A Real-Time DDS-Based Chest X-Ray Decision Support System for Resource-Constrained Clinics

TL;DR

This work tackles the need for real-time chest X-ray diagnostic support in resource-constrained clinics by combining a fine-tuned ResNet50 classifier with FastDDS middleware in a two-node IoT-style architecture. A doctor node on a Raspberry Pi publishes X-ray images to an inference node on a laptop, which runs the four-class classifier (Covid, Lung Opacity, Normal, Viral) and returns results via a low-latency publish–subscribe pipeline. The system achieves 88.61% accuracy, 88.76% precision, and 88.49% recall, with 65 ms end-to-end latency and 3.2 KB/s throughput, demonstrating the practicality of DDS-based real-time decision support in bandwidth-constrained environments. This work highlights how DDS middleware can enable reliable, scalable, and low-latency AI-assisted clinical workflows for remote healthcare applications.

Abstract

Internet of Things (IoT)-based healthcare systems offer significant potential for improving healthcare delivery in humanitarian and resource-constrained environments, providing essential services to underserved populations in remote areas. However, limited network infrastructure in such regions makes reliable communication challenging for traditional IoT systems. This paper presents a real-time chest X-ray decision support system designed for hospitals in remote locations. The proposed system integrates a fine-tuned ResNet50 deep learning model for disease classification with Fast DDS real-time middleware to ensure reliable and low-latency communication between healthcare practitioners and the inference system. Experimental results show that the model achieves an accuracy of 88.61%, precision of 88.76%, and recall of 88.49%. The system attains an average throughput of 3.2 KB/s and an average latency of 65 ms, demonstrating its suitability for deployment in bandwidth-constrained environments. These results highlight the effectiveness of DDS-based middleware in enabling real-time medical decision support for remote healthcare applications.

Figures (6)

• Figure 1: A Typical Middleware
• Figure 2: The proposed system.
• Figure 3: Deep Learning Model for Inference Node
• Figure 4: Performance of ResNet50
• Figure 5: Throughput profile