SlideShare ist ein Scribd-Unternehmen logo
1 von 40
Downloaden Sie, um offline zu lesen
Machine learning versus traditional
statistical modeling and medical
doctors
Maarten van Smeden
Leiden University Medical Center
IBS ROeS - Lausanne
September 10, 2019
IBS ROeS 2019, Lausanne MaartenvSmeden
Left out artificial intelligence?
In medical research, “artificial intelligence” usually
just means “machine learning” or “algorithm”
IBS ROeS 2019, Lausanne MaartenvSmeden
Tech company business model
IBS ROeS 2019, Lausanne MaartenvSmeden
Tech company business model
https://bit.ly/2HSp8X5; https://bit.ly/2Z0Pfop; https://bit.ly/2KIcpHG; https://bit.ly/33IJhr9
IBS ROeS 2019, Lausanne MaartenvSmeden
Other success stories
https://go.nature.com/2VG2hS7; https://bbc.in/2Z1drXQ
IBS ROeS 2019, Lausanne MaartenvSmeden
IBM Watson winning Jeopardy
https://bbc.in/2TMvV8I
IBS ROeS 2019, Lausanne MaartenvSmeden
IBM Watson for oncology
https://bit.ly/2LxiWGj
IBS ROeS 2019, Lausanne MaartenvSmedenForsting, J Nuc Med, 2017, DOI: 10.2967/jnumed.117.190397
IBS ROeS 2019, Lausanne MaartenvSmeden
Machine learning everywhere (selection of last month)
https://bit.ly/2ka0HLq; https://go.nature.com/33TQgO6; https://bit.ly/2kp6X23; https://bit.ly/2lZuKWt; https://bit.ly/2lI298g
What are these
Machine Learning methods?
IBS ROeS 2019, Lausanne MaartenvSmeden
“Everything is an ML method”
https://bit.ly/2lEVn33
IBS ROeS 2019, Lausanne MaartenvSmeden
“ML methods come from computer science”
https://bit.ly/2zhbwPv; https://stanford.io/2TVp1xK; https://stanford.io/2ZfED0k
Leo Breiman Jerome H Friedman Trevor Hastie
CART, random forest Gradient boosting Elements of statistical learning
Education Physics/Math Physics Statistics
Job title Professor of Statistics Professor of Statistics Professor of Statistics
IBS ROeS 2019, Lausanne MaartenvSmeden
“ML methods for prediction, statistics for explaining”
Damen, BMJ, 2016, DOI:10.1136/bmj.i2416
363 developed models how many?
Decision trees 0
Random forests 0
Support vector machines 0
Nearest neighbor algorithms 0
Neural networks 1
IBS ROeS 2019, Lausanne MaartenvSmeden
“ML methods for prediction, statistics for explaining”
1See further: Kreiff and Diaz Ordaz; https://bit.ly/2m1eYdK
ML and causal inference, small selection1
• Superlearner (e.g. van der Laan)
• High dimensional propensity scores (e.g. Schneeweiss)
• The book of why (Pearl)
Wednesday 10:40-12:10 Keynote Session 3
Els Goetghebeur: Plea for a marriage of
machine learning and causal inference
IBS ROeS 2019, Lausanne MaartenvSmeden
Two cultures
Breiman, Stat Sci, 2001, DOI: 10.1214/ss/1009213726
IBS ROeS 2019, Lausanne MaartenvSmedenRobert Tibshirani: https://stanford.io/2zqEGfr
Machine learning: large grant = $1,000,000
Statistics: large grant = $50,000
IBS ROeS 2019, Lausanne MaartenvSmeden
Statistics Machine learning
Covariates Features
Outcome variable Target
Model Network, graphs
Parameters Weights
Model for discrete var. Classifier
Model for continuous var. Regression
Log-likelihood Loss
Multinomial regression Softmax
Measurement error Noise
Subject/observation Sample/instance
Dummy coding One-hot encoding
Measurement invariance Concept drift
Statistics Machine learning
Prediction Supervised learning
Latent variable modeling Unsupervised learning
Fitting Learning
Prediction error Error
Sensitivity Recall
Positive predictive value Precision
Contingency table Confusion matrix
Measurement error model Noise-aware ML
Structural equation model Gaussian Bayesian network
Gold standard Ground truth
Derivation–validation Training–test
Experiment A/B test
Adapted from Daniel Obserski: https://bit.ly/2YN12Xf and Robert Tibshirani: https://stanford.io/2zqEGfr
Language
IBS ROeS 2019, Lausanne MaartenvSmeden
ML refers to a culture, not to methods
Distinguishing between statistics and machine learning
• Substantial overlap methods used by both cultures
• Substantial overlap analysis goals
• Attempts to separate the two frequently result in disagreement
Pragmatic approach:
I’ll use “ML” to refer to models roughly outside of the traditional regression
types of analysis: decision trees (and descendants), SVMs, neural networks,
boosting etc.
Examples where “ML” has
done well
IBS ROeS 2019, Lausanne MaartenvSmeden
IBS ROeS 2019, Lausanne MaartenvSmeden
Example: retinal disease
Gulshan et al, JAMA, 2016, 10.1001/jama.2016.17216; Picture retinopathy: https://bit.ly/2kB3X2w
Diabetic retinopathy
Deep learning (= Neural network)
• 128,000 images
• Transfer learning (preinitialization)
• Sensitivity and specificity > .90
• Estimated from training data
