Medical Oral Image Super-Resolution Reconstruction Algorithm Based on Stable Diffusion Model

To address the limitations of traditional approaches in super-resolution reconstruction of medical oral images, we have devised a novel method for medical oral image super-resolution reconstruction using a stable diffusion model called Stable Oral Reconstruction Technique (SORT). In comparison to conventional methods, our approach has demonstrated Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), recall®, and F1 scores of 92.42%, 92.44%, 91.89% and 92.37%, respectively. Additionally, we have successfully transformed Cone Beam Computed Tomography (CBCT) into high-resolution images of a resolution comparable to Micro-CT, significantly enhancing the spatial resolution and visual quality of the images. The study encompasses an in-depth exploration of the mathematical model and the development of efficient algorithms for image reconstruction. Rigorously validated experimental results underscore the foundation laid by this project for the advancement of medical oral image processing technology. Compared to traditional high-resolution image reconstruction techniques, our method provides more reliable and precise image support for clinical oral diagnostics and treatment, demonstrating superior performance and promising application prospects.