Ultrasound imaging has been used to investigate compression of the median nerve in carpal tunnel syndrome patients. Ultrasound imaging and the extraction of median nerve parameters from ultrasound images are crucial and are usually performed manually by experts. The manual annotation of ultrasound images relies on experience, and intra- and interrater reliability may vary among studies. In this study, two types of convolutional neural networks (CNNs), U-Net and SegNet, were used to extract the median nerve morphology. To the best of our knowledge, the application of these methods to ultrasound imaging of the median nerve has not yet been investigated. Spearman’s correlation and Bland–Altman analyses were performed to investigate the correlation and agreement between manual annotation and CNN estimation, namely, the cross-sectional area, circumference, and diameter of the median nerve. The results showed that the intersection over union (IoU) of U-Net (0.717) was greater than that of SegNet (0.625). A few images in SegNet had an IoU below 0.6, decreasing the average IoU. In both models, the IoU decreased when the median nerve was elongated longitudinally with a blurred outline. The Bland–Altman analysis revealed that, in general, both the U-Net- and SegNet-estimated measurements showed 95% limits of agreement with manual annotation. These results show that these CNN models are promising tools for median nerve ultrasound imaging analysis.
To clarify the effects of reaction mechanisms on soot formation, a two-dimensional numerical simulation was performed for a burner-stabilized stagnation flame. The GRI-3.0, Appel-Bockhorn-Frenklach (ABF), and DLR models were employed as representative reaction mechanisms. The soot particle formation model was a two-equation model, and only acetylene was assumed to be the soot-nucleating species. The ambient pressure and inlet temperature were 0.1 MPa and 473 K, respectively, and the distance between the burner surface and the stagnation plate (Hp) varied from 5.5 to 12 mm. The simulation was conducted using OpenFOAM, and the results were validated with experimental data based on the temperature distribution and soot volume fraction at Hp. After soot nucleation began, the number density became large in the preheating zone. As the soot transported downstream, coagulation and surface growth accelerated, and the mass density became large. Furthermore, while the effect of the reaction mechanism on temperature distribution was small, the effect on the soot growth rate was large. In the preheating zone, the DLR model showed the largest nucleation rate, which could be due to the abundant acetylene formation reaction. At the downstream, the DLR model showed the smallest nucleation rate, which is probably because many acetylene consumption reactions were included in the DLR model. As a result, the soot volume fraction of the DLR model at Hp was smaller than that of the GRI-3.0 and ABF models. These results suggest that the reaction mechanism, especially the fuel-pyrolysis sub-mechanism, has a significant impact on the prediction of soot formation.
28 October 2026 in Proceedings of the Combustion Institute