A depth-resolved quantitative evaluation method for non-carious cervical lesions treatment with optical coherence tomography.

OBJECTIVES: The aim of this study is prognostic assessment of surface smoothness and the presence of internal bubbles after treatment of non-cancerous cervical lesions (NCCLs) using optical coherence tomography (OCT). METHODS: After treatment with NCCLs, cross-sectional images of the lesion parts of the sample were non-invasively acquired and analyzed. The surface smoothness between tooth and resin, resin and cemento-enamel junction, and the presence bubble inside resin was confirmed. In addition, using an algorithm that distinguishes between resin and dental structure based on OCT cross-sectional images, we quantitatively analyzed the amount of resin used in treating NCCLs and acquired 3D images. RESULTS: The inner structure of the resin in each sample was checked, and the presence of bubbles was confirmed. In addition, the resin sections were separated from the tomographic images acquired by OCT to visualize 3D images. The volume of resin used in the treatment part of each NCCLs samples was quantitatively analyzed as 3.7216 ∼ 14.889 mm3. CONCLUSIONS: OCT is able to measure not only the surface abrasion provided by existing intraoral scanner, but also the size and depth location of interal bubbles, which is distinctive advantage of our method. Based on our results, OCT is a significant tool for qualitative and quantitative analysis of dental NCCLs treatment before and after treatment. CLINICAL SIGNIFICANCE: The study used OCT, a non-destructive diagnostic, to reveal the structure of the resin and the location and size of bubbles after NCCLs treatment. These findings could be golden standard in determining the prognosis of NCCLs treatment.

Three-dimensional reconstructing undersampled photoacoustic microscopy images using deep learning.

Spatial sampling density and data size are important determinants of the imaging speed of photoacoustic microscopy (PAM). Therefore, undersampling methods that reduce the number of scanning points are typically adopted to enhance the imaging speed of PAM by increasing the scanning step size. Since undersampling methods sacrifice spatial sampling density, by considering the number of data points, data size, and the characteristics of PAM that provides three-dimensional (3D) volume data, in this study, we newly reported deep learning-based fully reconstructing the undersampled 3D PAM data. The results of quantitative analyses demonstrate that the proposed method exhibits robustness and outperforms interpolation-based reconstruction methods at various undersampling ratios, enhancing the PAM system performance with 80-times faster-imaging speed and 800-times lower data size. The proposed method is demonstrated to be the closest model that can be used under experimental conditions, effectively shortening the imaging time with significantly reduced data size for processing.