Deep Learning-Based Postoperative Recovery and Nursing of Total Hip Arthroplasty

Abstract

Objective. To develop a deep learning-assisted recovery and nursing system after total hip arthroplasty and to conduct clinical trials in order to verify its accuracy. Methods. In our study, based on manual labeling, the human hip X-ray image library was established, and the deep neural network based on Mask R-CNN was built. The labeled medical images were used to train the model, providing reference for nursing decision after hip replacement. A total of 80 patients with hip injury from 2016 to 2019 were selected for the study. In our paper, the patients were divided into experimental group and control group. The pertinence and effectiveness of the model for postoperative care were evaluated by comparing the hip pain (VAS index), recovery (Harris score), self-care ability (Barthel index), and postoperative complication rate between the two groups. Results. The pain and complications in the experimental group were significantly lower than those in the control group, the difference being statistically significant (P<0.05); the recovery of hip joint and self-care ability were higher than those in the control group, the difference being statistically significant (P<0.05); the other differences were not statistically significant (P>0.05). Conclusion. The application of deep learning method in the rapid nursing after total hip replacement can significantly improve the nursing ability. Compared with the traditional method, it has stronger pertinence, faster postoperative recovery, lower incidence of complications, and greatly improves the postoperative quality of life of patients with hip injury.