Contour-Based Bone Axis Detection for X-Ray Guided Surgery on the Knee

Abstract: Contour-based Bone Axis Detection for X-ray-guided Surgery on the Knee

“…The approach is evaluated for the femur and tibia in the knee joint and achieves a median angulation error of 0.19 • and 0.33 • respectively. An inter-rater study with three trauma surgery experts confirms reliability of the method and recommends further clinical application [1].…”

Software‐based method for automated intraoperative planning of Schoettle Point in surgical medial patellofemoral ligament reconstruction: A comparative validation study

“…To alleviate these issues, the contour can be extracted from an additional segmentation mask of the femur. 22,23 � No cross-evaluation over different dataset splits was performed.…”

“…The network can also be trained to additionally predict segmentation masks of the bones to limit the influence of joint angulation during the registration step. 22,23 The detailed, technical aspects of automated, software-based planning of the Schoettle Point have already been published and can be found in the publications by Kordon et al (2019Kordon et al ( -2022). [22][23][24]…”

“…To obtain LM3, first the femur bone contour is extracted from the original image using Canny edge detection, which is then masked with the predicted region of interest of the cortical line. To obtain a line representation that can be used for further geometric calculations, an ordinary least squares regression is performed on the resulting contour points 22,23 …”

“…The parameterisation for Canny edge detection was based on manual optimization and does not automatically adapt to different acquisition settings. To alleviate these issues, the contour can be extracted from an additional segmentation mask of the femur 22,23 …”