IBS ROeS 2019, Lausanne MaartenvSmeden
Example: lymph node metastases
Bejnordi et al, JAMA, 2018, doi: 10.1001/jama.2017.14585. See our letter to the editor for a critical discussion: https://bit.ly/2kcYS0e
Deep learning competition
• 390 teams signed up, 23 submitted
• Only 270 images for training
• Test AUC range: 0.56 to 0.99
IBS ROeS 2019, Lausanne MaartenvSmeden
Deep learning on images
Many similar studies and challenges in radiology, pathology,
dermatology, opthalmology, gastroenterology, cardiology, ….
Topol, Nature Medicine, 2019, DOI: 10.1038/s41591-018-0300-7
IBS ROeS 2019, Lausanne MaartenvSmeden
Other sources of “medical” data
• Large scale gene expression data
• e.g. diagnosis of acute myeloid leukemia
https://bit.ly/2k8Ao8e
• Prognostication by text mining electronic health records
• e.g. predicting life expectancy
https://bit.ly/2k8Ao8e
• Analyzing social media posts
• e.g. pharmacovigilance, adverse events monitoring via Twitter posts
https://bit.ly/2m0KKrg
Examples where “ML” has
done poorly
IBS ROeS 2019, Lausanne MaartenvSmeden
Skin cancer and rulers
Esteva et al., Nature, 2016, DOI: 10.1038/nature21056; https://bit.ly/2lE0vV0
IBS ROeS 2019, Lausanne MaartenvSmeden
Predicting mortality – the conclusion
PlosOne, 2018, DOI: 10.1371/journal.pone.0202344
IBS ROeS 2019, Lausanne MaartenvSmeden
Predicting mortality – the results
PlosOne, 2018, DOI: 10.1371/journal.pone.0202344
IBS ROeS 2019, Lausanne MaartenvSmeden
Predicting mortality – the media
PlosOne, 2018, DOI: 10.1371/journal.pone.0202344; https://bit.ly/2Q6H41R; https://bit.ly/2m3RLrn
IBS ROeS 2019, Lausanne MaartenvSmeden
HYPE!
IBS ROeS 2019, Lausanne MaartenvSmeden
Systematic review clinical prediction models
Christodoulou et al. Journal of Clinical Epidemiology, 2019, doi: 10.1016/j.jclinepi.2019.02.004
“ML” versus traditional
statistics and medical
doctors
IBS ROeS 2019, Lausanne MaartenvSmeden
Comparison “ML” vs statistical models
• Machine learning methods versus statistical models is a false
dichotomy
• Advanced “ML” shows promise, especially in areas that are
not the traditional “tabular data” (e.g. images)
• Tabular data settings where “ML” can be compared with
traditional regression model techniques show little added value
in medical applications
IBS ROeS 2019, Lausanne MaartenvSmeden
Classification versus risk prediction
Most ML “classifiers” don’t come naturally with risk prediction, i.e.
a probability estimate of predicted outcome for individuals
• Possibly much large sample size needed to obtain reliable
(calibrated) risk predictions1 than reliable classifications
• Models can be trained to be optimized for a certain predictive
performance (e.g. AUC, sensitivity, calibration)
• Which performance to use to compare models are optimized
for different types of performance?
• What about the patient outcomes?
Van Smeden et al., Stat Meth Med Res, 2019
IBS ROeS 2019, Lausanne MaartenvSmeden
Where do we stand on “ML” vs doctors?
Domain: radiology and pathology
• Article hits: 12,000
• After screening: 22
• Out-of-sample comparison “ML” vs doctors: 2
Faes et al., Lancet preprint, 2019, https://ssrn.com/abstract=3384923
IBS ROeS 2019, Lausanne MaartenvSmeden
Fair “ML” vs doctor comparisons
Three basic principles
• Doctors should work under realistic time constraints and have
access to all regular diagnostic information, including relevant
additional diagnostic testing, unless there are compelling
reasons not to do so
• The output generated by algorithms and physicians should be
evaluated on the same scale
• Performance over-optimism should be avoided
Van Smeden et al., JAMA, 2018, doi:
IBS ROeS 2019, Lausanne MaartenvSmeden
Fair “ML” vs doctor comparisons
Several barriers for diagnosis/prognosis
• Absence of a gold standard for most diseases1
• Errors/unclear category are to be expected
• Errors are transferred to algorithm
• Risk overestimating the performance
• Which performance measures should we be looking at?
• AUC, sens/spec, predictive values, F1?
• What about patient outcomes?
1See: Reitsma, Journal of Clinical Epidemiology, 2009, doi: 10.1016/j.jclinepi.2009.02.005
IBS ROeS 2019, Lausanne MaartenvSmeden
My plea
To big data (and use it) and back to trials
• There is a need to evaluate and compare the performance of
well developed statistical learning models on patient outcomes
(e.g. survival, response to treatment, PROs, etc.)
• The analogue of test-treatment trials in diagnostic research:
algorithm-treatment trials
IBS ROeS 2019, Lausanne MaartenvSmeden
IBS ROeS 2019, Lausanne MaartenvSmeden

