Intelligent Clinical Response Orchestration via Adaptive AI-Driven Inference for Emergency Service Coordination

Abstract

The growing prevalence of chronic conditions such as hypertension and diabetes has intensified the need for intelligent healthcare solutions capable of supporting early-stage detection and informed medical decision-making. Continuous monitoring of multiple patient health indicators is essential, yet traditional diagnostic practices largely depend on manual assessment and basic statistical analysis, limiting their effectiveness in large-scale and data-intensive environments. These conventional systems often fail to capture complex interdependencies within medical data, struggle with imbalanced datasets, and lack the capability to deliver timely predictions in distributed settings. To overcome these challenges, this study presents a real-time clinical decision support framework driven by Artificial Intelligence (AI) and structured on a dual client–server model. The backend server is responsible for data preprocessing, model training, and prediction using Machine Learning (ML) techniques, including Complement Naive Bayes (CNB), Multinomial Naive Bayes (MNB), Perceptron, and a Tao Tree Classifier (TTC). Data preprocessing is performed using Label Encoding and K-Means Synthetic Minority Oversampling Technique (KMeans-SMOTE) to address categorical data handling and class imbalance issues. Communication between system components is facilitated through a Flask-based Application Programming Interface (API) utilizing Hypertext Transfer Protocol (HTTP), ensuring efficient data exchange. The client interface enables users to upload medical datasets, which are remotely processed to generate predictions for blood pressure levels and diabetic conditions. Additionally, Lightning Memory-Mapped Database (LMDB) is integrated to provide secure, highspeed data storage and retrieval. The proposed system enhances predictive accuracy, ensures real-time accessibility, and supports scalable, reliable healthcare delivery.