Download IBM Content and Predictive Analytics for Healthcare

IBM Content and Predictive Analytics for Healthcare
The first Watson Ready solution
© 2011 IBM Corporation
Working together
IBM Content and Predictive Analytics for Healthcare and IBM Watson for Healthcare
Evidence Based Learned Knowledge
IBM Content and
Predictive Analytics for
Healthcare
Books, clinical guidelines,
web resources, journals
and other healthcare
authoritative resources
Leverage learned knowledge
with QA-style interactions for
clinical applications such as
diagnosis
Past, present and future
analysis compliments
Watson – with focus on
customer data for clinical
and operational outcomes
Ready for Watson
Clinical
Outcomes
2
Operational
Outcomes
IBM Watson for
Healthcare
WellPoint and IBM Announce
Agreement to Put Watson to
Work in Health Care
September 12, 2011
“… clinical best practices to help a
physician advance a diagnosis and
guide a course of treatment"
© 2011 IBM Corporation
IBM Content and Predictive Analytics … Ready for Watson
Complements IBM Watson to analyze and visualize past,
present and future scenarios in context
Question What is Known
Complement ICPA with Watson for Healthcare to
get real time, confidence based answers with
evidence based learning
3
Analyze and Visualize the Past
See the Present
Predict the Future
Understand trends,
patterns, deviations, anomalies, context
and more in large corpuses of historical
clinical and operational information to
reveal new insights
Analyze and extract text from
in-process documents or other
information to find structured
data errors … feed the results
to other cases and systems
Use predictive models and
scoring to make more
informed decisions through
predictive and future
scenario modeling
© 2011 IBM Corporation
IBM at 100: Innovation That Matters in Healthcare
100-year track record of innovation and progress
TAKMI Bringing Order to Unstructured Data
• In 1997, IBM researchers at the company’s Tokyo
Research Laboratory pioneered a prototype for a
powerful new tool capable of analyzing text. The
system, known as TAKMI (for Text Analysis and
Knowledge Mining) was a watershed development:
for the first time, researchers could efficiently
capture and utilize the wealth of buried knowledge
residing in enormous databases of text
• In 2009, TAKMI was commercialized as IBM Content
Analytics
4
© 2011 IBM Corporation
What is UIMA and NLP?
 Natural Language Processing (NLP) is the cornerstone to translate interactions
between computers and human (natural) languages
– Watson uses IBM Content Analytics to perform critical NLP functions
 Unstructured Information Management Architecture (UIMA) is an open
framework for processing text and building analytic solutions
– Many IBM products leverage UIMA text analytics processing including
Watson and IBM Content and Predictive Analytics for Healthcare
– Now an OASIS open source standard
– Apache UIMA is sponsored project by the Apache Software Foundation
5
© 2011 IBM Corporation
Medical Transcription Discharge Summary Sample # 2:
Cardiology Consultation Transcribed Medical Transcription Sample Reports
REFERRING PHYSICIAN: John Doe, MD
CONSULTING PHYSICIAN: Jane Doe, MD
HISTORY OF PRESENT ILLNESS: This (XX)-year-old lady is seen in consultation for Dr. John
Doe. She has been under consideration for ventral hernia repair and has a background of aortic
DATE OF DISCHARGE: MM/DD/YYYY
valve replacement and known coronary artery disease. The patient was admitted with complaints
of abdominal pain, anorexia, and vomiting. She underwent a CT scan of the abdomen and pelvis
ADMITTING DIAGNOSIS: Syncope.
and this showed the ventral hernia involving the transverse colon, but without strangulation. There
was an atrophic right kidney. She had bilateral renal cysts. The hepatic flexure wall was thickened.
There was sigmoid diverticulosis without diverticulitis. It has been recommended to her that she
CHIEF COMPLAINT: Vertigo or dizziness.
undergo repair of the ventral hernia. For this reason, cardiology consult is obtained to assess
whether she can be cared from the cardiac standpoint.
HISTORY OF PRESENT ILLNESS: This is an (XX)-year-old male with a past medical history of coronary artery disease, CABG done a few
PAST CARDIAC HISTORY: Bypass surgery. She underwent echocardiography and cardiac
years ago, atrial fibrillation, peripheral arterial disease, peripheral neuropathy, recently retired one year ago secondary to leg pain. The
catheterization prior to the operation. Echocardiography showed an ejection fraction of 50%. There
patient came to the ER for an episode of vertigo while reaching for some books. The patient was able to reach the books, to support
was marked left ventricular hypertrophy with septal wall 1.60 cm and posterior wall 1.55 cm.
self, but did not have any syncope. No nausea or vomiting. No chest pain. No shortness of breath. Came to ER and had a CT head,
Coronary arteriography showed 90% stenosis in the anterior descending artery, situated distally
just before the apex of the left ventricle. Only mild to moderate narrowing was seen elsewhere in
which was within normal limits. The impression was atrophy with old ischemic changes but no acute intracranial findings. No focal
the coronary circulation.
weakness, headache, vision changes or speech changes. The patient has had similar episodes since one year. Peripheral neuropathy
CORONARY RISK FACTORS: Her father had an irregular heartbeat and her brother had a fatal
since one year and not relieved with multiple medications. The patient also complains of weight loss of 25 pounds in the last 6
heart attack. She herself has had high blood pressure for 20 years. She has elevated cholesterol
Echocardiogram Sample Report:
months. No colonoscopy done. Recent history of hematochezia but believes it was secondary to proctitis and secondary to decreased
and takes Lipitor. She has had diabetes for 20 years. She is not a cigarette smoker. She does little
appetite. No nausea, vomiting, no abdominal pain.
physical exercise.
