To improve operational efficiency within the healthcare sector, this study explores the development of a medical information diagnostic platform leveraging swarm intelligence and evolutionary algorithms. The paper first reviews the current state of medical diagnostic platforms that integrate Chat Generative Pre-trained Transformer (ChatGPT) models with Internet of Things (IoT) technology. It then provides a detailed analysis of the strengths and limitations of employing swarm and evolutionary algorithms in such platforms. Optimization of the swarm algorithm is achieved using reverse learning and Gaussian functions. The validity and effectiveness of this optimized approach are demonstrated through horizontal comparative experiments. Results indicate that the enhanced model performs well across minimum, average, and maximum algorithm fitness metrics. Furthermore, preprocessing data on a 10 × 10 server arrangement further improves algorithm fitness, with the optimized algorithm achieving a minimum fitness value of 3.56—a 3% improvement compared to unsorted data. Stability tests reveal that the optimized algorithm maintains superior stability, which is further strengthened by applying sorting techniques. Overall, this study provides new insights into medical information diagnostics and offers practical technical support for applications in medical data processing.
