automatically explaining machine learning prediction
play

Automatically Explaining Machine Learning Prediction Results: A Demonstration on Type 2 Diabetes Risk Prediction Gang Luo Department of Biomedical Informatics and Medical Education University of Washington luogang@uw.edu Clinical Big Data

0 downloads 2 Views 174 KB Size Report
  1. Automatically Explaining Machine Learning Prediction Results: A Demonstration on Type 2 Diabetes Risk Prediction Gang Luo Department of Biomedical Informatics and Medical Education University of Washington luogang@uw.edu

  2. Clinical Big Data • Volume of healthcare data – Increase 50-fold in 8 years to 25,000 petabytes by 2020 • Diverse sources – Electronic medical records – Sensors – Mobile devices • Opportunities to advance clinical care and biomedical research 2

  3. Predictive Modeling • Leverage these large, heterogeneous data sets to advance knowledge and foster discovery • Facilitate appropriate and timely care by forecasting – Health risk: Put high-risk patients into care management – Clinical course: Guide appropriate admission of bronchiolitis patients in the emergency department – Outcome: Assist with timely asthma diagnoses in children with clinically significant bronchiolitis 3

  4. Approaches to Predictive Modeling • Statistical methods – E.g., logistic regression • Machine learning algorithms that improve automatically through experience (model training) – E.g., support vector machine – Neural network – Decision tree – Random forest 4

  5. Outline • Pros and cons of machine learning • Our approach to address the challenge • Some results 5

  6. Pros of Machine Learning • Often achieves higher prediction accuracy than statistical methods – Sometimes doubles prediction accuracy • With less strict assumptions on data distribution 6

  7. Cons of Machine Learning • Use in healthcare is challenging • Most machine learning models give no explanation of prediction results – Most models are complex 7

  8. Cons of Machine Learning – Cont. • Explanation is essential for clinicians to – Trust prediction results – Determine appropriate, tailored interventions • E.g., provide transportation for patients who live far from their physicians and have difficulty accessing care – Defend their decisions in court if sued for medical negligence – Formulate new theories or hypotheses for biomedical research 8

  9. Challenge • Prediction accuracy and giving explanation of prediction results are frequently two conflicting goals • Need to achieve both goals simultaneously – Explain prediction results without sacrificing prediction accuracy 9

  10. Outline • Pros and cons of machine learning • Our approach to address the challenge • Some results 10

  11. Main Ideas • A model achieving high accuracy is usually complex and gives no explanation of prediction results • Challenge: Need to achieve high prediction accuracy as well as explain prediction results • Key idea: Separate prediction and explanation by using two models concurrently – The first model makes predictions and targets maximizing accuracy – The second model is rule-based • Used to explain the first model’s results rather than make predictions 11

  12. Main Ideas – Cont. • The rules used in the second model are mined directly from historical data • Use one or more rules to explain the prediction result for a patient • Suggest tailored interventions based on the reasons listed in the rules 12

  13. Outline • Pros and cons of machine learning • Our approach to address the challenge • Some results 13

  14. Some Results • Test case: Predicting type 2 diabetes diagnosis within the next year • Electronic medical record data of 10K patients • Can explain prediction results for 87% of patients who were correctly predicted by a champion machine learning model to have type 2 diabetes diagnosis within the next year 14

  15. Example Rule 1 • The patient had prescriptions of angiotensin- converting-enzyme (ACE) inhibitor in the past three years AND the patient’s maximum body mass index recorded in the past three years is ≥35  the patient will have type 2 diabetes diagnosis within the next year – ACE inhibitor is used mainly for treating hypertension and congestive heart failure – Obesity, hypertension, and congestive heart failure are known to correlate with type 2 diabetes • Example intervention: Enroll the patient in a weight loss program 15

  16. Example Rule 2 • The patient had prescriptions of loop diuretics in the past three years AND the patient had ≥ 23 diagnoses in total in the past three years  the patient will have type 2 diabetes diagnosis within the next year – Loop diuretics are used for treating hypertension – Hypertension and having a large number of diagnoses are known to correlate with type 2 diabetes 16