Weitere ähnliche Inhalte

Was ist angesagt?

Was ist angesagt? (20)

Introduction to prediction modelling - Berlin 2018 - Part II
Introduction to prediction modelling - Berlin 2018 - Part IIIntroduction to prediction modelling - Berlin 2018 - Part II
Introduction to prediction modelling - Berlin 2018 - Part II
 
Uncertainty in AI
Uncertainty in AIUncertainty in AI
Uncertainty in AI
 
Prognosis-based medicine: merits and pitfalls of forecasting patient health
Prognosis-based medicine: merits and pitfalls of forecasting patient healthPrognosis-based medicine: merits and pitfalls of forecasting patient health
Prognosis-based medicine: merits and pitfalls of forecasting patient health
 
Five questions about artificial intelligence
Five questions about artificial intelligenceFive questions about artificial intelligence
Five questions about artificial intelligence
 
Correcting for missing data, measurement error and confounding
Correcting for missing data, measurement error and confoundingCorrecting for missing data, measurement error and confounding
Correcting for missing data, measurement error and confounding
 
Algorithm based medicine: old statistics wine in new machine learning bottles?
Algorithm based medicine: old statistics wine in new machine learning bottles?Algorithm based medicine: old statistics wine in new machine learning bottles?
Algorithm based medicine: old statistics wine in new machine learning bottles?
 
How to establish and evaluate clinical prediction models - Statswork
How to establish and evaluate clinical prediction models - StatsworkHow to establish and evaluate clinical prediction models - Statswork
How to establish and evaluate clinical prediction models - Statswork
 
Clinical prediction models: development, validation and beyond
Clinical prediction models:development, validation and beyondClinical prediction models:development, validation and beyond
Clinical prediction models: development, validation and beyond
 
Is it causal, is it prediction or is it neither?
Is it causal, is it prediction or is it neither?Is it causal, is it prediction or is it neither?
Is it causal, is it prediction or is it neither?
 
Predictimands
PredictimandsPredictimands
Predictimands
 
Guideline for high-quality diagnostic and prognostic applications of AI in he...
Guideline for high-quality diagnostic and prognostic applications of AI in he...Guideline for high-quality diagnostic and prognostic applications of AI in he...
Guideline for high-quality diagnostic and prognostic applications of AI in he...
 