DATE OF STUDY: MM/DD/YYYY
REVIEW OF SYMPTOMS: CARDIOVASCULAR AND RESPIRATORY: She has no chest pain. She
sometimes becomes short of breath if she walks too far. No cough. She has occasional swelling of
her feet. Occasionally, she gets mildly lightheaded. Has not lost consciousness. She tends to be
PROCEDURES
PERFORMED:
The patient had
x-ray, which showed cardiomegaly with atherosclerotic heart disease, pleural
DATE
OF INTERPRETATION
OFa chest
STUDY:
aware of her heartbeat when she is tired. She has no history of heart murmur or rheumatic fever.
thickening and small pleural effusion, a left costophrenic angle which has not changed when compared to prior examination, COPD
GASTROINTESTINAL: Recent GI symptoms as noted above, but she does not usually have such
pattern. TheEchocardiogram
patient also had awas
headobtained
CT which for
showed
atrophy with
old ventricular
ischemic changes. No acute intracranial findings.
problems. She has had no hematemesis. She has no history of ulcer or jaundice. She sometimes
assessment
of left
has loose stools. No constipation and no blood in the stool. GENITOURINARY: She tends to have
function. The patient has been admitted with diagnosis of
urinary frequency. She gets up once at night to pass urine. No dysuria, incontinence. She has had
Cardiology Consultation Transcribed Medical Transcription Sample Reports
syncope.
Overall,
the
study
was
suboptimal
due
to
poor
sonic
window.
previous urinary infections. No stones noted. NEUROLOGIC: She has occasional headaches. No
CONSULTS OBTAINED: A rehab consult was done.
DATE OF CONSULTATION: MM/DD/YYYY
seizures. No trouble with vision, hearing, or speech. No limb weakness. MUSCULOSKELETAL: She
REFERRING PHYSICIAN: John Doe, MD
tends to have joint and muscle pains and has a history of gout. HEMATOLOGIC: No anemia,
CONSULTING PHYSICIAN: Jane Doe, MD
FINDINGS:
abnormal bleeding, or previous blood transfusion. GYNECOLOGIC: No gynecologic or breast
REASON FOR CONSULTATION: Surgical evaluation for coronary artery disease.
PAST MEDICAL/SURGICAL HISTORY: Positive for atrial fibrillation.
The
patient
had
AVR
6
years
ago.
Peripheral
arterial
disease
with
HISTORY OF PRESENT ILLNESS: The patient is a (XX)-year-old female who has a known history
of coronary artery
problems.
disease. She
underwent
previous
PTCA and stenting procedures in December and most recently
in August.
Since that
time,
hypertension,
peripheral
neuropathy,
atherosclerosis,
hemorrhoids,
proctitis,
CABG,
and
cholecystectomy.
PAST
MEDICAL
HISTORY:
She has had shoulder and hand injuries and has had carpal tunnel
1. Aortic root appears normal.
she has been relatively stable with medical management. However, in the past several weeks, she started to notice some
surgery.
has been diabetic and has been on insulin. She has chronic renal insufficiency with
exertional dyspnea
with chestispain. For the most part, the pain subsides with rest. For this reason,
she She
was re-evaluated
2. Left atrium is mildly dilated. No gross intraluminal
pathology
creatinine
around
2.2.
She
has
had hypothyroidism. She has had morbid obesity. She has chronic
with a cardiac catheterization. This demonstrated 3-vessel coronary artery disease with a 70% lesion to the right coronary
recognized, although subtle abnormalitiesartery;
could
not
excluded.
Right
this
wasbe
a proximal
left main had a 70% stenosis. The circumflex also hadobstructive
a 99% stenosis.
Overall
leftand uses BiPAP. She has had hysterectomy and oophorectomy in the past.
sleep
apnea
FAMILY HISTORY: Positive for atherosclerosis, hypertension,
autoimmune
diseaseslesion.
in the The
family.
ventricular function was mildly reduced with an ejection fraction of about 45%. The left ventriculogram
didas
note
someabove.
apical
Otherwise
noted
atrium is of normal dimension.
hypokinesis. In view of these findings, surgical consultation was requested and the patient was
seen and evaluatedPrior
by Dr.to hospital, she was taking glipizide XL 2.5 mg daily, metoprolol 50 mg
MEDICATIONS:
3. There is echo dropout of the interatrialDoe.
septum. Atrial septal defects
b.i.d., Cipro 250 mg b.i.d., atorvastatin 40 mg daily, Synthroid 75 mcg daily, aspirin 81 mg daily,
PAST MEDICAL HISTORY:
could Never
not besmoked.
excluded.
SOCIAL HISTORY:
Alcohol socially. No drugs.
