2. Analytics uses Big Data
?
Big
Next gen-seq,
iPOP,
Data Size
Small
Claims,
EMR, Clinical
notes
Small Large
Number of samples
3. Think about samples vs.
variables
Variables Drugs
Disease
s
Devices Procedures
Patients
Millions
??
Genome Expression Metabolome
Millions About 100,000
(Billions?)
4. Themes in Healthcare Analytics
Precision of Quality, cost and
diagnosis/treatment operations
A lot of research and disruptive Real time monitoring of acuity,
technologies are emerging to staffing, resources & outcomes
enable data-driven decisions. for operational transparency.
• Personalization is the mantra • System efficiency is the mantra
• Total awareness about patients
• Frequent fliers to the ER .. how
to provide the care they need
to identify them and intervene?
proactively. Using behavior
profile, social profile, contextual / • End of life care is 50% of the
spatial orientation spend .. how to form a 'circle of
• Patients come in seeking care' at home?
particular option .. how do you • What fraction of care is
counter that? 'defensive care'?
Both of these are data problems in some fashion and can be tackled with either
lots of data on one patient, or some data on lots of people.
Notes de l'éditeur
A dumb algorithm with lots of data beats a clever one with modest amounts of it