+ All Categories
Home > Documents > Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg...

Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg...

Date post: 17-Sep-2018
Category:
Upload: duongnhi
View: 221 times
Download: 0 times
Share this document with a friend
57
Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics and Intelligent Systems Medical University of Vienna [email protected] Gastpräsentation: Matthias Samwald Section for Medical Expert and Knowledge-based Systems
Transcript
Page 1: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Methoden zur medizinischen Datenmodellierung

Georg Dorffner

Section for Artificial Intelligence, Center for Medical Statistics, Informatics and Intelligent Systems

Medical University of Vienna

[email protected]

Gastpräsentation: Matthias SamwaldSection for Medical Expert and Knowledge-based Systems

Page 2: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Using information technologies to make effective treatment optimization accessible to every patient

Ass.‐Prof. Mag. Dr. Matthias SamwaldCeMSIIS, Medical University of Vienna

The Medication Safety Code project

Funded by Austrian Science Fund (FWF): [P 25608‐N15]

Page 3: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Drug safety and effectiveness vary drastically between patients

Up to 100,000 deaths and 2 million hospitalisations per year in the United States (Lazarou, 1998)

Page 4: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Part of this variability can be explained bygenetic variation influencing individual pharmacokinetics and –dynamics

The frequencies of actionable pharmacogenes. Over 95% of the population carries at least oneactionable genotype for one of the genes covered by the DPWG guidelines (Dunnenberger, 2015)

Page 5: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

However, several barriers to widespreadimplementation of pharmacogenomics remain

×Lack of selection of a panel of clinically relevant PGx markers

×Lack of information on cost‐effectiveness and cost‐consequences of PGx testing

×Lack of knowledge when test needs to be conducted; organizational overheads discourage testing

×Lack of appropriate information technologies for interpretation and incorporation in the workflow of physicians and pharmacists

Page 6: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

► We are developing methods for making essential pharmacogenomic data and decision support available

without barriers and at low cost, to all patients in diverse medical settings.

1) Evaluate pre-emptive testing options2) Evaluate potential impact and analyse cost-utility tradeoff

3) Develop solutions for integration in routine care

Page 7: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

In pre‐emptive pharmacogenomic testing, data on all essential pharmacogenes are tested in one go and aremade available for immediate use in the future

Gene Substances associated with gene in pharmacogenomic guidelinesCore list

CYP2C19 amitriptylinea,b, clomipraminea,b, clopidogrela,b, desipraminea, doxepinb, imipraminea,b, nortriptylinea,b, trimipraminea, citalopramb, escitalopramb, esomeprazoleb, lansoprazoleb, moclobemideb, omeprazoleb, pantoprazoleb, rabeprazoleb, sertralineb, voriconazoleb

CYP2C9 warfarina, acenocoumarolb, glibenclamideb, gliclazideb, glimepirideb, phenprocoumonb, phenytoinb, tolbutamideb

CYP2D6 amitriptylinea,b, clomipraminea,b, codeinea,b, desipraminea, doxepina,b, imipraminea,b, nortriptylinea,b, trimipraminea, aripiprazoleb, atomoxetine b, carvedilol b, clozapineb, codeinea,b, Duloxetineb, flecainideb, flupenthixolb, haloperidolb, metoprololb, mirtazapineb, olanzapineb, oxycodoneb, paroxetineb, propafenoneb, risperidoneb, tamoxifenb, tramadolb, venlafaxineb, zuclopenthixolb

CYP3A5 tacrolimusb

DPYD capecitabinea,b, fluorouracila,b, tegafura,b

TPMT azathioprinea,b, mercaptopurinea,b, thioguaninea,b

UGT1A1 irinotecanb

SLCO1B1 simvastatina

VKORC1 warfarina, phenprocoumonb

OthersF5 estrogen-containing oral contraceptivesb

HLA-B abacavira,b, allopurinola, carbamazepinea, ribavirina,b

IFNL3 peginterferon alfa-2aa, peginterferon alfa-2ba, ribavirina,b

a: substance covered by CPIC guidelines, b: substance covered by DPWG guidelines. Data as ofmid‐2014.

