Build predictive models on top of automatically extracted real-time information from unstructured documents. Enable data-driven decisions with accuracy and speed not previously available.
01
Extract
Extract in real time information from unstructured text
02
Build
Build a predictive model using the extracted information
03
Apply
Get timely and accurate support in clinical & operational decisions
Case Study #1: Predict Hospital Bed Demand by Real Time Analysis of Clinical Notes
Key factors that influence a patient’s flow (How likely they are to admitted? For how long? For what?):
Volume of arrivals
Outpatient
Referrals
Emergency Room
Operation Room
Admission specialty
Oncology
Hip-replacement
Renal Disease
Cardiology, …
Timing of arrival
Hour of the day
Day of the week
Holidays
Seasonal variables
Flu season
Natural disasters
Patient's length of stay
per unit (ICU, CVICU, …)
Nurse staffing levels & skill mix:
Certified Nurses
Licensed N.P.’s
Unlicensed Staff
Unique certifications
John Snow Labs enabled real-time decision-making and strategic planning, by predicting:
Demos
