Real-Time Clinical Warning for Hospitalized Patients via Data Mining

• Clinical study has found that 4–17% of patients will undergo cardiopulmonary or respiratory arrest while in the GHW of hospital.
Project mission
• Sudden deteriorations (e.g. septic shock, cardiopulmonary or respiratory arrest) of GHW patients can often be severe and life threatening.
Comparison of various models
Method RPART SVM (Linear kernel SVM (Quadratic kernel SVM (Cubic kernel) SVM(RBF kernel Our method 5 AUC 0.6703 0.6879 0.6851 SPEC 0.93 0.9762 0.9675 SENS 0.55 0.3997 0.4028 PPV 0.287 0.4405 0.3676 NPV 0.977 0.9719 0.9718 ACCU 0.912 0.95033 0.94216
0.97345
0.9751 0.97871 0.97948 0.97992
0.92747
0.9293 0.93269 0.93342 0.93384
At specificity=0.95
1: bucketing + logistic regression
2: bucketing + logistic regression + bagging
Agenda
1 2 3
Background and overview
Early warning system (EWS)
Real-time data sensing (RDS)
5
Future work
Medical Record (34 vital signs: pulse, temperature, oxygen saturation, shock index, respirations, age, blood pressure …)
Exponential Moving Average (EMA)
Evaluation Criteria
AUC (Area Under receives operating characteristic (ROC) Curve) represents the probability that a randomly chosen positive example is correctly rated with greater suspicion than a randomly chosen negative example.
Bucket bagging
Logistic Regression No
Final Model
Converge? Yes Predict Model
EMA Smoothing
No > iteration count? Yes Final Model (A) Model building phase
> threshold? Yes Alarm Warning
Challenges: • Classification of highdimensional time series data • Irregular data gaps • measurement errors • class imba0 15000 10000 5000 0 Non-ICU ICU
3: bucketing + logistic regression + bucket bagging 4: bucketing + logistic regression + biased bucket bagging 5: bucketing + logistic regression + biased bucket bagging + exploratory undersampling
No
(B) Deployment phase
Data Preprocessing
Outlier removal
Normalization
Temporal Bucketing
Bucket 1
Bucket 2
Bucket 3
Bucket 4
Bucket 5
Bucket 6
We retain data in a sliding window of the last 24 hours and divided it evenly into 6 buckets In order to capture temporal variations, we compute several feature values for each bucket, including the minimum, maximum, and average
• A NIH-ICTS funded project: currently under clinical trials at Barnes-Jewish Hospital, St. Louis, MO
What exactly do we predict
Is he going to die?
What exactly do we predict
Workflow of the System
Data set D,T Real-time data stream Data Preprocessing Generate a 24-hour window
Bucketing Data Preprocessing Exploratory Undersampling Bucketing
• Goal: Provide early detection and intervention based on data mining
– to prevent these serious, often lifethreatening events. – Using both clinical data and wireless body sensor data
Overview of EWS
Goal: Design an data mining algorithm that can automatically
identify patients at risk of clinical deterioration based on their existing electronic medical records time-series.
Temporal Bucketing
Classification Algo.
Output Model, Threshold
Aggregated Bootstrapping (bagging)
Advantages: 1. Handles outliers 2. Avoid over-fitting 3. Better model quality
…………. ………….

Final Model
Biased Bucket Bagging
Bucketing
………….
………….

Final Model
Exploratory Undersampling
Predict model
Class balance
Remove the right record from the majority class according to the predicted value
Discriminative Classification
• Logistic regression (LR)
Clinical data
• Support vector machine (SVM)
Data preprocessing
• Use max, min, and avg of each bucket and each vital sign as the input features. (~ 400 features in total) • Use the training data to learn the model parameters.
Real-Time Clinical Warning for Hospitalized Patients via Data Mining （数据挖掘实现的住院病人的实时预警）
Department of Computer Science and Engineering
Yixin Chen （陈一昕）, Yi Mao, Minmin Chen, Rahav Dor, Greg Hackermann, Zhicheng Yang, Chengyang Lu
School of Medicine
Kelly Faulkner, Kevin Heard, Marin Kollef, Thomas Bailey
Background
• The ICU direct costs per day for survivors is between six and seven times those for non-ICU care. • Unlike patients at ICUs, general hospital wards (GHW) patients are not under extensive electronic monitoring and nurse care.