Fractures of the scaphoid bone may be treated in a minimally-invasive fashion. Conventionally, fluoroscopy is required to guide the placement of an osteosynthesis screw. In this work, an alternative method based on volumetric ultrasound is validated. Methods: The fully automatic and fast image processing pipeline involves two machine learning architectures for segmentation and registration. A pre-operatively acquired 3D bone model is registered to the 3D bone surface segmented from the intra-operative ultrasound. Screw positioning is planned in an automated fashion and evaluated in an in-vitro setting: Volumetric ultrasound images of a 3D-printed phantom of a human wrist are acquired for 22 different probe poses. For 220 test runs with different initial displacements, the resulting screw placement within a defined safe zone is evaluated. If the screw lies within the safe zone, its placement is assumed to be successful. Results: An isolated analysis of the registration results in a surface distance error of the registered meshes of 0.49 ± 0.01mm, with successful screw placement in all of the evaluated 220 test runs. The full pipeline, combining segmentation and registration, achieves a mean surface distance error of 0.79 ± 0.37mm, leading to successful screw placements for 149 out of 220 test runs. Poses not suited for the registration could be determined. Excluding these from the analysis, 139 out of 160 test runs are successful. Conclusion: The method proves to be promising when evaluating the registration alone, even given the challenging setup of sub-optimal probe positions. The experiments also demonstrate that further improvement regarding the segmentation is necessary.

For the percutaneous fixation of scaphoid fractures, navigated approaches have been proposed to facilitate screw placement. Based on ultrasound imaging, navigation can be carried out in a cost-effective and fast manner, furthermore avoiding harmful radiation. For this purpose, a fast and efficient architecture for the automated segmentation of scaphoid bone in ultrasound volume images is needed. Methods: For 2D segmentation of the scaphoid, two architectures are taken into account: 2D nnUNet and Deeplabv3+. These architectures are trained and evaluated on a newly created dataset consisting of 67 annotated in-vivo ultrasound volume images (4576 slice images). Results: In terms of Dice coefficient, the 2D nnUNet achieves 0.67 compared to 0.57 for the Deeplabv3+. In terms of distance metrics, the 2D nnUNet shows an average symmetric surface distance error of 0.66mm, while the Deeplabv3+ achieves 0.55mm. Conclusion: Fast and accurate segmentation of the scaphoid in ultrasound volumes is feasible. Both architectures show competitive results.

The number of total knee arthroplasties performed world-wide are on a rise. Patient-specific planning and implants may improve surgical outcomes, but require 3D models of the bones involved. Ultrasound may become a cheap and non-harmful imaging modality, if the short comings of segmentation techniques in terms of automation, accuracy and robustness are overcome. Furthermore, any kind of ultrasound-based bone reconstruction must involve some kind of model completion, in order to handle occluded areas, e.g., the frontal femur. A fully-automatic and robust processing pipeline is proposed, generating full bone models from 3D freehand ultrasound scanning. A convolutional neural network is combined with a statistical shape model to segment and extrapolate the bone surface. We evaluate the method in-vivo on 10 subjects, comparing the ultrasound-based model to a magnetic resonance imaging (MRI) reference. The partial freehand 3D record of the femur and tibia bones deviate by 0.7 to 0.8mm from the MRI reference. After completion and on average, the full bone model shows sub-millimetric error in case of the femur and 1.24mm in case of the tibia. Processing of the images is performed in real-time, and the final model fitting step is computed in less than one minute. On average, it took 22 minutes for a full record per subject.