Page 8: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

We are creating a barrier‐free system for storing and interpreting personal pharmacogenomic information: The Medication Safety Code

Current version contains data on ~15 pharmacogeneswith clinically actionable variants

Page 9: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

We are creating a barrier‐free system for storing and interpreting personal pharmacogenomic information: The Medication Safety Code 

Page 10: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Technology

• Compressed pharmacogenomic data in QR codes• Can be readily decoded and interpreted with common 

mobile devices• No central database • Data remains anonymous • Backed by a sophisticated knowledge base we created

(Samwald, 2013; Miñarro‐Giménez, 2014; Samwald, 2015)

Page 11: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics
Page 12: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

We are evaluating the accuracy of various targeted, low‐cost assays suitable for pre‐emptive testingcompared to next‐gen sequencing

Fraction of tested genes resulting in abberations in haplotype calling with restricted assaycompared to next‐gen sequencing. Based on full genome sequences of 2504 persons. Manuscriptcurrently under review at ‘Pharmacogenomics’.

Page 13: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Age CCAE Medicaid Medicare

0-13 0.77% 1.40% N/A

14-39 9.94% 14.71% N/A

40-64 13.70% 32.20% N/A

>=65 N/A N/A 26.80%

We analyzed prescriptions of89 million patients in the US to evaluatethe potential utility of pre‐emptive testing

Fraction of patients receiving two or more new pharmacogenomic drugs within a 4‐year timewindow (2009‐2012). Manuscript currently under review at ‘The Pharmacogenomics Journal’.

Page 14: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

We analyzed prescriptions of89 million patients in the US to evaluatethe potential utility of pre‐emptive testing

Fraction of tested patients that would receive two or more new pharmacogenomic drugs withina 4‐year time window (2009‐2012) in a hypothetical scenario where testing is initiated at firstincident prescription of a pharmacogenomic drug. Manuscript currently under review at ‘ThePharmacogenomics Journal’.

Age CCAE Medicaid Medicare

0-13 9.60% 12.10% N/A

14-39 31.10% 37.90% N/A

40-64 43.30% 59.00% N/A

>=65 N/A N/A 54.60%

Page 15: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Join us athttp://safety‐code.org/

Page 16: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

http://upgx.eu/ European project starting January 2016

Page 17: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Local team (Medical University of Vienna) 

Ass.‐Prof. Mag. Dr. Matthias Samwald (principal investigator)

Mag. Sebastian Hofer

Dr. Kathrin Blagec 

Wolfgang Kuch

Web

http://samwald.info/

http://safety‐code.org/

http://upgx.eu

Thanks!

Page 18: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

References

Dunnenberger HM, Crews KR, Hoffman JM, Caudle KE, Broeckel U, Howard SC, Hunkler RJ, Klein TE, Evans WE, Relling MV: Preemptive Clinical Pharmacogenetics Implementation: CurrentPrograms in Five US Medical Centers. Annu Rev Pharmacol Toxicol 2015, 55:89–106.

Lazarou J: Incidence of Adverse Drug Reactions in Hospitalized Patients: A Meta‐analysis of Prospective Studies. JAMA: The Journal of the American Medical Association 1998, 279:1200–1205.

Miñarro‐Giménez JA, Blagec K, Boyce RD, Adlassnig K‐P, Samwald M: An Ontology‐Based, Mobile‐Optimized System for Pharmacogenomic Decision Support at the Point‐of‐Care. PLoSONE 2014, 9:e93769.

Samwald M, Adlassnig K‐P: Pharmacogenomics in the pocket of every patient? A prototype based on quick response codes. J Am Med Inform Assoc 2013, 20:409–412.

Samwald M, Blagec K, Hofer S, Freimuth R: “Analysing the potential for incorrect haplotype callswith different pharmacogenomic assays in different populations: a simulation based on 1000 Genomes data.” Pharmacogenomics, September 30, 2015. 

Samwald M, Giménez JAM, Boyce RD, Freimuth RR, Adlassnig K‐P, Dumontier M: Pharmacogenomic knowledge representation, reasoning and genome‐based clinical decisionsupport based on OWL 2 DL ontologies. BMC Medical Informatics and Decision Making 2015, 15:12.