1. Coronary artery disease as described above with previous PTCA and stenting procedures. and Lantus 36 units daily. Currently, she is taking Lipitor 40 mg daily, Lantus 10 units at bedtime,
2. Dyslipidemia.
Synthroid 75 mcg daily, metoprolol 50 mg b.i.d., and Zosyn 2.25 grams q.6h.
4. Right and left ventricles are normal in internal
dimension. Overall left
3. Hypertension.
ventricular systolic function appears to be4.normal.
Eyeball
ejection for cancer with followup radiation therapy to the chest. SOCIAL HISTORY: She does not drink alcohol.
Status post
breast lumpectomy
PHYSICAL EXAMINATION:
ALLERGIES: fraction
NO KNOWN
DRUG ALLERGIES.
ALLERGIES:
None. wall motion
is around
55%. Again, due to poor
sonic window,
GENERAL APPEARANCE: She is not currently dyspneic, in no distress. She is alert, oriented, and
MEDICATIONS: Aspirin 81 mg daily, Plavix 75 mg daily, Altace 2.5 mg daily, metoprolol 50 mg b.i.d. and Lipitor 10 mg
pleasant.
abnormalities in the distribution of lateralq.h.s.
and apical wall could not be
HEENT:
are normal and react normally. No icterus. Mucous membranes well colored.
SOCIAL HISTORY: She quit smoking approximately 8 months ago. Prior to that time, she had
about aPupils
35- to 40-pack-year
excluded.
REVIEW OF SYMPTOMS: Weight loss of 25 pounds within the
last She
6 months,
shortness
of breath, constipation, bleeding from
history.
does not
abuse alcohol.
NECK: Supple. No lymphadenopathy. Jugular venous pressure not elevated. Carotids equal.
FAMILY
MEDICAL
HISTORY:
Mother
died
prematurely
of
breast
cancer.
Her
father
died
prematurely
of
gastric
5.
Aortic
valve
is
sclerotic
with
normal
excursion.
Color
flow
imaging
and
hemorrhoids, increased frequency of urination, muscle aches, dizziness and faintness, focal weakness and numbness in both legs, knees
HEART: The heart rate is 82 per minute and regular and the blood pressure 132/78. The cardiac
carcinoma.
impulseNohas
a normal
and feet. Doppler study demonstrates trace aortic regurgitation.
REVIEW OF SYMPTOMS: There is no history of any CVAs, TIAs or seizures. No chronic headaches.
asthma,
TB, quality. There is a grade 3/6 ejection systolic murmur heard medial to the
and
athas
thenoaortic area, with well heard radiation to the neck vessels.
hemoptysis
or productive
cough.
There
is no congenital heart abnormality or rheumatic fever apex
history.
She
6. Mitral valve leaflets are also sclerotic with
normal
excursion.
Color
flow
clear to percussion and auscultation. Normal respiratory effort.
palpitations. She notes no nausea, vomiting, constipation, diarrhea, but immediately prior to CHEST:
admission,Chest
she didisdevelop
imaging and Doppler study demonstratessome
tracediffuse
to mild
degree
of mitral
abdominal
discomfort.
She says that since then, this has resolved. No diabetes ABDOMEN:
or thyroid problem.
Therenontender.
is
Soft and
The presence of a large ventral hernia is noted.
PHYSICAL EXAMINATION:
VITAL
SIGNS:
Blood
pressure
188/74,
pulse
62,
respirations
18
and
saturation
of
98%
on
room
air.
General
no depression
or psychiatric
problems.
There
is no musculoskeletal
disorders
or history of gout.
There
are
no
hematologic
EXTREMITIES: There is no edema. Posterior tibial pulses were felt bilaterally, but I did not feel the
regurgitation.
problemsConjunctivae
or blood dyscrasias.
No bleeding
tendencies.
sheNo
had a history of breast cancer and underwent
Appearance: The patient is a pleasant man, comfortable. HEENT:
are normal.
PERRLA.
EOMI.Again,
NECK:
pedis.
lumpectomyPulmonic
procedures valve
for this is
with
followup radiation therapy. She has been followed in thedorsalis
past 10 years
and
7. Tricuspid valve is delicate and opens normally.
not
masses. Trachea is central. No thyromegaly. LUNGS: Clear
to auscultation
andnopercussion
Irregular
SKIN:
No rash
or significant
lesions are noted.
mammography
shows
evidence of bilaterally.
any recurrentHEART:
problems.
There is no recent fevers, malaise,
changes
in appetite
or
clearly seen.
No evidence
pericardial effusion.
LABORATORY AND DIAGNOSTIC DATA: Electrolytes are normal. BUN and creatinine 18/2.2.
changes
in weight.
rhythm. ABDOMEN:
Soft, nontender,
andof
nondistended.
Bowel
sounds
are positive. GENITOURINARY: Prostate is hypertrophic with
PHYSICAL EXAMINATION: Her blood pressure is 120/70, pulse is 80. She is in a sinus rhythm
on the
EKG150. White count is 7.6, hemoglobin 11.7 with hematocrit 34.9, platelets 187,000.
