A Construction Method of an Intelligent Treatment Guidance System for ICU Ward
Fujie Deng; Feng Wang
National Foundation Research Laboratory of Fault Prevention and Hazardous Chemicals Production System, Beijing University of Chemical Technology, Beijing, China
Objectives: With the emergence of new viruses, new cases, new patients with different new constitutions, etc., it becomes more and more difficult to diagnose patients. With the advent of new doctors, new prescription drugs, new treatment devices, etc., patients' treatment options are increasingly diverse. Therefore, it is helpful and important to provide prescriptive guidance to attending physicians and nurses. Especially in the ICU wards, the patients, who are receiving treatment with a ventilator, generally have serious illnesses. The patients are staying in the stage of weaker resistance and at higher risk of complications and serious infections. It is necessary to develop an intelligent treatment guidance system based on big data such as drug structure components, medication instructions, prescriptions, monitoring data, and the disease progresses so that doctors can be provided with reference and intelligent guidance during treatment.
Methods: The paper puts forward a construction method of an intelligent treatment guidance system for the ICU ward. The system includes all kinds of drugs, patients' illnesses, clinical symptoms, equipment usage guidance, treatment plans, medication contraindications, etc. The information on the composition, shape, specification, dosage, adverse reactions, contraindications, drug interactions, pharmacology, and toxicology, etc., for most of the drugs, has been recorded in the database. The information of equipment names, operation procedures, usage methods, maintenance records for the treatment devices has been recorded in the database. Daily treatment plans are recorded in detail in a database. By using the method of machine learning, such as association ****ysis and Bayesian calculation method, the information of drugs, patients' illnesses, equipment usage guidance, treatment plans, medication contraindications, etc., can be matched. The system can provide doctors with the basis for comprehensive treatment and diagnosis. At the same time, the system can realize information queries and statistics, and recommend treatment and medication plans for doctors through intelligent data matching. It also can provide information on corresponding plans and times for usage drugs, etc.
Conclusions: According to the application of the method, it can be concluded that the system can to some extent guide doctors to make comprehensive judgments of diagnosis and treatment by referring to various types of information, reducing the time and energy of doctors in learning and practice, and avoiding misoperation and doctor-patient disputes.
Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grant No. 51775029) & the Chinese universities' scientific fund (Grants No. PYBZ1809).
Correspondence Author Feng Wang, National Foundation Research Laboratory of Fault Prevention and Hazardous Chemicals Production System, Beijing University of Chemical Technology, Beijing, China (e-mail: wangfeng991@163.com).