Improving predictions: Lasso, Ridge and Stein's paradox
Improving predictions: Lasso, Ridge and Stein's paradoxImproving predictions: Lasso, Ridge and Stein's paradox
Improving predictions: Lasso, Ridge and Stein's paradox
 
Why the EPV≥10 sample size rule is rubbish and what to use instead
Why the EPV≥10 sample size rule is rubbish and what to use instead Why the EPV≥10 sample size rule is rubbish and what to use instead
Why the EPV≥10 sample size rule is rubbish and what to use instead
 
Data Visualization in Exploratory Data Analysis
Data Visualization in Exploratory Data AnalysisData Visualization in Exploratory Data Analysis
Data Visualization in Exploratory Data Analysis
 
Prediction, Big Data, and AI: Steyerberg, Basel Nov 1, 2019
Prediction, Big Data, and AI: Steyerberg, Basel Nov 1, 2019Prediction, Big Data, and AI: Steyerberg, Basel Nov 1, 2019
Prediction, Big Data, and AI: Steyerberg, Basel Nov 1, 2019
 
The basics of prediction modeling
The basics of prediction modeling The basics of prediction modeling
The basics of prediction modeling
 
Choosing Regression Models
Choosing Regression ModelsChoosing Regression Models
Choosing Regression Models
 
Exploratory data analysis
Exploratory data analysisExploratory data analysis
Exploratory data analysis
 
Measurement error in medical research
Measurement error in medical researchMeasurement error in medical research
Measurement error in medical research
 
Machine learning in medicine: calm down
Machine learning in medicine: calm downMachine learning in medicine: calm down
Machine learning in medicine: calm down
 

Ähnlich wie Machine learning versus traditional statistical modeling and medical doctors

Ähnlich wie Machine learning versus traditional statistical modeling and medical doctors (20)

Using Healthcare Data for Research @ The Hyve - Campus Party 2016
Using Healthcare Data for Research @ The Hyve - Campus Party 2016Using Healthcare Data for Research @ The Hyve - Campus Party 2016
Using Healthcare Data for Research @ The Hyve - Campus Party 2016
 
Big data in research: possibilities and pitfalls
Big data in research: possibilities and pitfallsBig data in research: possibilities and pitfalls
Big data in research: possibilities and pitfalls
 
Deep learning for biomedicine
Deep learning for biomedicineDeep learning for biomedicine
Deep learning for biomedicine
 
2016 ACS Semantic Approaches for Biochemical Knowledge Discovery
2016 ACS Semantic Approaches for Biochemical Knowledge Discovery2016 ACS Semantic Approaches for Biochemical Knowledge Discovery
2016 ACS Semantic Approaches for Biochemical Knowledge Discovery
 
Bias in covid 19 models
Bias in covid 19 modelsBias in covid 19 models
Bias in covid 19 models
 
Final APEC ERW 25 Aug 2022.pdf
Final APEC ERW 25 Aug 2022.pdfFinal APEC ERW 25 Aug 2022.pdf
Final APEC ERW 25 Aug 2022.pdf
 
IRJET- Develop Futuristic Prediction Regarding Details of Health System for H...
IRJET- Develop Futuristic Prediction Regarding Details of Health System for H...IRJET- Develop Futuristic Prediction Regarding Details of Health System for H...
IRJET- Develop Futuristic Prediction Regarding Details of Health System for H...
 
IVD Market Size and Growth Trend
IVD Market Size and Growth TrendIVD Market Size and Growth Trend
IVD Market Size and Growth Trend
 
How to compare typing techniques: do’s and Don’t’s
How to compare typing techniques:do’s and Don’t’sHow to compare typing techniques:do’s and Don’t’s
How to compare typing techniques: do’s and Don’t’s
 
Possibilities and pitfalls of AI in PICU
Possibilities and pitfalls of AI in PICUPossibilities and pitfalls of AI in PICU
Possibilities and pitfalls of AI in PICU
 
The big data challenge in healthcare and how can business intelligence best d...
The big data challenge in healthcare and how can business intelligence best d...The big data challenge in healthcare and how can business intelligence best d...
The big data challenge in healthcare and how can business intelligence best d...
 
InSTEDD: TED Prize Follow Up
InSTEDD: TED Prize Follow UpInSTEDD: TED Prize Follow Up
InSTEDD: TED Prize Follow Up
 
Developing and validating statistical models for clinical prediction and prog...
Developing and validating statistical models for clinical prediction and prog...Developing and validating statistical models for clinical prediction and prog...
Developing and validating statistical models for clinical prediction and prog...
 