Blood
sugar
smooth margin. EXTREMITIES: Upper and lower limbs bilaterally
normal. SKIN: Normal. NEUROLOGIC: Cranial nerves are grossly
monitor. Respirations are 18 and unlabored. Temperature is 98.2 degrees Fahrenheit. She weighs
pounds,
she Hemoglobin
is 5 feet
LFTs 160
were
normal.
A1c 7.7. TSH 1.82. Troponin I was normal on three occasions.
CONCLUSIONS:
within normal
limits. No nystagmus. DTRs are normal. Good
sensation.
The patient
awake, and pleasant
orientedfemale
x3. Mild
4 inches.
In general,
this was is
analert,
elderly-appearing,
who confusion.
currently is not in acute distress. Skin color and
Chest
x-ray
showed
turgor are good. Pupils were equal and reactive to light. Conjunctivae clear. Throat is benign.
Mucosa
was moist
and an enlarged heart with postoperative changes, but no evidence of acute
pathology.
EKG
shows
probable
left atrial enlargement. Low voltage QRS, probable inferior wall
noncyanotic. Neck veins not distended at 90 degrees. Carotids had 2+ upstrokes bilaterally without bruits. No
lymphadenopathy was appreciated. Chest had a normal AP diameter. The lungs were clear inmyocardial
the apices and
bases, no and anterior wall infarction, age undetermined.
infarction
1. Poor quality study.
wheezing
or egophony
heart
had a normal
S1, MCV
S2. No
murmurs, clicks or gallops.
The abdomen was soft,
LABORATORY
AND
RADIOLOGICAL
hemoglobin
13.4,appreciated.
hematocritThe
39.8,
platelets
207,000,
91.6,
ASSESSMENT:
2. DATA
Eyeball
ejection
fractionRESULTS:
is 55%. WBC 8.6,
nontender, nondistended. Good bowel sounds present. No hepatosplenomegaly was appreciated.
No pulsatile
were
1. Aortic
valvemasses
replacement
with bioprosthetic valve. Residual systolic murmur.
neutrophil percentage of 72.6%. Sodium 133, potassium 4.7,
chloride 104. Blood urea nitrogen of 18 and creatinine of 1.1. PT 17.4, INR
felt. No abdominal bruits were heard. Her pulses are 2+ and equal bilaterally in the upper and lower extremities. No
3.
Trace
to
mild
degree
of
mitral
regurgitation.
2. Arteriosclerotic heart disease with severe stenosis in anterior descending artery, but this is
1.6, PTT 33.
clubbing is appreciated. She is oriented x3. Demonstrated a good amount of strength in the upper and lower
situated
distally
and
subtends
only a small mass of myocardium.
4. Trace aortic regurgitation.
extremities. Face was symmetrical. She had a normal gait.
3. Well
systolic function. The EKG appearance of previous myocardial
IMPRESSION: This is a (XX)-year-old female with significant multivessel coronary artery disease.
Thepreserved
patient alsoleft
has ventricular
a
left main lesion. She has undergone several PTCA and stenting procedures within the last year
to year and
half. At this
infarction
is aprobably
serious, indicating multiple other medical problems as listed above
DATE OF ADMISSION: MM/DD/YYYY
Unstructured data is messy but
filled with key medical facts
Medications, diseases,
symptoms, non-symptoms,
lab measurements, social
history, family history and
much more
in order to reduceheart
the risk
of any pleural
possible thickening
ischemia in the
The patient had a chest x-ray, which showed cardiomegaly point,
with atherosclerotic
disease,
andfuture,
small surgical
pleural myocardial revascularization is
and also documented in the chart.
recommended.
effusion, a left costophrenic angle which has not changed when
compared to prior examination, COPD pattern. The patient also had a
IBM Corporation
PLAN: We will plan to proceed with surgical myocardial revascularization. The risks and benefits
of this procedure were
RECOMMENDATIONS:
It appears that she does not wish©
to2011
proceed
with the
head CT, which showed atrophy with old ischemic changes.explained
No acute
findings.
to intracranial
the patient. All
questions pertaining to this procedure were answered.
surgery at this time, and if such surgery is not
IBM is helping to transform healthcare
Revealing clinical and operational insights in the high impact overlap between
clinical and operational – enabling low cost accountable care
IBM Content and Predictive Analytics for Healthcare
 Readmission prevention
Diagnostic assistance
 Claims management
Clinical treatment
effectiveness
Critical care intervention
Research for improved
disease management
 Fraud detection and
prevention
Clinical
Outcomes
Operational
Outcomes
 Voice of the patient
 Patient discharge and
follow-up care
 Improved patient satisfaction at lower costs
 Enhanced patient care with optimized outcomes
7
© 2011 IBM Corporation
California Pacific Medical Center
Using advanced cardiac risk models to reduce
risks and recovery times for patients
"IBM SPSS offers many time-saving features that
enable us to obtain and publish results faster… With
data analysis, we have been able to reduce heart
attack patient mortality by 8%"
Dr. Richard Shaw, Director of Cardiac Research, Division
of Cardiology, California Pacific Medical
Business Challenge
Cardiac research program needed to manage, track
and analyze vast amounts of disparate patient data
collected from departments throughout the hospital.
What’s Smart?
Implemented models that improved patients’
long-term outcomes and shorten patients’ length of
stay and reduced costs of treatment.
