Digital holography has emerged as a powerful tool for investigating dynamic particle behaviors in the spatiotemporal domain. However, the performance of this technique is fundamentally limited by the system's space-time bandwidth, leading to a trade-off between accuracy and speed. This study presents a novel, to the best of our knowledge, approach integrating event-driven neuromorphic imaging with digital holography for dynamic particle imaging. Through comprehensive analysis of synthetic and real-world experiments, we demonstrate that the proposed method significantly improves data efficiency and hologram contrast by over tenfold. This technique, termed "neuromorphic holography," offers promising applications for efficient dynamic particle imaging.