The Effect of Humanized Nursing Intervention Guided by Computed Tomography Images on Elderly Patients Undergoing Anesthesia for Femur Intertrochanteric Fractures under Intelligent Reconstruction Algorithm

This research was aimed at analyzing the effect of humanized nursing intervention combined with computed tomography (CT) imaging in the surgical anesthesia of femur intertrochanteric fractures (FIF) in the elderly. An image reconstruction algorithm was proposed based on nonlocal mean (NLM) algorithm, which was named as ONLM, and its performance was analyzed. A total of 114 elderly patients with FIF were equally and randomly divided into a humanized nursing group (57 cases) and a routine nursing group (57 cases). They were performed with CT imaging scan based on the ONLM algorithm, and the clinical indicators of the two groups of patients were recorded. The root mean square error (RMSE) and mean absolute error (MAE) of the CT images constructed using the ONLM algorithm were significantly lower than those using NLM algorithm, edge filtering algorithm, and total variation model, while the peak signal-to-noise ratio (PSNR) was the opposite ( P < 0.05 ). The operation time, hospitalization days, intraoperative blood loss, postoperative drainage, and anesthesia preparation time of patients in the humanized nursing group were significantly lower than those in the routine nursing group. The number of patients with excellent Harris scores in the humanized nursing group was higher than that in the routine nursing group, and the number of patients with poor Harris scores was lower ( P < 0.05 ). The language pain score, facial pain score, and visual analog simulation (VAS) scores of patients in the humanized nursing group were significantly lower than those in the routine nursing group. The numbers of postoperative hip varus and fracture nonunion cases in the humanized nursing group were significantly more than those in the routine nursing group. In short, CT images constructed by the ONLM showed higher performance than those by the traditional algorithm. In addition, CT images constructed by ONLM combined with humanized nursing intervention could more effectively improve the cooperation of patients with surgical anesthesia, reduce surgical pain and fear of patients, improve the prognosis of patients, and lower the occurrence of adverse events.