Page 19: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Beispiel 2: Kontinuierliche Schlafmodellierung

Roman Rosipal, Achim Lewandowski, GD

Page 20: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Continuous sleep profile

Automatic sleep analysis

Page 21: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Continuous model

Continuous probability vectors

Preprocessing and feature extraction

Supervised learning of class-conditional GMMs

Unsupervised reshuffling of GMMs

AR(10) coefficients

GMMs

Calculate posteriors

GMMs

Polysomnographic recordings

Single channel EEG data

R&K labels

Spindle

Process

Artifacts

Detection

Page 22: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

b003302

0 50 100 150 200 250 300 350 400 4500

1

tim e [m in]

deep

0

1

s2

0

1

s1

0

1

wak

e

C4

C3

otherswake

s1s2s3s4

REM

B003302: Female, 76 years

Page 23: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

“Subjective sleep quality” versus “Objective sleep quality”R&K

SSA-1 Number of stage shifts (/hr TST)

-10.00 0.00 10.00

s_qua_21

-30.00

-20.00

-10.00

0.00

10.00

20.00

fsts

t00c

fstst00c = -0.42 + 0.53 * s_qua_21R-Square = 0.19

20-39 40-59 >=60

-10.00 0.00 10.00

s_qua_21

fstst00c = -1.30 + 0.21 * s_qua_21R-Square = 0.03

-10.00 0.00 10.00

s_qua_21

fstst00c = -1.53 + 0.01 * s_qua_21R-Square = 0.00

Page 24: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

“Subjective sleep quality” versus “Objective sleep quality”hGMM

SSA-1 Number of stage shifts “deep – S2”

-10.00 0.00 10.00

s_qua_21

-0.0200

0.0000

0.0200

0.0400

sc_d

_s2c

sc_d_s2c = 0.00 + -0.00 * s_qua_21R-Square = 0.15

20-39 40-59 >=60

-10.00 0.00 10.00

s_qua_21

sc_d_s2c = 0.00 + -0.00 * s_qua_21R-Square = 0.14

-10.00 0.00 10.00

s_qua_21

sc_d_s2c = 0.00 + -0.00 * s_qua_21R-Square = 0.13

Page 25: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Result• Measures for sleep continuity and

architecture based on R&K showed significant correlations with subjective sleep quality only in young subjects.

• In contrast, measures for sleep continuity and architecture based on hGMM showed significant correlations in all age-groups

© Alle Rechte liegen bei den Autoren.

Page 26: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Continuous probability model II

• z: state• x: AR(10) vector• c: RK class• s: spindle class

Model assumption: given the state z, x,c, and s are independent

Page 27: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Applying the modelExpress the current sleep as R&K posterior for given x and s

Page 28: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Applying the modelExpress the current sleep as ‚raw‘ state posterior for given x and s

Page 29: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Relative time spent in a state

N

i ii NaxzpzRTS1

/),|()(

• subject with observations (x1,a1),...,(XN,aN)

• Looking for a measure judging the time spent in a state z, weighted by ‚intensity‘ of a visit:

e.g. calculate sum of posteriors for each state z, relative to length of recording

Page 30: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Correlation to outside criteria• can compare given sleep quality

variable (e.g. result of concentration test) with RTS(z) for a list of subjects

=> use Spearman-Rank correlation to detect monotonic relationships

Page 31: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Results: sleep stages

Page 32: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Results: probabilistic sleep model

Rosipal et al., Biol Psychol, 2013

Page 33: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

HMM trajectories

Same subject(1st/2nd night)

Page 34: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Individual differences

• Q: Is it really justified to ask for more correlations?

Page 35: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Conclusions• PSG/EEG objectively describes

physiology of sleep• Visual approaches lead to „fuzzy“

ground truth, automation leads to reliability

• Data-based approaches can extract more information

• But relationship to outside criteria about sleep quality due to other effects (context, individual characteristics

Page 36: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Beispiel 3: Vorhersage der Mortalität nach Herzstillstand

Fritz Sterz, Stefan Aschauer, GD

Page 37: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me out of hospital cardiac arrest

(OOHCA)

• major health problem

• 500.000 patients in United States and Europe /year

• overall mortality: 8% - 11%

background

Page 38: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

background• OOHCA has a very uncertain

outcome

• no valid outcome scoring system• problem in giving reliable outcome

estimation • delicate decisions

based on experience and gut feeling

Page 39: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

aim

• to assess the predictability of outcome after OOHCA, based on a number of observational variables

• to identify variables with high predictive power

• to assess whether a multivariateapproach is superior to a univariateone

• to derive a OOHCA outcomeprediction score tool

Page 40: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

benefit• improvement of the predictability of

patient’s survival would be of major medical and socioeconomic interest.

