Abstract
The periprosthetic acetabular osteotomy (PAO) is a commonly used technique in orthopedics for treating developmental hip dysplasia and hip dislocation, as the most effective treatment for developmental dysplasia of the hip (DDH). However, performing PAO can be challenging for surgeons due to limited visibility and difficulty in detecting any deformations of osteotome chisels when they are deeply immersed in the pelvis. These challenges can result in serious complications, such as excessive bleeding and nerve injuries. We propose a novel precision tracking system to mitigate these risks by acquiring the chisel deformation in real-time. This system consists of a newly designed osteotome chisel with five built-in microsensors, which are finely chosen with the help of Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). We propose a fast finite element method (FFEM) model to calculate the deformation of the chisel from flexibility information collected by these five sensors, where the model deformation can be predicted from a well-designed light deep neural network (DNN) model. Our model has achieved an impressive R2 value of 0.98781 and an average deformation error of only 0.07 mm in nodes compared to the experiment. The prediction time of FFEM model has been shortened to 0.33 s, and the total time including three-dimensional reconstruction and visualization has been shortened to 3.84 s. Implementing such an osteotome chisel with a deformation tracking system has shown immense potential in increasing surgical accuracy and reducing medical negligence for PAO operations.