Smarter Business Outcomes
Cal Pacific Medical Center has been able to shorten
patients’ length of stay and reduce costs of treatment.
IBM SPSS provides the power, flexibility and ease of
use the research team needed to support multidisciplinary, data-intensive projects.
8
© 2011 IBM Corporation
BJC Healthcare and Washington University Partnership
Improving care and increasing revenue while
lowering research costs
"We anticipate this solution to be a game changer in
biomedical research and patient care … accelerate the
pace of clinical and translational research …"
Dr. Rakesh Nagarajan, MD, PhD, Associate Professor,
Department of Pathology and Immunology,
Washington University
Business Challenge
Existing Biomedical Informatics (BMI) resources
were disjointed, siloed, redundant and only available
to a few researchers - key insights not accessible,
trapped in unstructured clinical notes, diagnostic
reports, etc.
What’s Smart?
Leverage unstructured information along with
structured data by using IBM Content Analytics with
IBM InfoSphere Warehouse
Smarter Business Outcomes
Researchers now able to see new trends, patterns
and find answers in days instead of weeks or
months eliminating manual methods also enables
new grant revenue
9
© 2011 IBM Corporation
What IBM is Announcing
IBM Content and Predictive Analytics for
Healthcare (ICPA)
Helps transform healthcare clinical and operational decision
making for improved outcomes by uniquely applying multiple
analytics services to derive and act on new insights in ways not
previously possible
New IBM BAO Solution Services for ICPA
Core Industry Strategy
Solution services that enable 3 paths to solution value based
on client needs leveraging a new UIMA center of competence
• Consumable, repeatable,
scalable offerings with
solution assets built on a
common industry
framework
New IBM POWER7 Workload Optimized for ICPA
Deliver ICPA solutions faster, with higher performance and
superior economics
IBM Content and Predictive Analytics
for Healthcare is the first “Ready for
Watson” solution … to complement
and leverage IBM Watson
10
BAO = Business Analytics and Optimization
UIMA = Unstructured Information Management Architecture
• Reduces integration
costs and accelerates
time-to-solution value
© 2011 IBM Corporation
IBM Content and Predictive Analytics for Healthcare
An overview
Extract medical facts and relationships from multiple
clinical and operational information sources
Analyze
Visualize
Analyze and Visualize past, present and future
scenarios to create an evidence based corpus of information
• Content Analytics with natural language processing to analyze trends, patterns,
deviations, anomalies and more unstructured information – detect discrepancies
in structured data
• Predictive Analytics for predictive scoring and probability analysis
• Healthcare Solution Accelerator enabling medical fact and relationship extraction through
pre-built annotators – built from expertise in previous engagements
• Planned integration to IBM Watson for Healthcare for deep question answering capability*
Enable clinicians, executives and knowledge workers to
Interact with information and derive insight in new ways
Integrate and leverage other systems to turn insight to action
11
* Future optional capability
© 2011 IBM Corporation
IBM Content and Predictive Analytics for Healthcare
How it works
IBM Watson for
Healthcare
Raw
Information
Unstructured Data
(Nurses notes, discharge
notes, etc.)
Structured Data
(Billing data, EMR, etc.)
Confirm hypotheses or seek alternative
ideas with confidence based responses
from learned knowledge*
IBM Content
and Predictive
Analytics
Analyzed and
Visualized
Information
Content Analytics
Predictive Analytics
• Natural Language Processing
• Medical Fact and Relationship
Extraction (Annotation)
• Trend, Pattern, Anomaly,
Deviation Analysis
• Predictive Scoring and
Probability Analysis
Health
Integration
Framework
Dynamic
Multimode
Interaction
Search and Visually
Explore (Mine)
Monitor, Dashboard and
Report
Question and Answer*
Data Warehouse and Model
Custom Solutions
Master Data Management
Advanced Case Management
Partners (HLI)
12
* Future optional capability
Specialized Research
Business Analytics
© 2011 IBM Corporation
Seton Healthcare Family
Reducing CHF readmission to improve care
“IBM Content and Predictive Analytics for Healthcare uses the same type of natural language processing as IBM Watson,
enabling us to leverage information in new ways not possible before. We can access an integrated view of relevant clinical
and operational information to drive more informed decision making and optimize patient and operational outcomes.”
Charles J. Barnett, FACHE, President/Chief Executive Officer, Seton Healthcare Family
Business Challenge
Seton Healthcare strives to reduce the occurrence of high
cost Congestive Heart Failure (CHF) readmissions by
proactively identifying patients likely to be readmitted on
an emergent basis.
What’s Smart?
IBM Content and Predictive Analytics for Healthcare
solution will help to better target and understand high-risk
CHF patients for care management programs by:
• Utilizing natural language processing to extract key elements
from unstructured History and Physical, Discharge Summaries,
Echocardiogram Reports, and Consult Notes
• Leveraging predictive models that have demonstrated high
positive predictive value against extracted elements of
structured and unstructured data
• Providing an interface through which providers can intuitively
navigate, interpret and take action
13
Smarter Business Outcomes
• Seton will be able to proactively target care management
and reduce re-admission of CHF patients.
