers (IEEE)
Number of pages5
Publication statusPublished - 1 Jun 2018

Cite this