Search

Accurate inversion of PM2.5 concentration is crucial for haze management. Currently commonly used inversion methods cannot accurately invert the concentration in non-site areas, so this paper proposes a PM2.5 concentration inversion method based on an integrated learning model. The method utilises the Top Atmospheric Reflectance (TOAR), observation angle and meteorological element data as input features, and screens the important features by Random Forest, and constructs an integrated inversion model using XGBoost and LightGBM. The results show that the model built by TOAR improves R2 by 2.9% and reduces RMSE and MAE by 2.67 and 1.45, respectively, compared with the AOD-based model, and our model has an inversion accuracy of 0.95, which is better than other models. We used the model to estimate and analyse the historical PM2.5 concentration changes at Huaihe station in Tianjin, China, and the results were consistent with the trend of the actual PM2.5 concentration distribution, and it is clear that the proposed model has a high inversion accuracy.

Coronary artery stenosis detection remains a challenging task due to the complex vascular structure, poor quality of imaging pictures, poor vessel contouring caused by breathing artifacts and stenotic lesions that often appear in a small region of the image. In order to improve the accuracy and efficiency of detection, a new deep-learning technique based on a coronary artery stenosis detection framework (DCA-YOLOv8) is proposed in this paper. The framework consists of a histogram equalization and canny edge detection preprocessing (HEC) enhancement module, a double coordinate attention (DCA) feature extraction module and an output module that combines a newly designed loss function, named adaptive inner-CIoU (AICI). This new framework is called DCA-YOLOv8. The experimental results show that the DCA-YOLOv8 framework performs better than existing object detection algorithms in coronary artery stenosis detection, with precision, recall, F1-score and mean average precision (mAP) at 96.62%, 95.06%, 95.83% and 97.6%, respectively. In addition, the framework performs better in the classification task, with accuracy at 93.2%, precision at 92.94%, recall at 93.5% and F1-score at 93.22%. Despite the limitations of data volume and labeled data, the proposed framework is valuable in applications for assisting the cardiac team in making decisions by using coronary angiography results. [ABSTRACT FROM AUTHOR]