• Teaming unstructured content with predictive analytics,
Seton will be able to identify patients likely for readmission and introduce early interventions to reduce
cost, mortality rates, and improved patient quality of life.
IBM solution
• IBM Content and
Predictive Analytics
for Healthcare
• IBM Cognos Business
Intelligence
• IBM BAO solution
services
© 2011 IBM Corporation
IBM Content and Predictive Analytics for Healthcare
What’s so innovative?
Patient
Procedure
A 42-year old white male
presents for a physical. He
recently had a right
hemicolectomy invasive
grade 2 (of 4) adenocarcinoma
in the ilocecal valve was found
and excised. At the same time
he had an appendectomy.
The appendix showed no
diagnostic abnormality.
Accurately extract
buried medical facts
and relationships with
medical annotators
Age: 42
Gender: Male
Race: White
Physicians
Other Clinicians
Care Coordinators
Researchers
hemicolectomy
diagnosis: invasive
adenocarcinoma
anatomical site:
ileocecal valve
grade: 2 (of 4)
Executives
Business Analysts
Claims
Fraud
Knowledge Workers
Procedure
appendectomy
diagnosis: normal
anatomical site:
appendix
Analyze compiled information for trends,
patterns, deviations, anomalies and
relationships in aggregate to reveal new
insights with content analytics
Model, score and predict the probability of
outcomes with predictive analytics
Other Systems and
Applications
Make insights accessible and
actionable for all clinical and
operational knowledge
workers (and systems)
Confirm hypotheses or seek alternative ideas from learned knowledge
via Watson for Healthcare from the same user interfaces*
14
* Future capability
© 2011 IBM Corporation
The Healthcare Solution Accelerator explained
Healthcare Solution Accelerator
• Based on expertise drawn from previous solution engagements
• Enables medical facts and information relationships (context) to be extracted from
information sources through pre-built annotators designed accelerate time-to-value
and shorten the solution development process
• Example: Extraction of clinical notes by identifying types of clinical named entities
including medications, dosages, diseases/disorders, signs/symptoms, lifestyle indicators
and other contextually relevant information
15
© 2011 IBM Corporation
Using Content and Predictive Analytics to Impact Outcomes
Starting in 2012 … Hospitals will be penalized for having high readmission rates - Medicare
discharge payments will be reduced in key areas such as Congestive Heart Failure (CHF)
Use Information
to Determine
High Risk Profile
Simple Theory,
Harder Reality
Identify
Key
Indicators
What is
accepted
and known
Examples:
Smoking Status
LVEF Levels
Errors in
Smoking
Status
Status Fixed,
Unidentified
Smokers Found
LVEF Levels
Found Only in
Textual Reports
LVEF Levels
Extracted from
Lab Reports
Do we have all the
answers today?
What is
unknown
16
Identify Patients
That Match
Profile
What about any
unknown factors?
Can we improve
outcomes even more?
Analyzed
Information
Visually
Explore
Information
in New Ways
Coordinate Care
to Prevent
Readmission
Predict High
Risk of CHF
Readmission
with Accepted
Indicators
Intervene and
Coordinate
Care – Optimal
Outcome?
Predict High
Risk of CHF
Readmission
with Optimal
Indicators
Intervene and
Coordinate
Care –
Optimized
Outcome
Uncover new
2D Insights Not
Possible Before
•
•
•
•
Improve care
Reduce readmissions
Lower costs
Avoid penalties
© 2011 IBM Corporation
ICPA Live Solution Demo
© 2011 IBM Corporation
Optimizing Insight Discovery
End Users
Content (NLP) Modeling
Analysts
Text Mining / Discovery
Predictive Modeling
Analyzed
Information
Step 1: Search and explore
(or mine) information to
understand source data
Step 2: Customize by building
content (NLP) and predictive
models
Step 3: Expose insights to multiple users
and systems (e.g. custom apps, mobile
devices, dashboards)
© 2011 IBM Corporation
Solution Demo
© 2011 IBM Corporation
The IBM Content Analytics User Interface
Explained
Search Query Exploration
Views, Filters and
Thresholds
Automatically Extracted
Facts and Relationships
Analyzed Concepts,
Entities, Meta Data,
Classifications, etc.
Visualization with Drill Down for Exploration and Mining
© 2011 IBM Corporation
Demo Scenario: Medical Device
Manufacturer
2
Interactive Discovery
(ex: trend & outlier analysis)
1
Source Content
(ex: device incident reports,
customer emails)
Action
3
(ex: prepare for recall)
4
Advanced Case Management
(trigger case process to manage intervention)
Operational Reporting
(ex: BI dashboard including content metrics)
© 2011 IBM Corporation
MedWatch: FDA Public Safety Information
1
© 2011 IBM Corporation
2
© 2011 IBM Corporation
2
© 2011 IBM Corporation
2
© 2011 IBM Corporation
2
© 2011 IBM Corporation
2
© 2011 IBM Corporation
2
© 2011 IBM Corporation
4
© 2011 IBM Corporation
Demo Recap: Medical Device Manufacturer
2
Interactive Discovery
(ex: trend & outlier analysis)
1
Source Content
(ex: device incident reports,
customer emails)
Action
3
(ex: prepare for recall)
4
Advanced Case Management
(trigger case process to manage intervention)
Operational Reporting
(ex: BI dashboard including content metrics)
© 2011 IBM Corporation
Tuesday May 4th, 2010
 Baxter International Inc. said Monday
it would recall the approximately
200,000 Colleague brand drug-infusion
pumps that are on the market, after
years of malfunctions with the device,
along with patient injuries and deaths.
