Fusion of Efficient Transformer-CNN Hybrid Network (ETCHNet) and Adaptive Deep Clustering Optimization Network (ADCONet) for Advanced IoMT-Based Surgical Monitoring With Robotics and Artificial Intelligence

Abstract

Traditional health care systems face the challenge of real-time data processing and decision-making, particularly in the surgery domain. Health gadgets based on advanced technologies like IoMT, blockchain, and AI-based systems are used for improving surgical precision and real-time monitoring. While these systems aim to enhance surgical monitoring accuracy, scalability, and efficiency, handling high-dimensional data and real-time analysis still present challenges. This study focuses on the development of a robust and scalable enhanced IoMT-based surgical monitoring system that improves precision. The framework enhances feature extraction, clustering, and decision-making using real-time feedback and blockchain for secure data storage. The proposed framework integrates ETCHNet and ADCONet with robotic automation. The first algorithm, ADCONet, clusters dynamically and builds decisions based on optimised approaches and ETCHNet extracts hybrid features using transformers and CNNs. Robotic automation provides real-time feedback and a fluid surgical interaction. The proposed method achieved 96% classification accuracy and 95% clustering accuracy, which exceeded the existing methods. It was so effective that the method detected 94.5% of anomalies and prevented 97% of collisions. Even more impressively, it did so with an inference latency of only 120 milliseconds. As a result, the solution is still ideal upon surgical real time monitoring where split second decisions are needed. These impressive results clearly prove that the system improves healthcare accuracy and effectiveness.