Bayesian networks An emerging tool for disease risk estimation and - - PowerPoint PPT Presentation

▶

Jun 26, 2023 540 likes •765 views

Bayesian networks An emerging tool for disease risk estimation and clinical decision- making with application to coronary artery disease Concurrent Session #B6 Session name: HTA Methods Monday, April 15, 2019 1:30-2:45 PM Salon 5&6

SLIDE 1

Bayesian networks

An emerging tool for disease risk estimation and clinical decision- making with application to coronary artery disease

lighthouseoutcomes.com

Concurrent Session #B6 Session name: HTA Methods Monday, April 15, 2019 1:30-2:45 PM Salon 5&6

SLIDE 2

Disclosure

I have the following relevant financial relationships to disclose:
Employed by Lighthouse Outcomes Inc., Toronto, Ontario, Canada

SLIDE 3

Challenges

Key challenges in HEOR analyses

SLIDE 4

Key challenges in HEOR analyses

Precision medicine
Right treatment for the right person at the right time
Modelling problems are increasingly becoming complex
Biomarkers, patient characteristics, socio-demographics, genetics,

study and treatment factors

Balancing multiple objectives
Maximizing treatment effectiveness, safety and minimizing cost
Accounting for patient preference and expert opinion

SLIDE 5

Bayesian networks

With application to a coronary artery disease dataset

SLIDE 6

Bayesian network

Probabilistic graphical models
Useful for transparent reasoning under uncertainty
Flexible – fit to data or problem
Can combine data-driven computational methods with expert

input and background knowledge P(A,B,C,D,E) = P(A) P(B|A) P(C) P(D|B,C) P(E|D)

SLIDE 7

Step 1 – Structure learning

Methods for learning structure of Bayesian network
Machine learning
Maximize model likelihood penalized by model complexity
Perform conditional independence tests
Expert elicitation

Edge directions do not imply cause-and-effect relationships

Koller, D., Friedman, N., & Bach, F. (2009). Probabilistic graphical models: principles and techniques. MIT press.

SLIDE 8

Step 2 – Parameter estimation

Represent the conditional probability tables (i.e. probabilities of

a node given values of its “parents”)

Maximum likelihood-based methods

+ Data

SLIDE 9

Z-Alizadehsani dataset

Publicly available dataset from UCI ML repository1,2
Contains records of 303 visitors to cardiovascular research

center in Iran

55 predictors:
Demographic
Symptoms
ECG results
Biomarkers/lab measures
Coronary artery disease (CAD) is the outcome of interest
Result of angiography

1 Alizadehsani, R., Habibi, J., Hosseini, M. J., Mashayekhi, H., Boghrati, R., Ghandeharioun, A., ... & Sani, Z. A. (2013). A data mining approach for diagnosis of coronary artery disease. Computer methods and programs in biomedicine, 111(1), 52-61. 2 Alizadehsani, R., Habibi, J., Bahadorian, B., Mashayekhi, H., Ghandeharioun, A., Boghrati, R., & Sani, Z. A. (2012). Diagnosis of coronary arteries stenosis using data mining. Journal of medical signals and sensors, 2(3), 153.

SLIDE 10

CAD Bayesian network

Biomarkers/lab measures ECG measurements Outcome Demographic Symptoms

Legend

Unconnected nodes (independent) are not shown for simplicity

SLIDE 11

Network parameters (example)

Biomarkers/lab measures ECG measurements Outcome Demographic Symptoms

Legend

SLIDE 12

Network parameters (example)

Outcome Demographic Biomarkers/lab measures ECG measurements Symptoms

Legend

Values of parent nodes Probability of CAD

SLIDE 13

Markov blanket of CAD

Outcome Demographic Biomarkers/lab measures ECG measurements Symptoms

Legend

The Markov blanket of CAD contains variables that summarize all information about CAD

SLIDE 14

Marginal probability of variables

SLIDE 15

Personalized probability update

Observed in clinic for patient X

SLIDE 16

Prognostic values (PV)

Demographics Biomarkers/lab measures ECG measurements Symptoms

Prognostic value (max value = 1.0) Variables (ordered by median PV) PV of zero represents uninformative

Cover, T. M., & Thomas, J. A. (2012). Elements of information theory. John Wiley & Sons.

SLIDE 17

Classification performance

Model type Feature name AUC (Mean ± standard deviation) Logistic regression Simple 0.84 ± 0.10 L2 penalty (ridge regression) 0.92 ± 0.06 L1 penalty (Lasso) 0.92 ± 0.05 Backward selection 0.86 ± 0.11 Bayesian network Edgeless network 0.48 ± 0.13 Naïve Bayes 0.93 ± 0.04 Hill climbing 0.93 ± 0.04

SLIDE 18

Other uses of Bayesian networks

Ideal responders
Select a patient profile (jointly) based on “most

likely” to have CAD

Effect modification
Backwards reasoning (“diagnostic reasoning”)
Find probability that a person with CAD has

ventricular hypertrophy

Decision & cost-effectiveness models
Adaptive questionnaires & clinical decision

support tools1,2

1 Seixas, F. L., Zadrozny, B., Laks, J., Conci, A., & Saade, D. C. M. (2014). A Bayesian network decision model for supporting the diagnosis of dementia, Alzheimer׳ s disease and mild cognitive impairment. Computers in biology and medicine, 51, 140-158. 2 Sesen, M. B., Nicholson, A. E., Banares-Alcantara, R., Kadir, T., & Brady, M. (2013). Bayesian networks for clinical decision support in lung cancer care. PloS one, 8(12), e82349.

SLIDE 19

Conclusion

Key messages

SLIDE 20

Key messages

Individual risk prediction for personalized medicine
Transparent models
Ability to discover and represent relationships between

variables

Incorporation of background knowledge
“Living” models that adapt new data become available

SLIDE 21

Thank you

Primary contacts:

agupta@lighthouseoutcomes.com parora@lighthouseoutcomes.com

info@lighthouseoutcomes.com

+1 800 535 9760

1 University Avenue. 3rd Floor Toronto, Ontario M5CJ 2P1 Canada @lighthouseheor