PRESAGE® by MOKAAL p.c.c.
REAL TIME PREDICTION MODEL ON REMOTE COLLECTED VITAL SIGNS
A system utilizes time series of metric data to produce metadata through which is attempting to predict in real time health threatening events. Streams of time series data are processed to generate a set of independent metadata sets. Advanced machine learning and statistical techniques are used to automatically find anomalies and outliers from the combination of the independent metadata sets by revealing latent and hidden patterns in the data streams. The trends of each data set may also be analyzed and the trends for each characteristic may be learned. The system can automatically detect latent and hidden patterns in metadata including weekly, daily, holiday and other application specific patterns. The present technology applies on any platform of wearable device that monitors vital signs on a BAN (Body Area Network) complex. The wearable device can perform real-time measurement of a number of physiological and environmental parameters including heart rate, pulse oximetry, respiration, movement, environmental particulate matter, moisture, temperature (e.g., ambient air and body temperatures) and geospatial location, by using related algorithms and software that are tied to a portable electronic device for readout. The system may establish a physiological baseline for a patient by measuring the above parameters during a healthy state. Collected data can be wirelessly transmitted to a portable electronic device or monitoring and feedback platform where software will analyze the data and make assessments of the device wearer’s health based upon the wearer’s baseline and by using these data in conjunction with the bearer’s EHR (Electronic Health Record), possibly predict upcoming health threatening events for this person.
