Artificial Intelligence Image-Assisted Knee Ligament Trauma Repair Efficacy Analysis And Postoperative Femoral Nerve Block Analgesia Effect Research

OBJECTIVE: To analyze artificial intelligence image-assisted knee ligament injury repair and femoral nerve block analgesia after surgery.METHODS: Data-driven and artificial intelligence methods were adopted to systematically study magnetic resonance imaging image reconstruction, processing, and analysis. First, knee ligament reconstruction and femoral arteriography images were studied. Using the prior knowledge that the full width at half maximum of the contrast image does not change with the resolution, a constrained data exploration algorithm was proposed combined with the iterative algorithm. The algorithm could reconstruct high-resolution images using the collected low-frequency data of k-space. The experimental data and results were simulated with the enhanced knee ligaments and femoral nerve angiography images. Combining the spatial continuity of knee ligaments and femoral nerve, a multilayer input segmentation network was designed. The multisupervised network was adopted for output and had good segmentation results for the knee ligaments and femoral nerve. On this basis, a multiparametric image input speaker net was proposed to detect knee ligament injuries.RESULTS: The area under the receiver operating characteristic curve of the constructed model under the test set was 0.824, and the sensitivity and specificity were 0.800 and 0.836, respectively. The image was better than compressed sensing to reconstruct the image, which was more accurate for knee ligament and femoral nerve stenosis. The network also had higher sensitivity for knee joint trauma detection, which could aid clinicians. The postoperative femoral nerve block had a good detection effect, which could provide important information for clinical analgesia.CONCLUSIONS: The artificial intelligence image-assisted diagnosis system for analysis and processing of multiparametric magnetic resonance images is useful for clinical decision making, reducing physicians' labor intensity, improving efficiency, and lowering the rate of misdiagnosis.