Evaluation of effects of small-incision approach treatment on proximal tibia fracture by deep learning algorithm-based magnetic resonance imaging.

研究概要

40 patients with proximal tibial fractures were randomized to receive either a small-incision approach (n=22) or standard of care (common access; n=18). To evaluate the outcomes, a deep learning algorithm (super-resolution reconstruction) was used to reconstruct the magnetic resonance imaging (MRI) images. Outcomes of interest included operative time, intra-operative blood loss, knee range of motion, full load time, complete healing time, therapeutic effects, peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM). Operative time, intra-operative blood loss, and knee range of motion all favored the small-incision group compared to the common access group. Therapeutic effects (excellent, good, and different) all favored the small-incision group. Overall, PSNR and SSIM were significantly better after 3D reconstruction than before.