Germination Project Data Science at Penn Medicine

Slide 1

Slide 1 text

Data Science and Predictive Healthcare Corey Chivers, PhD Senior Data Scientist @CjBayesian

Slide 2

Slide 2 text

2 Outline wWhat is Data Science? wBecoming a data scientist wTools of a data scientist wPredictive Healthcare at Penn Medicine

Slide 3

Slide 3 text

3 wWhat is Data Science?

Slide 4

Slide 4 text

4 wWhat is Data Science? wWhat do you need to study to become a Data Scientist?

Slide 5

Slide 5 text

5 Data Science Venn Diagram http://drewconway.com/zia/2013/3/26/the-data-science-venn-diagram

Slide 6

Slide 6 text

6 Popular Expectations

Slide 7

Slide 7 text

Slide 8

Slide 8 text

8 Using data and computation to give people superpowers

Slide 9

Slide 9 text

9 One Data Scientist’s Story w Undergraduate Chemistry McGill University

Slide 10

Slide 10 text

10 One Data Scientist’s Story w Undergraduate Chemistry Atmospheric Physics McGill University

Slide 11

Slide 11 text

11 One Data Scientist’s Story w Undergraduate Chemistry Atmospheric Physics Environmental Science McGill University

Slide 12

Slide 12 text

12 One Data Scientist’s Story w Undergraduate Chemistry Atmospheric Physics Environmental Science w Non-Profit research & activism around public space use in Toronto McGill University

Slide 13

Slide 13 text

13 One Data Scientist’s Story w Undergraduate Chemistry Atmospheric Physics Environmental Science w Non-Profit research & activism around public space use in Toronto w PhD Computational Biology McGill University

Slide 14

Slide 14 text

14 One Data Scientist’s Story w Undergraduate Chemistry Atmospheric Physics Environmental Science w Non-Profit research & activism around public space use in Toronto w PhD Computational Biology Modeling population dynamics for conservation McGill University

Slide 15

Slide 15 text

Slide 16

Slide 16 text

Slide 17

Slide 17 text

Slide 18

Slide 18 text

Slide 19

Slide 19 text

19 The medical record is full of data* *There’s a whole bunch that is not in the EHR, too! There are all kinds of patterns in there! Provides predictions to aid in clinical decision making

Slide 20

Slide 20 text

20 “Predictive Health Care” is the development and integration of Data Science* into clinical and operational processes and workflows to deliver better outcomes at lower cost. * Machine Learning, data visualization, computational simulation, AI, etc.

Slide 21

Slide 21 text

21 Cat Cat Cat Dog Dog Doge Machine Learning Generalizing from many examples

Slide 22

Slide 22 text

22 0.7 Dog 0.25 Cat 0.05 … Cat Dog Machine Learning Generalizing from many examples

Slide 23

Slide 23 text

23 Time Encounter with First Diagnosis of Valve Disease (‘ground truth’ ICD code) Patient Encounters Diagnoses (non-valve disease) Labs Etc. Input data Time Input data Positive Case Negative Case No encounters with diagnosis of Valve Disease (‘ground truth’ ICD code) Clinical Example

Slide 24

Slide 24 text

24 Penn Medicine Data Data Type 2011-Present estimated Total Vital (Manual) 20GB Vital (Telemetry) 1.5TB/year Labs 36 GB Meds and Orders 42 GB Radiology Raw 250TB Radiology Meta 200 GB Clinical Notes 250 GB/year Twitter 100TB Other social media (est.) 1TB Wearables (est.) 5GB/year Genomics 800TB Lots of examples to generalize!!

Slide 25

Slide 25 text

25 Data Visualization

Slide 26

Slide 26 text

26 Data Visualization

Slide 27

Slide 27 text

27 Data Visualization

Slide 28

Slide 28 text

28 Computational Simulation Using randomization to predict trajectories of complex dynamic systems Allows you to experiment with ‘what if’ scenarios and to select from alternative actions. ‘Writing down’ the hypothesized system dynamics also sharpens understanding and assumptions. (Chivers, 2014)

Slide 29

Slide 29 text

29 w Creating ‘Digital Twins’ that allow a team to explore efficiency gains • Build optimized scheduling templates • Test process changes to drive clinical workflow redesign • Identify bottlenecks w We’ve created digital twins to support Emergency Department and OB/GYN clinic operations Patient Volume Minute Census Wait Times OB/GYN Simulated Output Computational Simulation

Slide 30

Slide 30 text

30 Tools

Slide 31

Slide 31 text

31 Skills Successful Data Science is highly collaborative

Slide 32

Slide 32 text

32 As a ____________ ROLE when I’m ____________ CONTEXT if I knew ____________ INFORMATION I would do _____________ INTERVENTION to improve __________ OUTCOME Data Science “MadLibs” Solving the right problem