• valid outcome estimation could facilitate decision-making for persons in authority and could save medical resources

Page 41: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

methods

• based on a cardiac arrest-registry with > 4000 patients which were resuscitated from OOHCA and which were admitted to the Department of Emergency Medicine at a large University Hospital

• multivariate logistic regression was applied on 20 variables before ROSC deemed to have high predictive power

Page 42: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

methods

• the framework of machine learning was chosen

• a 10-fold cross-validation was done for reliable estimates and confidence intervals

• main performance parameter was the area under the ROC curve (AUC)

Page 43: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

variablesVariable name Description Value Scalesex Sex of the patient Male=0, Female=1 binaryage Age of the patient In years, at the time of cardiac arrest metricbmi Body Mass Index Weight (kg) / Size (m) Squared metricdiabetes Previous diagnosis of diabetes Diabetes = 1, no diabetes = 0 binarysmoker Patient is a smoker Smoker=1, nonsmoker=0 binarymyocinfarct Patient previously had a myocardial infarction Infarction=1, no infarction=0 binarykhk Previous diagnosis of Coronary Artery Disease CAD=1, no CAD=0 binaryhypertension Previous diagnosis of hypertension Hypertension=1, no hypertension=0 binaryheartfail Previous diagnosis of heart failure Heart failure=1, no heart failure=0 binarycvi Previous diagnosis of chronic venous insufficiency CVI=1, no CVI=0 binary

copd Previous diagnosis of chronic obstructive pulmonary disease COPD=1, no COPD=0 binaryopcpre OPC score prior to cardiac arrest Score 1 to 5 ordinal, treated as metricnyh5pre NYH5 score prior to cardiac arrest Score 1 to 5 ordinal, treated as metricnoflow Minutes between cardiac arrest and first aid (length of "no

flow" time) in minutes metricmin2srosc Minutes between cardiac arrest and SROSC in minutes metriccause Main cause of cardiac arrest Cardiac=1, non-cardicac=0 binary

firstaidFirst aid performed by physician, family member, paramedic or layman Physician=1, non-Physician=0 binary

nodefi Number of defibrillation shots Count of shots metricadrenaline Amount of adrenaline applied Total amount (in …) metric

shockable Shockability of rhythm in first defibrillation Shockable=1, non-shockable=0 binarydefireaction Reaction tp the first defibrillation Not shockable=0, shockable and VT/VF (as reaction

to first defi)=1, shockable and PEA=2, shockable+Asystole=3, shockable+SR/RHY/SVES/VES/AVES+ no pulse=4, shockable+pulse=5

ordinal, treated as metric

Page 44: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Initial variables, histograms

0 0.5 10

500

1000

1500

0 0.5 10

500

1000

1500

0 0.5 10

500

1000

1500

0 0.5 10

1000

2000

-5 0 50

200

400

0 0.5 10

500

1000

1500

-5 0 50

500

1000

1500

0 0.5 10

1000

2000

-10 0 100

500

1000

0 0.5 10

500

1000

1500

-5 0 50

500

1000

-10 0 100

500

1000

1500

0 0.5 10

500

1000

1500

0 0.5 10

1000

2000

0 0.5 10

500

1000

1500

-10 0 100

500

1000

1500

0 0.5 10

500

1000

1500

0 0.5 10

1000

2000

0 0.5 10

1000

2000

0 0.5 10

500

1000

1500

0 0.5 10

500

1000

1500

0 0.5 10

500

1000

1500

0 0.5 10

1000

2000

0 0.5 10

500

1000

Page 45: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Data sets

Page 46: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

witnessed Trainingset median 25% percentile

75% percentile Percent 1 Percent 0

sex 27.53% 72.47%age 59 49 69 0 0bmi 26.12 23.88 29.22 0 0diabetes 16.20% 83.80%smoker 30.90% 69.10%myocinfarct 12.92% 87.08%cad 21.91% 78.09%hypertension 32.21% 67.79%heartfail 11.05% 88.95%cvi 5.99% 94.01%copd 9.74% 90.26%opcpre 1 1 1 nyh5pre 1 1 2 noflow 1 0 6.5 min2srosc 20 10 30 cause 69.76% 30.24%firstaid 34.18% 65.82%nodefi 2 0 4 adrenaline 2 0 4 defireaction 1 0 2 shockable 59.83% 40.17%cpc30d 3 1 5 mortality 39.89% 60.11%

