CTA-UNet: CNN-transformer architecture UNet for dental CBCT images segmentation.

In view of the limitations of current deep learning models in segmenting dental cone-beam computed tomography (CBCT) images, specifically dealing with complex root morphological features, fuzzy boundaries between tooth roots and alveolar bone, and the need for costly annotation of dental CBCT images. We collected dental CBCT data from 200 patients and annotated 45 of them for network training, and proposed a CNN-Transformer Architecture UNet network, which combines the advantages of CNN and Transformer. The CNN component effectively extracts local features, while the Transformer captures long-range dependencies. Multiple spatial attention modules were included to enhance the network's ability to extract and represent spatial information. Additionally, we introduced a novel Masked image modeling method to pre-train the CNN and Transformer modules simultaneously, mitigating limitations due to a smaller amount of labeled training data. Experimental results demonstrate that the proposed method achieved superior performance (DSC of 87.12%, IoU of 78.90%, HD95 of 0.525 mm, ASSD of 0.199 mm), and provides a more efficient and effective approach to automatically and accurately segment dental CBCT images, has real-world applicability in orthodontics and dental implants.

[Vessels' Air Pollutant Emissions Inventory and Emission Characteristics in the Xiamen Emission Control Area].

Based on the automatic identification system (AIS) data and large field survey datasets for Xiamen port, the activity-based approach was used to calculate the emissions from each sailing ship in the Xiamen Emission Control Area (XECA), and to obtain the 2018 air emissions inventory for the XECA. This study subsequently analyzed the emission characteristics and spatiotemporal distribution characteristics of pollutants. The results showed that in 2018, the total amount of pollutants discharged from ships in the XECA was 16413 t, of which 82.2% were from ships entering and leaving the port and 17.8% were from ships outside of the port. NOx emissions were the highest among all of the pollutants and accounted for 64.2% of the total. Comparing the results of the five modes, emissions at berth were the highest, which was followed by the cruise mode, reduced speed-zone mode and maneuvering mode, and finally, the hoteling mode. In addition, the analysis indicated that the main source of pollutant emissions in Xiamen Port was cargo ships, of which, container ships contributed the most. The peak period of pollutant emissions from ships was between 09:00 and 16:00. The emission value during February was the lowest over the year, whereas the highest emission values occurred mostly during March and May. In terms of the spatial distribution, this study revealed that the main channel and port coastline had the highest emission values.