Scalable Data and AI Architectures for Intelligent Healthcare and Vision-Driven Systems

Abstract

The rapid expansion of digital healthcare ecosystems has intensified the need for scalable data and artificial intelligence (AI) architectures capable of supporting intelligent, vision-driven clinical systems. This study investigates how architectural design influences data processing efficiency, vision-based AI performance, infrastructure utilization, and governance readiness in modern healthcare environments. A mixed-method, system-centric framework was employed to evaluate centralized, cloud-native, hybrid edge–cloud, and distributed edge architectures using heterogeneous healthcare data and vision-driven AI workloads. Results demonstrate that hybrid edge–cloud architectures consistently outperform centralized designs, achieving higher data ingestion throughput, lower inference latency, improved model accuracy, and superior resource efficiency under increasing workload intensity. Furthermore, scalable architectures exhibit enhanced reliability, governance compliance, and operational sustainability, which are critical for safety-sensitive healthcare applications. The findings highlight that intelligent healthcare transformation depends not only on advanced AI models but also on robust, scalable, and governance-aware architectural foundations capable of sustaining real-world clinical intelligence at scale.