Computational Health Care 1. Motivation “We have to do our own research” – effective health care is ethnic, cultural and environmental dependent NUS has.

1

Computational Health Care

2

Motivation• “We have to do our own research” – effective

health care is ethnic, cultural and environmental dependent

• NUS has core technology

3

Knowledge Structuring Layer

Secured Layer

Media Analytics Layer

Application Layer

Sense-Making Cloud

Predictive Analytics

Prescriptive Analytics

Drug / Food Safety

OntologyHealthcare

Professionals, patients and

caregivers

Hospital Operating

DataDevice

Social Media

InputApplication

Preventive Healthcare

Public Layer

4

Sense Making Cloud

5

A “Learning” Sense Making Cloud – technology and methodology evolve!

Sense Making Cloud

LEARNS !

6

• Learning depends on data, and data is ethnic, cultural and environmental dependent

• Leverage on NUS strength in knowledge learning and discovery

7

Secured Layer

8

Data and Cyber Security

• Medical Record• confidentiality• data integrity• holistic and timely

• Provide a secure and trusted environment for safe housekeeping of patients’ data – prevent hacking and malicious software

9

Knowledge Structuring Layer

10

Nature of Data• Unstructured• in natural language• knowledge structuring for meaningful exploitation

• non-textual• visualization

11

• Semi-Structured - Quantified Self

12

• uncertain & inconsistent

• doctors, nurses, care givers, patients may give conflicting data or interpretation of data over the years due to changes in context (eg. situation changes, knowledge enhances)

• Need to do analytics to form a logical picture out of these unstructured, inconsistent data

13

• Need clever means of structuring knowledge and reasoning from uncertain/inconsistent data• thru natural language processing• thru crowd sourcing

• make the unstructured and uncertain/inconsistent data exploitable

14

Media Analytics Layer

15

• Predictive Analytics (Clinical)• hospital bed• prediction to reduce re-admission• medical, social conditions

• prediction of which patient needs more urgent attention (eg. ICU, Emergency Deparment)

Predictive Analytics

16

• Predictive Analytics (Resources)• ambulance• prediction and optimization of ambulance

placement• stream data from ambulance to hospital for

timely action upon arrival at hospital

17

• Predictive Analytics (Wellness)

• wellness/mobile device

• behavioral

• drug and supplements

• food

18

• Diagnosis and treatment• prognosis prediction• develop hypotheses• eg. for uncommon diseases

• develop better diagnostic plan• also reduce unnecessary dosage (imaging and

drugs)• knowledge management

Prescriptive Analytics

19

• Drug Safety

• interactions • drug-drug• drug-food

• allergies

Drug and Food Safety

20

• Food Safety• Pesticide Screening• mass screening of peticide using advanced imaging

• Genetically modified food• Organic food• develop technology to certify organic food

• Meat safety

21

• Wellness: Aging and Handicap Assistive Technologies

Preventive Health Care

21

22

• Wellness: Sensors & Devices• Mobile Phones as sensing device

• Other wearable gadgets• transmitting health data including

vital signs

• Predictive and Prescriptive Analytics

23

• Wellness: Music • Music Enhanced Preventive Exercise• Music Therapy

• Wellness: Evidence-based design computing & human-centered computing

24

Pioneering a Computational Healthcare Industry• Inspired by our success in water technology

• Pioneer a computational healthcare industry, put Singapore on global map of health care with Asian context

25

Implementation Plan

• Medical is about Human

• Need to build confidence before adoption• practice on archival data• training of doctors• do a pilot study

• Leverage on crowd sourcing for general public health

26

Thank You !