  17. Example Rule 3 • The patient had ≥6 diagnoses of hyperlipidemia in the past three years AND the patient had prescriptions of statins in the past three years AND the patient had ≥9 prescriptions in the past three years  the patient will have type 2 diabetes diagnosis within the next year – Hyperlipidemia: high lipid (fat) level in the blood – Statins are used for lowering cholesterol • Hyperlipidemia, high cholesterol level, and using many medications are known to correlate with type 2 diabetes 17

  18. Example Rule 4 • The patient had ≥5 diagnoses of hypertension in the past three years AND the patient had prescriptions of statins in the past three years AND the patient had ≥11 doctor visits in the past three years  the patient will have type 2 diabetes diagnosis within the next year – Hypertension, high cholesterol level, and frequent doctor visits are known to correlate with type 2 diabetes • Example intervention: Suggest the patient to make lifestyle changes to help lower his/her blood pressure 18

  19. Thank you 19

Recommend Documents


1
1 Dont Make Me Get Non -Linear! A

What is Machine Learning? A (Very Short) List of Applications Many different

decision tree learning
Decision tree learning Introduction to

INTRODUCTION TO MACHINE LEARNING Decision tree learning Introduction to

amr and machine learning
AMR and machine-learning Prediction of

AMR and machine-learning Prediction of AMR from metagenomes among other things

machine learning methods for metabolic pathway prediction
Machine Learning Methods for Metabolic

PathoLogic Machine Learning Methods for Prediction Evaluation Conclusions

introduction to machine learning
Introduction to Machine Learning

Introduction to Machine Learning COMPSCI 371D Machine Learning COMPSCI 371D

stop explaining black box machine learning models for
Stop Explaining Black Box Machine

Stop Explaining Black Box Machine Learning Models for High Stakes Decisions

stealing machine learning models via prediction apis
Stealing Machine Learning Models via

Stealing Machine Learning Models via Prediction APIs Florian Tramr, Fan

explaining complex energy systems
Explaining Complex Energy Systems A

Explaining Complex Energy Systems A Challenge Tackeling Climate Change with

gpu accelerated machine learning for bond price prediction
GPU Accelerated Machine Learning for

GPU Accelerated Machine Learning for Bond Price Prediction Venkat Bala

what can statistical machine translation teach neural
What can Statistical Machine

What can Statistical Machine Translation teach Neural Machine Translation

covid 19 pandemic
COVID-19 Pandemic ChABSA and AMHP

COVID-19 Pandemic ChABSA and AMHP Cyber Chapter Webinar March 10, 2020

gender differences in the use of cardiovascular disease
Gender Differences in the Use of

12 th International Congress on Drug Therapy in HIV Infection Gender

experience amp learning event
Experience & Learning Event

Biannual Regional Shared Experience & Learning Event September 18, 2014

vaccine and therapeutic initiatives for covid 19 an
Vaccine and Therapeutic Initiatives

Vaccine and Therapeutic Initiatives for COVID-19: an Investor Call with

what we will do look at what we know about covid 19 so
What we will do - look at what we know

What we will do - look at what we know about COVID-19 so far in relation to

national rural health resource center
National Rural Health Resource Center

National Rural Health Resource Center 2019 Delta Regional Community Health

extraction of proteins from palmaria palmata r sa j nsd
Extraction of proteins from Palmaria

Extraction of proteins from Palmaria palmata Rsa Jnsdttir Background

medical management of coronary artery disease focus on
Medical Management of Coronary Artery

Medical Management of Coronary Artery Disease Focus on Residual Risk Duane

public funding of research eu perspective
Public funding of research EU

Public funding of research EU perspective Presented by: Stefanie Prilla

eating amp swallowing deficits in individuals with
Eating & Swallowing Deficits in

2013/06/10 Eating & Swallowing Deficits in Individuals with Dementia

presenter disclosures
Presenter Disclosures Dr. Gordon Moe

Presenter Disclosures Dr. Gordon Moe Presenter Topic: Management of HFrEF: the

april 22 2020 6 00 7 00 pm
April 22, 2020 6:00-7:00 pm

April 22, 2020 6:00-7:00 pm Teleconference: (647) 951-8467 / Toll Free: 1