i3PosNet: instrument pose estimation from X-ray in temporal bone surgery

• DOI: 10.1007/s11548-020-02157-4
• Publication Date: 2020-05-21T05:02:41Z
• Authors: David Kügler; Jannik Sehring; Andrei Stefanov; Igor Stenin; Julia Kristin; Thomas Klenzner; Jörg Schipper; Anirban Mukhopadhyay
• Keywords: Image-guided surgery; Position (finance)

Purpose: Accurate estimation of the position and orientation (pose) surgical instruments is crucial for delicate minimally invasive temporal bone surgery. Current techniques lack in accuracy and/or line-of-sight constraints (conventional tracking systems) or expose patient to prohibitive ionizing radiation (intra-operative CT). A possible solution capture instrument with a c-arm at irregular intervals recover pose from image. Methods: i3PosNet infers images using network. Said framework considers localized patches outputs pseudo-landmarks. The reconstructed pseudo-landmarks by geometric considerations. Results: We show reaches errors less than 0.05mm. It outperforms conventional image registration-based approaches reducing average maximum least two thirds. trained on synthetic generalizes real x-rays without any further adaptation. Conclusion: translation Deep Learning based methods applications difficult, because large representative datasets training testing are not available. This work empirically shows sub-millimeter solely data.