Artificial Intelligence

While it is well known that population differences from genetics, sex, race, and environmental factors contribute to disease, AI studies in medicine have largely focused on locoregional patient cohorts with less diverse data sources. Such limitation stems from barriers to large-scale data share and ethical concerns over data privacy. Federated learning (FL) is one potential pathway for AI development that enables learning across hospitals without data share.

This data set is from the pictures collected at the industrial production site of the power adapter.The dataset contains surface defects such as scratches, glue spills, dirt, dirty spots, and off-frame labels that occur during the production process of the power adapter.  The original dataset contains 235 images with a height and width of 2448 and 2048, which are annotated in VOC2007 format. These defects were categorized into five classes: label, mark, scratch, smudge, and spill, with each class containing 32, 38, 58, 80, and 32 images.

Completing the docking between two multi-rotor unmanned aerial vehicles (UAVs) in the air has great significance for expanding the application scenarios of multiple UAVs and improving their endurance capability. In this regard, this paper first designs a safe and autonomous docking scheme and constructs a sub-mother UAV system.

Predicting future events always comes with uncertainty, but traditional non-probabilistic methods cannot distinguish certain from uncertain predictions. In survival analysis, probabilistic methods applied to state-of-the-art solutions in the healthcare and biomedical field are still novel and their implications have not been fully evaluated. In this paper, we study the benefits of modeling uncertainty in deep neural networks for survival analysis with a focus on prediction and calibration performance.