Recent advances in foundation models (FMs) have enabled artificial intelligence systems to acquire generalizable capabilities, with language and visual models demonstrating adaptability across diverse healthcare applications and multimodal vision-language FMs (VLFMs) supporting complex tasks such as report generation and question answering. Building on these developments, autonomous agents, particularly vision-language agents (VLAs), represent the next frontier, extending AI from perception and recognition to cognition, decision-making, and action. By integrating multimodal understanding with planning, interaction, and tool use, VLAs introduce autonomous intelligence capable of managing multi-step clinical workflows, adapting over time, and collaborating with clinicians. This survey provides a comprehensive perspective on VLAs in healthcare. We examine their current progress, analyze the major challenges hindering their integration, and highlight promising directions for future development. Unlike prior surveys that primarily focus on diverse FMs, our work prioritizes emerging VLFMs and incorporates agentic concepts, emphasizing their transformative potential beyond narrow research prototypes. Specifically, we (ⅰ) evaluate VLAs in terms of VLFM architectures and adaptation strategies and agentic design, (ⅱ) assess their potential impact across diverse clinical workflows, and (ⅲ) discuss pathways towards next-generation clinical decision support, considering technical and clinical challenges.

Direct volume rendering (DVR) is a technique that emphasizes structures of interest (SOIs) within a volume visually, while simultaneously depicting adjacent regional information, e.g., the spatial location of a structure concerning its neighbors. In DVR, transfer function (TF) plays a key role by enabling accurate identification of SOIs interactively as well as ensuring appropriate visibility of them. TF generation typically involves non-intuitive trial-and-error optimization of rendering parameters, which is time-consuming and inefficient. Attempts at mitigating this manual process have led to approaches that make use of a knowledge database consisting of pre-designed TFs by domain experts. In these approaches, a user navigates the knowledge database to find the most suitable pre-designed TF for their input volume to visualize the SOIs. Although these approaches potentially reduce the workload to generate the TFs, they, however, require manual TF navigation of the knowledge database, as well as the likely fine tuning of the selected TF to suit the input. In this work, we propose a TF design approach, CBR-TF, where we introduce a new content-based retrieval (CBR) method to automatically navigate the knowledge database. Instead of pre-designed TFs, our knowledge database contains volumes with SOI labels. Given an input volume, our CBR-TF approach retrieves relevant volumes (with SOI labels) from the knowledge database; the retrieved labels are then used to generate and optimize TFs of the input. This approach largely reduces manual TF navigation and fine tuning. For our CBR-TF approach, we introduce a novel volumetric image feature which includes both a local primitive intensity profile along the SOIs and regional spatial semantics available from the co-planar images to the profile. For the regional spatial semantics, we adopt a convolutional neural network to obtain high-level image feature representations. For the intensity profile, we extend the dynamic time warping technique to address subtle alignment differences between similar profiles (SOIs). Finally, we propose a two-stage CBR scheme to enable the use of these two different feature representations in a complementary manner, thereby improving SOI retrieval performance. We demonstrate the capabilities of our CBR-TF approach with comparison with a conventional approach in visualization, where an intensity profile matching algorithm is used, and also with potential use-cases in medical volume visualization.