Sepsis

Transcript

1. Automating Sepsis
   Diagnosis
   Stephen Thomas

   BMED 6517 - Spring 2019
   

   View Slide

2. 40 features
   8
   vital signs
   26
   lab
   results
   6
   dem
   ographic
   values
   40,336
   patients
   sepsis
   label
   70% missing data
   2932
   sepsis
   positive
   patients
   8 to 336 hourly
   measurements
   

   View Slide

3. Data Preparation
   40 features
   8
   vital signs
   26
   lab
   results
   6
   dem
   ographic
   values
   40,336
   patients
   sepsis
   label
   70% missing data
   2932
   sepsis
   positive
   patients
   8 to 336 hourly
   measurements
   12,237 patients
   28,099 patients
   2064
   sepsis
   positive
   patients
   30% Holdout
   For Testing
   Impute Temporal
   Data by Linear
   Interpolation/
   Extrapolation
   Balance
   Training Data by
   Undersampling
   Augment
   Temporal Data
   with Rates of
   Change
   Normalize and
   Zero-Fill
   34
   tem
   poral
   variables
   velocity
   acceleration
   Δ(X) Δ²(X)
   Training Set: 199,686 observations × 108 features
   Test Set: 469,332 observations × 108 features
   19,669
   sepsis
   positive
   labels
   

   View Slide

4. 30% Holdout
   40 features
   8
   vital signs
   26
   lab
   results
   6
   dem
   ographic
   values
   40,336
   patients
   sepsis
   label
   70% missing data
   2932
   sepsis
   positive
   patients
   8 to 336 hourly
   measurements
   12,237 patients
   28,099 patients
   2064
   sepsis
   positive
   patients
   

   View Slide

5. Impute Temporal Data
   12,237 patients
   28,099 patients
   2064
   sepsis
   positive
   patients
   Linear
   Interpolation &
   Extrapolation
   

   View Slide

6. Balance Training Data
   Undersample
   sepsis negative
   26,035 → 2064
   

   View Slide

7. Capture Temporal History
   34
   tem
   poral
   variables
   velocity
   acceleration
   Δ(X) Δ²(X)
   

   View Slide

8. Normalize / Zero-Fill
   Training Set: 199,686 observations × 108 features
   Test Set: 469,332 observations × 108 features
   19,669
   sepsis
   positive
   labels
   

   View Slide

9. Is Data Separable?
   

   View Slide

10. Is Data Separable?
    

    View Slide

11. LSTM Network Results
    Sepsis Negative Sepsis Positive
    Predicted Class
    True Class
    297
    1172
    571
    10197
    0 0.2 0.4 0.6 0.8 1
    False prediction rate
    0
    0.2
    0.4
    0.6
    0.8
    1
    True prediction rate
    Sepsis Negative
    Sepsis Positive
    

    View Slide

12. Results
    • All models very sensitive to overfitting
    
    • Problem is worthy of computational challenge
    Utility Notes
    Support Vector Machine 0.00
    Gaussian kernel, other hyper-parameters had
    minimal effect
    Random Forest 0.02
    Large leaf size and minimal splits to minimize
    overfitting
    
    Long Short-Term Memory
    Recurrent Neural Network
    0.35
    Limited epochs and relatively high learning rate to
    minimize overfitting
    

    View Slide