Publications by authors named "Bingchuan Li"
Carbon materials are promising anode materials for rechargeable lithium and sodium-ion batteries, due to their low cost, high capacity, and structural designability. In this work, we selected a waste biomass, garlic stem, as the carbon precursor, and we systematically investigated the effect of pyrolysis temperature and time on their battery performance. We find that 800 °C and 2 h are the best pyrolysis conditions, which leads to the optimal carbon material (800C-2H) with a large layer spacing, abundant defect sites, high surface area, and sufficient micro/meso-porous structures.
View Article and Find Full Text PDFAttention-Guided Convolutional Neural Network for Detecting Pneumonia on Chest X-Rays.
Annu Int Conf IEEE Eng Med Biol Soc
July 2019
Article Synopsis
- Pneumonia is a widespread infectious disease, and chest X-ray (CXR) is the primary method for diagnosis, but interpreting these images can be challenging due to the similarity between various conditions.
- A new convolutional neural network (CNN) model is proposed to enhance pneumonia detection by erasing pneumonia areas in CXR images and treating them as non-pneumonia samples, helping the model focus specifically on the disease.
- The results indicate that this improved CNN model outperforms existing leading detection methods in both accuracy and minimizing false positives.