 The Colleague pumps have been
widely used in hospitals, especially in
the U.S., to deliver medication and
other fluids to patients.
 Approximately 200,000 units
recalled
 Estimated cost of recall between
$400-600 million
© 2011 IBM Corporation
Building Solutions and Getting Started
© 2011 IBM Corporation
Content Analytics Reference Architecture with Watson
Data Import
Data Layer
Structured Data
DB
Connect
B2B Data
Broker
Federation
ETL / Cleanse
B2B Data
Triple
Store
Data Warehouse
OLAP
NLP
Content
Analytics
Index
Watson
Content
Prep
33
Applications
Social Media
Analytics
Social Biz
Apps
Sentiment
Analysis
Industry
Apps
Entity
Analytics
BI
Reporting
Database
Unstructured
Content
Crawler
Data Fusion
Watson
Content
Business
Analytics
Predictive
Analytics
Other
Analytics
Watson
Services
Operational
HC Apps
Clinical HC
Apps
Rapid
Insight
Enterprise
Search
Watson
Specific
© 2011 IBM Corporation
Content Analytics Reference Architecture with Mappings
Data Import
Data Layer
Structured Data
DB
Connect
B2B Data
Broker
Federation
ETL / Cleanse
B2B Data
Triple
Store
Data Warehouse
OLAP
NLP
Content
Analytics
Index
ICPA for HC
Applications
Social Media
Analytics
Social Biz
Apps
Sentiment
Analysis
Industry
Apps
Entity
Analytics
BI
Reporting
Database
Unstructured
Content
Crawler
Data Fusion
Watson
Content
Prep
Watson
Content
Business
Analytics
Predictive
Analytics
Other
Analytics
Watson
Services
Operational
HC Apps
Clinical HC
Apps
Rapid
Insight
Enterprise
Search
Watson
Specific
Watson HC
HIF Ecosystem
34
© 2011 IBM Corporation
IBM BAO Content and Predictive Solution Services
3 paths to value based on your needs
Start finding actionable insights in 6+ weeks
OPTION 1: ICPA FAST PATH PILOT
• UIMA Center of Competence
• Business Value Assessments
• Start with no charge strategy
workshop
Define and
Implement
Pilot
6-8 Weeks
1-4 Weeks
OPTION 2: ICPA SOLUTION IMPLEMENTATION
Start Here
Strategy
Workshop
Evaluate
Pilot
Use Case
Definition
½ - 1 Day
Design and
Build
Solution
Destination
Complementary
Deploy
Solution to
Production*
Integrate
Watson
(Future)
OPTION 3: BUSINESS CASE and ROADMAP
Assess
Requirements
35
Develop
Roadmap
and Business
Case
Incrementally
Deploy to
Production
* Optionally expand
to other use cases
© 2011 IBM Corporation
How to get started
NOW
Unique Value Delivered
1
Start with IBM Content
and Predictive Analytics
Address pressing clinical and
operational issues today
2
Expand and integrate
ICPA-based solutions
Expand solution value by
integrating other systems and
capabilities
• IBM BAO Enterprise Services
Clinical
Outcomes
Operational
Outcomes
• IBM Content and Predictive
Analytics for Healthcare
• IBM BAO Solution Services –
UIMA Center of Competence
• Also available as a Workload
Optimized System
FUTURE
NOW
3
Complement with IBM
Watson for Healthcare
Maximize solution value by
extending with IBM Watson
for Healthcare for real-time
confidence based answers
• Advanced Case Management
• Business Analytics
• Data Warehouse and Data
Models
• Master Data Management
• Partner Solutions
KEY
Future Optional
Capability
ICPA for
Healthcare
Optional
capabilities
Value Maximized
36
© 2011 IBM Corporation
New Healthcare solutions from IBM
IBM Content and Predictive Analytics for Healthcare (ICPA) … a synergistic solution to IBM Watson
Helps transform healthcare clinical and operational decision making by uniquely applying multiple
analytics services to derive and act on new insights in ways not previously possible – also available as a
Workload Optimized System
The first solution designed to complement and leverage IBM Watson
ICM Care Management for IBM Case Manager
Extends IBM Case Manager with a patient-centric care management platform that empowers Care
Coordinators to create personalized care plans for patients, coordinate the care of a patient as they
move between care settings, monitor the progress of the plan and communicate / collaborate with
providers and patients within a PCMH (Patient Centered Medical Home) at an Accountable Care
Organization
Data Model for Healthcare Provider Data Warehouse
A single integrated Healthcare model for the Healthcare market that includes Provider, Payer, and
Supply Chain content
The data model comprises three tiers including a Business Data Model, Atomic Warehouse Model and
a Dimensional Warehouse Model - also included are HL7 mappings to the model and supporting
deployment methodology
© 2011 IBM Corporation
Turning insight into action for care coordination
Patient and
operational source
data including paper
based information
IBM Content and Predictive Analytics
identifies high risk patients for
readmission improving patient care
while reducing readmission
frequency and costs
Care based on IBM Care Management
Clinical
Outcomes
Operational
Outcomes
• Identified patients trigger care coordination interactions
• Care plan developed and implemented
