Application value of a deep learning method based on a 3D V-Net convolutional neural network in the recognition and segmentation of the auditory ossicles.

Objective: To explore the feasibility of a deep learning three-dimensional (3D) V-Net convolutional neural network to construct high-resolution computed tomography (HRCT)-based auditory ossicle structure recognition and segmentation models. Methods: The temporal bone HRCT images of 158 patients were collected retrospectively, and the malleus, incus, and stapes were manually segmented. The 3D V-Net and U-Net convolutional neural networks were selected as the deep learning methods for segmenting the auditory ossicles. The temporal bone images were randomized into a training set (126 cases), a test set (16 cases), and a validation set (16 cases). Taking the results of manual segmentation as a control, the segmentation results of each model were compared. Results: The Dice similarity coefficients (DSCs) of the malleus, incus, and stapes, which were automatically segmented with a 3D V-Net convolutional neural network and manually segmented from the HRCT images, were 0.920 ± 0.014, 0.925 ± 0.014, and 0.835 ± 0.035, respectively. The average surface distance (ASD) was 0.257 ± 0.054, 0.236 ± 0.047, and 0.258 ± 0.077, respectively. The Hausdorff distance (HD) 95 was 1.016 ± 0.080, 1.000 ± 0.000, and 1.027 ± 0.102, respectively. The DSCs of the malleus, incus, and stapes, which were automatically segmented using the 3D U-Net convolutional neural network and manually segmented from the HRCT images, were 0.876 ± 0.025, 0.889 ± 0.023, and 0.758 ± 0.044, respectively. The ASD was 0.439 ± 0.208, 0.361 ± 0.077, and 0.433 ± 0.108, respectively. The HD 95 was 1.361 ± 0.872, 1.174 ± 0.350, and 1.455 ± 0.618, respectively. As these results demonstrated, there was a statistically significant difference between the two groups (P < 0.001). Conclusion: The 3D V-Net convolutional neural network yielded automatic recognition and segmentation of the auditory ossicles and produced similar accuracy to manual segmentation results.

Osteopontin's colocalization with the adhesion molecule CEACAM5 in cytoplasm of carcinoma of tongue and its correlation with the invasion of that diease.

The purpose of this study was to investigate the expression of carcinoembryonic antigen-related cell adhesion molecule 5 (CEACAM5) and correlate it with OPN expression and function in squamous carcinoma of tongue.Paraffin were sections of 80 samples with squamous carcinoma of tongue and 40 samples with normal tissue of tongue for benign lesion having undergone surgery. Immunohistochemistry (IHC) was used to study the distribution of CEACAM5 and OPN, and double-labeling immunohistochemistry was used to observe the relationship between CEACAM5 and OPN expression.CEACAM5 and OPN are found in normal tissue of tongue, but with different expression pattern. CEACAM5 expression mainly with membranous staining is restricted on the superficial epithelium. However, OPN expression with mainly cytoplasmic staining is restricted on the deep epithelium. No colocalization of CEACAM5 and OPN have been observed in normal tissue of tongue. In squamous carcinoma of tongue, CEACAM5 expression with cytoplasmic staining is different from normal tongue tissue with membranous staining, and the transformation of CEACAM5 distribution from membrane to cytoplasm is an important incident for the invasion and differentiation of tumor. CEACAM5 and OPN are colocalized in cytoplasm, and a significant correlation was observed between the positive colocalization and the negative colocalization in the depth of invasion and the differentiation of the tumor.