Generative artificial intelligence (AI), which uses the probability distribution of real-world data and generates new content by learning it, is becoming a powerful tool for use in medicine. In this paper, we will systematically review the progress made in research, major applications, and challenges of the use of generative AI in the medical field. Firstly, we give a brief introduction to the probabilistic background of generative models and then describe four kinds of models, including generative adversarial network, diffusion model, variational autoencoder (VAE), and large language model. Technical features and their potential usage in the medical field will be discussed. Secondly, we systematically investigate their applications in five fields, which include clinical decision-making, synthetic medical imaging, drug discovery, processing of medical literature, and medical education. From our review, it is clear that generative AI has a lot of capabilities when it comes to increasing efficiency and accuracy in health care delivery, drug discovery, and medical education. Nonetheless, there are several challenges associated with this technology, including factual errors, hallucinations, data privacy concerns, algorithmic bias, and difficulties in attributing responsibility. In light of these limitations, it is vital for generative AI technology to play the role of an assistant to medical professionals as opposed to taking their roles. The future will require the creation of reliable technologies like uncertainty quantification, explainable AI, and retrieval-augmented generation (RAG).

Aluminum die-cast products often exhibit surface defects that vary in type and severity depending on product function and design requirements. Current defect detection primarily relies on manual inspection, which demands significant expertise, raises health and fatigue concerns, and is prone to human error. Automated defect detection offers a promising solution to reduce costs, improve efficiency, resolve occupational safety and health concerns, and mitigate challenges in labor shortages and training. This paper presents an ensemble deep learning (DL) approach for detecting surface defects in aluminum die-cast lids for residential gas meters, a quality-critical component with stringent safety standards. Specifically, we implement and evaluate three state-of-the-art DL architectures: a convolutional neural network (CNN), residual networks (ResNet-18), and Vision Transformer (ViT). In addition, we develop an ensemble model to further enhance performance. We leverage grid search and cross-validation for hyperparameter tuning and train/test each model ten times for comprehensive performance evaluation. Experiments on a large real-world dataset demonstrate that all models achieve high accuracy, precision, and recall, with CNN and ResNet-18 slightly outperforming ViT. The ensemble model further improves prediction accuracy and robustness. The paired t-tests showed that the ensemble model significantly performed better compared to CNN and ViT model. In summary, this study contributes to the advancement of automated inspection of surface defects in die-cast products by systematically comparing state-of-the-art deep learning methods, discussing model selection criteria, and optimizing ensemble strategies. Centered on CNN, ResNet-18, and ViT architectures, it proposes a rigorous methodological framework for surface defect detection and provides a foundational basis for subsequent research in in-situ quality control. Our codes are available at https://github.com/Alexruoyun/Aluminum-Die-Casting-Surface-Defect-Detection.

Autonomous vehicles require effective prediction of potential future trajectories of surrounding agents. The current trajectory prediction methods have limitations, firstly, traditional feature fusion methods merge scene features sequentially in a simplistic manner, often overlooking the intricate interrelations among scene elements, leading to incomplete selection and insufficient utilization of useful features; secondly, in multimodal trajectory prediction, the mode collapse issue inherent to probabilistic approaches results in inadequate expression of agent intent uncertainty, while overly anchor-dependent proposal-based methods can generate implausible trajectories. To address these limitations, We present a Dynamic scene and Relational reasoning Transformer (DRTR), a novel multimodal trajectory prediction method based on dynamic scene encoding and relational reasoning. A pivotal aspect of DRTR is the dynamic closed-loop modeling framework that effectively combines scene features to output three dynamic features: dynamic traffic flow, dynamic agents, and interactions between agents. This innovative framework ensures a comprehensive capture of the dynamic scene and its intricate interrelations. Then, DRTR initializes a set of trajectory suggestions representing various modalities and carefully refines these suggestions by sequentially fusing and querying dynamic scene features, ensuring predictions are both accurate and reflect multimodality. To further enhance model expressiveness, we introduce a feature selection network based on relational reasoning, which can recognize deep relationships between scene elements and select beneficial contextual features. Experiments on the Argoverse 1 dataset indicate that DRTR exhibits superior performance, particularly in multimodal trajectory prediction.