Application of Data Cloud Platform in Perioperative Risk Monitoring of Elderly Patients

As the population ages, the clinical diagnosis and treatment of elderly patients is gradually becoming a research hotspot. Especially to the elderly patients, due to their weak physical foundation and accompanied with chronic diseases commonly, the elderly patients have a higher risk of surgery in surgical operation. The assessment for risk of clinical surgery common used in clinic is realized by scale menthod, which is not designed for the elderly population and committed by manual operation, so is difficult to operate and has low analysis efficiency. It is unable to realize continuous risk monitoring in combination with the real-time changes of patients' conditions. The digital cloud technology has been widely used in medical diagnosis and scientific research, which can realize the functions of automatic data extraction, integration, data mining, data monitoring and early warning, and improve the analysis efficiency. In paper, we construct the perioperative monitoring data cloud platform for elderly patients and the data channel with the functions of automatic extraction, transformation and loading of perioperative data; the risk related features are extracted by stepwise regression method, and the perioperative real-time risk assessment model is established by online data mining tools. The iterative training can be realized with the business system data. In test, the accuracy and recall of the perioperative risk prediction model by the system are better than POSSUM evaluation method; the cost of single assessment was only 0.325s, which was faster than that of the control group. Compared with the traditional evaluation methods, the platform has obvious advantages in prediction accuracy, running speed, etc., and provides a broad prospect in the aspects of clinical application, such as perioperative monitoring of elderly patients.