Slide 33

Slide 33 text

33 Mechanical Ventilation

Slide 34

Slide 34 text

34 If I knew when mechanical ventilator patients were ready for a Spontaneous Breathing Trial (SBT) Mechanical Ventilation

Slide 35

Slide 35 text

35 I would reduce sedation and initiate an SBT trial earlier to decrease how much time patients spend in the ICU Mechanical Ventilation

Slide 36

Slide 36 text

36 Time on Ventilation Mechanical Ventilation Time in ICU

Slide 37

Slide 37 text

37 Lung Connect

Slide 38

Slide 38 text

38 If I knew which lung cancer patients will go to the ED Lung Connect

Slide 39

Slide 39 text

39 I would address their symptoms in the outpatient setting to lower ED usage Lung Connect

Slide 40

Slide 40 text

40 flic.kr/p/aDbss E Lower ED Usage Lung Connect

Slide 41

Slide 41 text

41 https://www.pennmedicine.org/news/news-blog/2018/june/palliative-connect-digitizing-the-physicians-intuition-to-prompt-critical-conversations Palliative Connect

Slide 42

Slide 42 text

42 https://www.pennmedicine.org/news/news-blog/2018/june/palliative-connect-digitizing-the-physicians-intuition-to-prompt-critical-conversations If I knew which patients have serious, life-limiting illnesses Palliative Connect

Slide 43

Slide 43 text

43 I would make sure they receive a palliative care consultation to ensure the care team understands their goals and desires https://www.pennmedicine.org/news/news-blog/2018/june/palliative-connect-digitizing-the-physicians-intuition-to-prompt-critical-conversations Palliative Connect

Slide 44

Slide 44 text

44 Earlier access to palliative care More documented advanced care plans Palliative Connect

Slide 45

Slide 45 text

45 Some Challenges w Data != Reality • Healthcare data is particularly challenging – Non-random missingness – Largely unstructured – Clinical concepts and their representations change over time w How good does the model need to be? • We use meta decision theory to decide whether the model makes better decisions than some alternative w How do we know it’s working? • When you’re trying to prevent the thing you’re predicting, is your prediction bad or your intervention good?

Slide 46

Slide 46 text

46 Changing Clinical Concepts Weekly Count

Slide 47

Slide 47 text

47 How good does the model need to be? Data Scientist: Finally, I built a model from the data! Doc: Awesome, is it good? Data Scientist: I used tensorflow, so, ya. Doc: Lets deploy this thing!* *Any resemblance between these fictional characters and any persons, living or dead, is a purely coincidental

Slide 48

Slide 48 text

48 How good does the model need to be? Data Scientist: Finally, I built a model from the data! Doc: Awesome, is it good? Data Scientist: I used tensorflow, so, ya. Doc: Lets deploy this thing!* *Any resemblance between these fictional characters and any persons, living or dead, is a purely coincidental

Slide 49

Slide 49 text

49 All models are wrong, but some are useful. - George Box

Slide 50

Slide 50 text

50 FP FN TP TN

Slide 51

Slide 51 text

51 How good tho? How good tho? How bad tho? How bad tho? FP FN TP TN

Slide 52

Slide 52 text

52 Goodness can be measured in any units More adorbs Less adorbs

Slide 53

Slide 53 text

53 When outcomes are uncertain, the best decision is the one that has the highest expected goodness.

Slide 54

Slide 54 text

54 When outcomes are uncertain, the best decision is the one that has the highest expected goodness. Machine Learning can only help you with this part!

Slide 55

Slide 55 text

55 When outcomes are uncertain, the best decision is the one that has the highest expected goodness. Machine Learning can only help you with this part!

Slide 56

Slide 56 text

56 Healthcare Example Predicted Sepsis Treated a true case (Potential to avoid bad outcome) Predicted Sepsis Treated a false case (unnecessary) Predicted No Sepsis Didn’t treat (all good) Predicted No Sepsis Failed to treat (Bad outcome)

Slide 57

Slide 57 text

57 Healthcare Example

Slide 58

Slide 58 text

58 Treat none Cost of Intervention ($) Cost of event ($) Treat all Pregnancy Related Hypertension (PRH) is the leading cause of maternal morbidity and mortality in the U.S. High-risk patients à remote blood pressure monitoring https://healthcareinnovation.upenn.edu/projects/heart-safe-motherhood

Slide 59

Slide 59 text

59 Takeaways w Predictive Healthcare is the integration of Data Science into clinical and operational workflows to improve healthcare quality w Data Science is a collaborative, multidisciplinary endeavor • Using data to make better decisions • There are many paths to becoming a data scientist! w Taking time to ensure you’re solving the right problem is essential

Slide 60

Slide 60 text

60 Thanks!!