Big data and machine learning: opportunità per la medicina di precisione e i ...
Big data and machine learning: opportunità per la medicina di precisione e i ...Big data and machine learning: opportunità per la medicina di precisione e i ...
Big data and machine learning: opportunità per la medicina di precisione e i ...
 
The absence of a gold standard: a measurement error problem
The absence of a gold standard: a measurement error problemThe absence of a gold standard: a measurement error problem
The absence of a gold standard: a measurement error problem
 
Journal for Clinical Studies: Close Cooperation Between Data Management and B...
Journal for Clinical Studies: Close Cooperation Between Data Management and B...Journal for Clinical Studies: Close Cooperation Between Data Management and B...
Journal for Clinical Studies: Close Cooperation Between Data Management and B...
 
ML, biomedical data & trust
ML, biomedical data & trustML, biomedical data & trust
ML, biomedical data & trust
 
Artificial Intelligence in Medicine.pdf
Artificial Intelligence in Medicine.pdfArtificial Intelligence in Medicine.pdf
Artificial Intelligence in Medicine.pdf
 
IRJET - Prediction and Analysis of Multiple Diseases using Machine Learni...
IRJET -  	  Prediction and Analysis of Multiple Diseases using Machine Learni...IRJET -  	  Prediction and Analysis of Multiple Diseases using Machine Learni...
IRJET - Prediction and Analysis of Multiple Diseases using Machine Learni...
 
IRJET- Cancer Disease Prediction using Machine Learning over Big Data
IRJET- Cancer Disease Prediction using Machine Learning over Big DataIRJET- Cancer Disease Prediction using Machine Learning over Big Data
IRJET- Cancer Disease Prediction using Machine Learning over Big Data
 

Mehr von Maarten van Smeden

Mehr von Maarten van Smeden (11)

UMC Utrecht AI Methods Lab
UMC Utrecht AI Methods LabUMC Utrecht AI Methods Lab
UMC Utrecht AI Methods Lab
 
A gentle introduction to AI for medicine
A gentle introduction to AI for medicineA gentle introduction to AI for medicine
A gentle introduction to AI for medicine
 
Associate professor lecture
Associate professor lectureAssociate professor lecture
Associate professor lecture
 
Algorithm based medicine
Algorithm based medicineAlgorithm based medicine
Algorithm based medicine
 
Clinical prediction models for covid-19: alarming results from a living syste...
Clinical prediction models for covid-19: alarming results from a living syste...Clinical prediction models for covid-19: alarming results from a living syste...
Clinical prediction models for covid-19: alarming results from a living syste...
 
Prediction models for diagnosis and prognosis related to COVID-19
Prediction models for diagnosis and prognosis related to COVID-19Prediction models for diagnosis and prognosis related to COVID-19
Prediction models for diagnosis and prognosis related to COVID-19
 
Living systematic reviews: now and in the future
Living systematic reviews: now and in the futureLiving systematic reviews: now and in the future
Living systematic reviews: now and in the future
 
Voorspelmodellen en COVID-19
Voorspelmodellen en COVID-19Voorspelmodellen en COVID-19
Voorspelmodellen en COVID-19
 
The statistics of the coronavirus
The statistics of the coronavirusThe statistics of the coronavirus
The statistics of the coronavirus
 
COVID-19 related prediction models for diagnosis and prognosis - a living sys...
COVID-19 related prediction models for diagnosis and prognosis - a living sys...COVID-19 related prediction models for diagnosis and prognosis - a living sys...
COVID-19 related prediction models for diagnosis and prognosis - a living sys...
 
Anatomy of a successful science thread
Anatomy of a successful science threadAnatomy of a successful science thread
Anatomy of a successful science thread
 

Kürzlich hochgeladen

SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptxSCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
RizalinePalanog2
 
biology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGYbiology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGY
1301aanya
 
Seismic Method Estimate velocity from seismic data.pptx
Seismic Method Estimate velocity from seismic  data.pptxSeismic Method Estimate velocity from seismic  data.pptx
Seismic Method Estimate velocity from seismic data.pptx
AlMamun560346
 
Introduction,importance and scope of horticulture.pptx
Introduction,importance and scope of horticulture.pptxIntroduction,importance and scope of horticulture.pptx
Introduction,importance and scope of horticulture.pptx
Bhagirath Gogikar
 
Pests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdf
PirithiRaju
 

Kürzlich hochgeladen (20)

Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.
 
Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verified
Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verifiedConnaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verified
Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verified
 
Forensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdfForensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdf
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
 
IDENTIFICATION OF THE LIVING- forensic medicine
IDENTIFICATION OF THE LIVING- forensic medicineIDENTIFICATION OF THE LIVING- forensic medicine
IDENTIFICATION OF THE LIVING- forensic medicine
 
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLKochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
 
STS-UNIT 4 CLIMATE CHANGE POWERPOINT PRESENTATION
STS-UNIT 4 CLIMATE CHANGE POWERPOINT PRESENTATIONSTS-UNIT 4 CLIMATE CHANGE POWERPOINT PRESENTATION
STS-UNIT 4 CLIMATE CHANGE POWERPOINT PRESENTATION
 
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptxSCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
 
GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)
 
Clean In Place(CIP).pptx .
Clean In Place(CIP).pptx                 .Clean In Place(CIP).pptx                 .
Clean In Place(CIP).pptx .
 
High Class Escorts in Hyderabad ₹7.5k Pick Up & Drop With Cash Payment 969456...
High Class Escorts in Hyderabad ₹7.5k Pick Up & Drop With Cash Payment 969456...High Class Escorts in Hyderabad ₹7.5k Pick Up & Drop With Cash Payment 969456...
High Class Escorts in Hyderabad ₹7.5k Pick Up & Drop With Cash Payment 969456...
 
biology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGYbiology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGY
 
Seismic Method Estimate velocity from seismic data.pptx
Seismic Method Estimate velocity from seismic  data.pptxSeismic Method Estimate velocity from seismic  data.pptx
Seismic Method Estimate velocity from seismic data.pptx
 
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
Vip profile Call Girls In Lonavala 9748763073 For Genuine Sex Service At Just...
 
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
 
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
 
Introduction,importance and scope of horticulture.pptx
Introduction,importance and scope of horticulture.pptxIntroduction,importance and scope of horticulture.pptx
Introduction,importance and scope of horticulture.pptx
 
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptxCOST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
 
Pests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdf
 
GBSN - Biochemistry (Unit 1)
GBSN - Biochemistry (Unit 1)GBSN - Biochemistry (Unit 1)
GBSN - Biochemistry (Unit 1)
 

