Abstract: X-ray detectors are one of the critical components in X-ray imaging systems, playing a decisive role in imaging quality.However, with the miniaturization of detector pixel dimensions and exponential expansion of array scale, conventional readout electronics face challenges including limited data transmission bandwidth, surging power consumption, and cumulative signal delays.The deployment of artificial intelligence at the detector front-end has become an inevitable trend.Based on this, systematically reviews the latest advancements in sensing-computing convergence technology for X-ray imaging detectors.Firstly, analyzes the technological breakthroughs in architecture optimization and energy efficiency enhancement of photon-counting readout application-specific integrated circuit (ASIC) .Secondly, explores the technical pathway of in-sensor computing technology for real-time signal processing at the front-end through analog neural networks.Finally, from the perspective of algorithm-circuit co-design, it evaluates innovative practices of computing-in-memory technology in overcoming von Neumann architecture bottlenecks, providing both theoretical foundation and technical roadmap for constructing high-efficiency, low-latency intelligent X-ray detection systems.