
RSIP Vision announced today that it is developing an AI-based solution for 3D knee reconstruction from X-ray images.
Tel Aviv, Israel-based RSIP Vision’s technology is designed to give physicians 3D modeling of each bone, providing data for surgery planning and implant fitting while also improving workflow and cutting costs compared to the currently used high-radiation methods.
Physicians could potentially receive an accurate 3D anatomical model of a knee, enabling optimal pre-op planning and precise implant tailoring, all from two X-ray images of AP and lateral views, according to RSIP. The model is based on convolutional neural networks, which, along with synthetic data creation, can create a 3D model that would be low-cost, widely available and offer low-radiation imaging modality, according to a news release.
“The most important factor for achieving high accuracy and success rates is preoperative planning, during which the orthopedic surgeon normally uses an expensive, time consuming, scarcely available modalities (CT, MRI…),” Dr. Rabeeh Fares, a radiologist at Tel Aviv’s Sourasky Medical Center, said in the news release. “To overcome this obstacle, RSIP Vision developed a new technology, based solely on X-ray beams that are readily available, accurate, cheap and reproducible, which could be applied in poor peripheral as in rich central areas, while preserving the segmentation quality.”