Machine learning versus traditional statistical modeling and medical doctors

  • 1. Machine learning versus traditional statistical modeling and medical doctors Maarten van Smeden Leiden University Medical Center IBS ROeS - Lausanne September 10, 2019
  • 2. IBS ROeS 2019, Lausanne MaartenvSmeden Left out artificial intelligence? In medical research, “artificial intelligence” usually just means “machine learning” or “algorithm”
  • 3. IBS ROeS 2019, Lausanne MaartenvSmeden Tech company business model
  • 4. IBS ROeS 2019, Lausanne MaartenvSmeden Tech company business model https://bit.ly/2HSp8X5; https://bit.ly/2Z0Pfop; https://bit.ly/2KIcpHG; https://bit.ly/33IJhr9
  • 5. IBS ROeS 2019, Lausanne MaartenvSmeden Other success stories https://go.nature.com/2VG2hS7; https://bbc.in/2Z1drXQ
  • 6. IBS ROeS 2019, Lausanne MaartenvSmeden IBM Watson winning Jeopardy https://bbc.in/2TMvV8I
  • 7. IBS ROeS 2019, Lausanne MaartenvSmeden IBM Watson for oncology https://bit.ly/2LxiWGj
  • 8. IBS ROeS 2019, Lausanne MaartenvSmedenForsting, J Nuc Med, 2017, DOI: 10.2967/jnumed.117.190397
  • 9. IBS ROeS 2019, Lausanne MaartenvSmeden Machine learning everywhere (selection of last month) https://bit.ly/2ka0HLq; https://go.nature.com/33TQgO6; https://bit.ly/2kp6X23; https://bit.ly/2lZuKWt; https://bit.ly/2lI298g
  • 10. What are these Machine Learning methods?
  • 11. IBS ROeS 2019, Lausanne MaartenvSmeden “Everything is an ML method” https://bit.ly/2lEVn33
  • 12. IBS ROeS 2019, Lausanne MaartenvSmeden “ML methods come from computer science” https://bit.ly/2zhbwPv; https://stanford.io/2TVp1xK; https://stanford.io/2ZfED0k Leo Breiman Jerome H Friedman Trevor Hastie CART, random forest Gradient boosting Elements of statistical learning Education Physics/Math Physics Statistics Job title Professor of Statistics Professor of Statistics Professor of Statistics
  • 13. IBS ROeS 2019, Lausanne MaartenvSmeden “ML methods for prediction, statistics for explaining” Damen, BMJ, 2016, DOI:10.1136/bmj.i2416 363 developed models how many? Decision trees 0 Random forests 0 Support vector machines 0 Nearest neighbor algorithms 0 Neural networks 1
  • 14. IBS ROeS 2019, Lausanne MaartenvSmeden “ML methods for prediction, statistics for explaining” 1See further: Kreiff and Diaz Ordaz; https://bit.ly/2m1eYdK ML and causal inference, small selection1 • Superlearner (e.g. van der Laan) • High dimensional propensity scores (e.g. Schneeweiss) • The book of why (Pearl) Wednesday 10:40-12:10 Keynote Session 3 Els Goetghebeur: Plea for a marriage of machine learning and causal inference
  • 15. IBS ROeS 2019, Lausanne MaartenvSmeden Two cultures Breiman, Stat Sci, 2001, DOI: 10.1214/ss/1009213726
  • 16. IBS ROeS 2019, Lausanne MaartenvSmedenRobert Tibshirani: https://stanford.io/2zqEGfr Machine learning: large grant = $1,000,000 Statistics: large grant = $50,000
  • 17. IBS ROeS 2019, Lausanne MaartenvSmeden Statistics Machine learning Covariates Features Outcome variable Target Model Network, graphs Parameters Weights Model for discrete var. Classifier Model for continuous var. Regression Log-likelihood Loss Multinomial regression Softmax Measurement error Noise Subject/observation Sample/instance Dummy coding One-hot encoding Measurement invariance Concept drift Statistics Machine learning Prediction Supervised learning Latent variable modeling Unsupervised learning Fitting Learning Prediction error Error Sensitivity Recall Positive predictive value Precision Contingency table Confusion matrix Measurement error model Noise-aware ML Structural equation model Gaussian Bayesian network Gold standard Ground truth Derivation–validation Training–test Experiment A/B test Adapted from Daniel Obserski: https://bit.ly/2YN12Xf and Robert Tibshirani: https://stanford.io/2zqEGfr Language
  • 18. IBS ROeS 2019, Lausanne MaartenvSmeden ML refers to a culture, not to methods Distinguishing between statistics and machine learning • Substantial overlap methods used by both cultures • Substantial overlap analysis goals • Attempts to separate the two frequently result in disagreement Pragmatic approach: I’ll use “ML” to refer to models roughly outside of the traditional regression types of analysis: decision trees (and descendants), SVMs, neural networks, boosting etc.
  • 19. Examples where “ML” has done well
  • 20. IBS ROeS 2019, Lausanne MaartenvSmeden
  • 21. IBS ROeS 2019, Lausanne MaartenvSmeden Example: retinal disease Gulshan et al, JAMA, 2016, 10.1001/jama.2016.17216; Picture retinopathy: https://bit.ly/2kB3X2w Diabetic retinopathy Deep learning (= Neural network) • 128,000 images • Transfer learning (preinitialization) • Sensitivity and specificity > .90 • Estimated from training data