• Single view of interactions with the patient
• Post-discharge follow-up proactively managed
IBM Case Manager Solution Accelerator
38
A patient-centric care
management platform that
empowers Care Coordinators
© 2011 IBM Corporation
Next steps
Find out more about IBM Content and Predictive
Analytics for Healthcare
http://www.ibm.com/software/ecm/content-analytics/predictive.html
Conduct a content and predictive analytics
strategy planning workshop
Contact your IBM client representative
39
© 2011
Corporation
© 2011
IBMIBM
Corporation
Use Case Discussion
© 2011 IBM Corporation
IBM is helping to transform healthcare
Revealing clinical and operational insights in the high impact overlap between
clinical and operational – enabling low cost accountable care
IBM Content and Predictive Analytics for Healthcare
 Readmission prevention
Diagnostic assistance
 Claims management
Clinical treatment
effectiveness
Critical care intervention
Research for improved
disease management
 Fraud detection and
prevention
Clinical
Outcomes
Operational
Outcomes
 Voice of the patient
 Patient discharge and
follow-up care
 Improved patient satisfaction at lower costs
 Enhanced patient care with optimized outcomes
41
© 2011 IBM Corporation
ICPA Clinical Use Cases
Clinical
Outcomes
Operational
Outcomes
 Diagnostic Assistance: Highly correlated symptom to health/disease analysis issues visualized with
predictive guidance on diagnosis to improve treatment and outcomes … with predicted or
forecasted costs.
 Clinical Treatment Effectiveness: Examine patient-specific factors against the effectiveness of a
healthcare organizations specific treatment options and protocols … including comparisons to
industry wide outcomes and best practices.
 Critical Care Intervention: Early detection of unmanageable or high risk cases in the hospital that
leads to interventions to reduce costs and maintain or improve clinical conditions … including case
based interventions.
 Research for Improved Disease Management: Perform analysis and predict outcomes by extracting
discreet facts from text, such as: patient smoking status, patient diet and patient exercise regime to
find new and better treatment options … use a mechanism for differentiation or to secure research
grants.
© 2011 IBM Corporation
ICPA Operational Use Cases
Clinical
Outcomes
Operational
Outcomes
 Claims Management: All claims involve unstructured data and manually intensive analysis. Analyze
claims information documented in cases, forms and web content to understand new trends and
patterns to identify areas … perfect for process improvement, cost reduction and optimal service
delivery.
 Fraud Detection and Prevention: Uncover eligibility, false assertions and fraud patterns trapped in
the unstructured data to reduce risk before payments are made … usually represented by a word or
combination of words in text that can’t be detected with just structured data.
 Voice of the Patient: Include unstructured data and sentiment analysis from surveys and web forms
in analysis of patient and member satisfaction … this will be key as the industry moves to a value
based model.
 Prevention of Readmissions: Discover key indicators which are indicative of readmission to alert
healthcare organizations to these so that protocols can be altered to avoid readmission … this is key
as new Medicare payment penalties go into effect in 2012.
 Patient Discharge and Follow-up Care: Understand and monitor patient behavior to proactively
inform preventative and follow-up care coordinators before situations get worse.
© 2011 IBM Corporation
Unstructured Information Use Landscape
Natural Language Processing is Needed to
Search
Deep QA
Analyze / Visualize
(Trends, Patterns, Relationships)
Commerce
Search
eDiscovery,
Legal
Enterprise
Search
Expertise
Locator
Product
Quality
Care
Care
Consideration
Consideration
Analysis
Analysis
Safety,
Defects,
Maintenance
Treatment
Treatment
Protocol
Protocol
Analysis
Analysis
What’s
What’s New?
New?
Consumer
Consumer
Portal
Portal
Clinical
Outcomes
Operational
Outcomes
Risk, Fraud,
Security
Claims
Analysis
Research
(Biz, Edu, Legal,
Social
Scientific)
Media,
Marketing
VoC, Churn,
Cust Svc
Coding
Coding
Automation
Automation
Patient
Patient
Inquiry
Inquiry
Patient
Patient
Workup
Workup
Treatment
Treatment
Authorization
Authorization
Longitudinal
Patient Electronic
Health Information
Differential
Differential
Diagnosis
Diagnosis
Treatment
Treatment Options
Options
Specialty
Specialty Diagnosis
Diagnosis &
&
Treatment
Treatment Options
Options
Caregiver
Caregiver Education
Education
Population
Population
Analysis
Analysis &
&
Care
Care Mgmt
Mgmt
Second
Second
Opinion
Opinion
On-going
On-going
Treatment
Treatment
Specialty
Specialty
Research
Research
GenomicGenomicbased
based Analysis
Analysis
Call Center,
Help, Self
Service
Knowledge
Mgt
Semantic Understanding, Ontology Mgt and Big Data
© 2011 IBM Corporation