Building a Model of a Patient State based on Machine Learning Methods in a Fog Computing Environment

The paper discusses a method for constructing a model that reflects the patient’s state based on image processing using machine learning in a fog computing environment. A review of existing methods for constructing models of patient states is carried out. The application of machine learning methods for diagnosing the conditions of patients under the supervision of medical personnel is considered. The problem of providing timely assistance to patients, increasing the organization and speed of medical personnel has been identified. A solution is proposed that provides the construction of a model of the patient’s state based on a factorial model of a person’s face. Images of patients’ faces are converted into feature vectors. According to these vectors, the patient’s condition is classified using machine learning methods. The rate of care for hospital patients increased by 24.4%, the accuracy of primary diagnosis by 16.37%, that indicates the effectiveness of the proposed method, which can be used to diagnose a state of a patient not only in hospitals, but also at home (outpatient), and in polyclinic.