Deep segmentation leverages geometric pose estimation in computer‐aided total knee arthroplasty

Abstract

Knee arthritis is a common joint disease that usually requires a total knee arthroplasty. There are multiple surgical variables that have a direct impact on the correct positioning of the implants, and an optimal combination of all these variables is the most challenging aspect of the procedure. Usually, preoperative planning using a computed tomography scan or magnetic resonance imaging helps the surgeon in deciding the most suitable resections to be made. This work is a proof of concept for a navigation system that supports the surgeon in following a preoperative plan. Existing solutions require costly sensors and special markers, fixed to the bones using additional incisions, which can interfere with the normal surgical flow. In contrast, the authors propose a computer-aided system that uses consumer RGB and depth cameras and do not require additional markers or tools to be tracked. They combine a deep learning approach for segmenting the bone surface with a recent registration algorithm for computing the pose of the navigation sensor with respect to the preoperative 3D model. Experimental validation using ex-vivo data shows that the method enables contactless pose estimation of the navigation sensor with the preoperative model, providing valuable information for guiding the surgeon during the medical procedure.

Towards Markerless 3D Pose Estimation for Computer Assisted Orthopaedic Surgery: a Comparison Study of Depth Cameras

“…Indeed, the Kinect Azure shows a distance map mean registration error of 1.13mm, which could be suitable for TKA. Moreover, the results are very encouraging compared to the state of the art: [4] is similar to our work and intends to register intraoperatively bone surface with depth camera with a precision of 6.18mm. [5] also reports an Figure 1: Statistical precision of compared depth cameras for 3D pose estimation higher average registration error of 11.46mm on real data for similar task.…”

Markerless Navigation System for Orthopaedic Knee Surgery: A Proof of Concept Study

“…Neural networks can be trained to "learn" comprehensive semantics using labelled dataset. A proof-of-concept study was outlined in [14], and a more systematic validation was provided by [13]. Both works used a consumer-level RGB-D camera to record surgical site.…”

“…Automatic markerless target tracking and registration are desired to reduce such risks and encourage the broader acceptance of CAOS systems. Markerless tracking and registration algorithm has been proposed in [13], [14] for knee surgeries. A deep neural network was trained to segment the target femur from the RGB-D capture of surgical site.…”

“…The best-reported registration accuracy is 6.66 • and 2.74 mm when the target is held static. Such accuracy (especially its angular component) is not acceptable for clinical applications [14]. Furthermore, such markerless technique has not been applied to surgical navigation.…”

A U-Net Based Approach for Automating Tribological Experiments

“…CNN-based algorithms are the top performing solutions for the PASCAL VOC 2012 [ 12 , 13 , 14 ] dataset, cityscapes [ 15 , 16 , 17 ], and ADE20K [ 10 , 18 , 19 ]. There have also been multiple proposals for using CNNs to analyze endoscopic camera images, predominantly in the medical field [ 20 , 21 , 22 ].…”