Analyse Cardio-respiratory Patterns in Order to Early Detect the Clinical Complications in the Hemodialysis Patients

Overview

The incidence and prevalence of dialysis in Taiwan remains higher as compared to past several years. The number of dialysis reported about 6-7 million people each year and the 2014 full-year cost estimation was about 330 million NTDs. According to age-standardized population statistics in 1995, the number of dialysis per million population in 2006 year were 372.2 people, 381.9 people in 2012 year and the annual growth rate of 0.43 percent recorded. EarlySense system is certified by TFDA, FDA and many other countries, which consists of mainly two parts. One for the sensing element, which placed under the mattress and the other one is AC-powered display panel bed. This system has the ability to monitor the patients physiological functions just be in contact via bed. When the patient laid on bed, the sensor which is placed under the bed would detects heart rate, breathing rate and activity level as well as other physiological signals. The system is also able to distinguish between patients in bed and out of bed during the operation to save the patient's records and information (including heart rate, respiratory rate and level of activity, and may render trends). The system is also have capability for real-time data transmission information which includes a warning to the nurses' station or other monitoring center of the screen by providing real-time information to nurse. In this research project, we will use EarlySense equipment for continuous monitoring the dialysis patient's physiological data along with clinical data such as A. acute complications such as rapid changes in blood pressure, respiratory rate, movements of patients and Nausea or vomiting, etc phenomenon's) B. Lab examination data C. Status of patient whether patients hospitalized or not and patient death occurred or not. The study is expected to have a pilot study for more than three months' duration. Through this EarlySense continuous monitoring and gather the data, we will analyse to develop a prediction model and confirmed with indicators. Evidences from these analytics would help to propose ways to improve it and implement Positive predictive validity models.

Conditions