  • 22. IBS ROeS 2019, Lausanne MaartenvSmeden Example: lymph node metastases Bejnordi et al, JAMA, 2018, doi: 10.1001/jama.2017.14585. See our letter to the editor for a critical discussion: https://bit.ly/2kcYS0e Deep learning competition • 390 teams signed up, 23 submitted • Only 270 images for training • Test AUC range: 0.56 to 0.99
  • 23. IBS ROeS 2019, Lausanne MaartenvSmeden Deep learning on images Many similar studies and challenges in radiology, pathology, dermatology, opthalmology, gastroenterology, cardiology, …. Topol, Nature Medicine, 2019, DOI: 10.1038/s41591-018-0300-7
  • 24. IBS ROeS 2019, Lausanne MaartenvSmeden Other sources of “medical” data • Large scale gene expression data • e.g. diagnosis of acute myeloid leukemia https://bit.ly/2k8Ao8e • Prognostication by text mining electronic health records • e.g. predicting life expectancy https://bit.ly/2k8Ao8e • Analyzing social media posts • e.g. pharmacovigilance, adverse events monitoring via Twitter posts https://bit.ly/2m0KKrg
  • 25. Examples where “ML” has done poorly
  • 26. IBS ROeS 2019, Lausanne MaartenvSmeden Skin cancer and rulers Esteva et al., Nature, 2016, DOI: 10.1038/nature21056; https://bit.ly/2lE0vV0
  • 27. IBS ROeS 2019, Lausanne MaartenvSmeden Predicting mortality – the conclusion PlosOne, 2018, DOI: 10.1371/journal.pone.0202344
  • 28. IBS ROeS 2019, Lausanne MaartenvSmeden Predicting mortality – the results PlosOne, 2018, DOI: 10.1371/journal.pone.0202344
  • 29. IBS ROeS 2019, Lausanne MaartenvSmeden Predicting mortality – the media PlosOne, 2018, DOI: 10.1371/journal.pone.0202344; https://bit.ly/2Q6H41R; https://bit.ly/2m3RLrn
  • 30. IBS ROeS 2019, Lausanne MaartenvSmeden HYPE!
  • 31. IBS ROeS 2019, Lausanne MaartenvSmeden Systematic review clinical prediction models Christodoulou et al. Journal of Clinical Epidemiology, 2019, doi: 10.1016/j.jclinepi.2019.02.004
  • 33. IBS ROeS 2019, Lausanne MaartenvSmeden Comparison “ML” vs statistical models • Machine learning methods versus statistical models is a false dichotomy • Advanced “ML” shows promise, especially in areas that are not the traditional “tabular data” (e.g. images) • Tabular data settings where “ML” can be compared with traditional regression model techniques show little added value in medical applications
  • 34. IBS ROeS 2019, Lausanne MaartenvSmeden Classification versus risk prediction Most ML “classifiers” don’t come naturally with risk prediction, i.e. a probability estimate of predicted outcome for individuals • Possibly much large sample size needed to obtain reliable (calibrated) risk predictions1 than reliable classifications • Models can be trained to be optimized for a certain predictive performance (e.g. AUC, sensitivity, calibration) • Which performance to use to compare models are optimized for different types of performance? • What about the patient outcomes? Van Smeden et al., Stat Meth Med Res, 2019
  • 35. IBS ROeS 2019, Lausanne MaartenvSmeden Where do we stand on “ML” vs doctors? Domain: radiology and pathology • Article hits: 12,000 • After screening: 22 • Out-of-sample comparison “ML” vs doctors: 2 Faes et al., Lancet preprint, 2019, https://ssrn.com/abstract=3384923
  • 36. IBS ROeS 2019, Lausanne MaartenvSmeden Fair “ML” vs doctor comparisons Three basic principles • Doctors should work under realistic time constraints and have access to all regular diagnostic information, including relevant additional diagnostic testing, unless there are compelling reasons not to do so • The output generated by algorithms and physicians should be evaluated on the same scale • Performance over-optimism should be avoided Van Smeden et al., JAMA, 2018, doi:
  • 37. IBS ROeS 2019, Lausanne MaartenvSmeden Fair “ML” vs doctor comparisons Several barriers for diagnosis/prognosis • Absence of a gold standard for most diseases1 • Errors/unclear category are to be expected • Errors are transferred to algorithm • Risk overestimating the performance • Which performance measures should we be looking at? • AUC, sens/spec, predictive values, F1? • What about patient outcomes? 1See: Reitsma, Journal of Clinical Epidemiology, 2009, doi: 10.1016/j.jclinepi.2009.02.005
  • 38. IBS ROeS 2019, Lausanne MaartenvSmeden My plea To big data (and use it) and back to trials • There is a need to evaluate and compare the performance of well developed statistical learning models on patient outcomes (e.g. survival, response to treatment, PROs, etc.) • The analogue of test-treatment trials in diagnostic research: algorithm-treatment trials
  • 39. IBS ROeS 2019, Lausanne MaartenvSmeden
  • 40. IBS ROeS 2019, Lausanne MaartenvSmeden