KneeBones3Dify: Open-source software for segmentation and 3D reconstruction of knee bones from MRI data
Year: 2024
Type: software-paper
Abstract: KneeBones3Dify is a Python software tool that supports detailed analysis of knee pathologies and preoperative planning for knee replacement surgery based on patient-specific 3D models. It produces printable 3D bones in a stereolithography file format by automatically segmenting the femur, patella, and tibia from high-resolution Magnetic Resonance (MR) images with nearly isotropic voxel dimensions. Our software avoids time-consuming and subjective manual segmentation by specialists, offering an accurate and efficient alternative employing GPU acceleration. We validated the results by computing objective metrics against the ground truth voxel-wise segmentation produced for a 3D MR image by specialists, who also confirmed the reconstruction accuracy qualitatively. KneeBones3Dify and annotated data are publicly available, enabling broader research and clinical practice use.
Source: SoftwareX
Institutions: Institute for High Performance Computing and Networking, National Research Council, Institute of Biostructure and Bioimaging, University of Florence, University of Naples Federico II, Istituto Nazionale Tumori IRCCS "Fondazione G. Pascale", Fondazione Idis Città della Scienza, Esaote (Italy)
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $970
OpenAlex ID: W4401820638