Testset median 25% percentile

75% percentile Percent 1 Percent 0

sex 27.84% 72.17%age 61 50 71 bmi 26.23 24.11 29.41 diabetes 20.62% 79.38%smoker 31.62% 68.39%myocinfarct 14.09% 85.91%cad 24.74% 75.26%hypertension 41.92% 58.08%heartfail 14.78% 85.22%cvi 4.81% 95.19%copd 6.53% 93.47%opcpre 1 1 2 nyh5pre 1 1 2 noflow 1 0 5 min2srosc 19 12 32 cause 62.54% 37.46%firstaid 49.49% 50.52%nodefi 1 0 3 adrenaline 1 0 3 defireaction 1 0 2 shockable 54.30% 45.70%cpc30d 3 1 5 mortality 42.27% 57.73% 

Page 47: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Non-witnessedTrainingset median 25% percentile 75% percentile Percent 1 Percent 0 sex 30.46% 69.54%age 56 41 68 bmi 25.94 22.86 28.72 diabetes 14.37% 85.63%smoker 27.01% 72.99%myocinfarct 9.77% 90.23%khk 16.09% 83.91%hypertension 26.44% 73.56%heartfail 10.92% 89.08%cvi 4.02% 95.98%copd 8.62% 91.38%opcpre 1 1 1 nyh5pre 1 1 1 cause 43.68% 56.32%nodefi 1 0 4 adrenaline 4 2 6.5 defireaction 0 0 1 shockable 36.78% 63.22%cpc30d 5 5 5 mortality 76.44% 23.56%Testset median 25% percentile 75% percentile Percent 1 Percent 0 sex 20.00% 80.00%age 54 44.5 64.25 bmi 26.23 23.32 28.18 diabetes 0.00% 100.00%smoker 32.00% 68.00%myocinfarct 4.00% 96.00%khk 12.00% 88.00%hypertension 44.00% 56.00%heartfail 4.00% 96.00%cvi 4.00% 96.00%copd 16.00% 84.00%opcpre 1 1 1 nyh5pre 1 1 1 cause 60.00% 40.00%nodefi 1 0 6.25 adrenaline 4 2 8 defireaction 1 0 1 shockable 60.00% 40.00%cpc30d 5 1.75 5 mortality 68.00% 32.00% 

Page 48: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Page 49: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Page 50: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Score

Page 51: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Simplifed scorePredictor Points Predictor Points

1. Age group 3. Minutes until SROSC>80 32 >100min 35>70 27 >50min 21>50 23 >40min 13>60 20 >30min 10>40 16 >20min 7≤40 11 >10min 4

>0min 12. Adrenalin administered 0min 0>10mg 24>5mg 12 4. Shockable rhythm?>4mg 7 Yes ‐15>3mg 5 No 0>2mg 4>1mg 2>0mg 10mg 0 Total score

Total score Probability for mortality

<13 10%13‐22 30%23‐30 50%31‐40 70%>40 90%

Page 52: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

predicted mortality

Page 53: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Beispiel 4: Simulation molekularer Dynamiken

Bernhard Knapp, Wolfgang Schreiner, GD

Page 54: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Molecular dynamics simulation –T-cells

Page 55: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Patterns within 50ns

Knapp et al., Plos One, 2013

Page 56: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Statistical significance

Page 57: Methoden zur medizinischen Datenmodellierung · Methoden zur medizinischen Datenmodellierung Georg Dorffner Section for Artificial Intelligence, Center for Medical Statistics, Informatics

Inst

itut f

ür A

rtific

ial I

ntel

ligen

ceZe

ntru

m fü

r Med

izin

isch

e S

tatis

tik,

Info

rmat

ik u

nd In

telli

gent

e S

yste

me

Affected regions


Recommended