Big data in the Cardiac ICU What can CNN can do for you? Kevin - - PowerPoint PPT Presentation

▶
big data in the cardiac icu
SMART_READER_LITE
LIVE PREVIEW

Big data in the Cardiac ICU What can CNN can do for you? Kevin - - PowerPoint PPT Presentation

Sep 10, 2023 313 likes •841 views

Big data in the Cardiac ICU What can CNN can do for you? Kevin Maher, MD Professor of Pediatrics, Emory University School of Medicine Director, Cardiac Intensive Care, Childrens Healthcare of Atlanta Medical Director, Pediatric

slide-1
SLIDE 1

Big data in the Cardiac ICU





What can CNN can do for you? 


Kevin Maher, MD Professor of Pediatrics, Emory University School of Medicine Director, Cardiac Intensive Care, Children’s Healthcare of Atlanta Medical Director, Pediatric Technology Center Georgia Institute of Technology

slide-2
SLIDE 2

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

  • No Disclosures
slide-3
SLIDE 3

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Overview

  • Clinical medicine and data in the CICU
  • CNN
  • Atlanta experience with CNN and CICU Data
slide-4
SLIDE 4

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-5
SLIDE 5

Quality of decisions

Critical Care Management/Decisions

Time

(+)

(-)

slide-6
SLIDE 6

Hospital Course

(+) (-)

Admission Discharge home

Death

health illness

slide-7
SLIDE 7

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Medical Decisions

  • History and Physical
  • Laboratory data
  • Radiologic data
  • Anatomy
  • Physiology
  • Operation
  • Current course/clinical state
  • Medications
  • Genetic data
  • Study results (echo, etc)
  • MD Experience
  • MD Knowledge (book)
slide-8
SLIDE 8

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-9
SLIDE 9

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-10
SLIDE 10

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-11
SLIDE 11

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Critical care data

  • About 150,000 waves of each type per day for each patient
  • Close to a million individual waves per patient, per day
  • In addition to laboratory, clinical diagnosis, radiologic, genetic,

historical, and interventional data

slide-12
SLIDE 12

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

ECMO support in Children

slide-13
SLIDE 13

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

25-30 patients

slide-14
SLIDE 14

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-15
SLIDE 15

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Critical Care Data

  • Too much data!
  • Most data is ignored
  • No way to visualize the data, understand or utilize
  • The subtleties of the waveforms that we simply don’t see
slide-16
SLIDE 16

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Overview

  • Clinical medicine and data in the CICU
  • CNN
  • Atlanta experience with CNN and CICU Data
slide-17
SLIDE 17

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Convolutional Neural Network

  • CNN is a deep learning algorithm that works well in image

analysis

  • “Computer Vision” , how images are evaluated, categorized,

identified

  • Does well with large data sets, and improves as data sets

increase in size

slide-18
SLIDE 18

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Consider a waveform as an image to be analyzed

slide-19
SLIDE 19

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Convolutional Neural Networks

“R” wave

Convolution: extracting features from the image using filters/kernels Pooling reduces the dimensions of the data

slide-20
SLIDE 20

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Overview

  • Clinical medicine and data in the CICU
  • CNN
  • Atlanta experience with CNN and CICU Data
slide-21
SLIDE 21

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Methods:

  • We conducted a binary classification for

differentiating whether a 30-sec ECG clips were from a ‘critical’ (post surgery)

  • r ‘healthy’ (ready to transfer step-

down) child.

  • 51 children into training set
  • 10 most recent children into test set.
  • We obtained about 253k valid training

ECG clips and 74k test clips per each

  • f the 3 leads.
slide-22
SLIDE 22

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Methods

A 64-layer 32 paths Convolution Neural Network (CNN)

slide-23
SLIDE 23

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Methods

  • Train CNN on each of the 8 waveforms
  • Waveforms include I, II, III, ABP

, RESP , Pleth, CVP , and SPO2

  • Each waveform is segmented to 30 seconds
  • Discard segmentations with missing values
  • I: 261M valid data points => contain 2 billion numerical values
  • II: 327M valid data points
  • III: 214M valid data points
slide-24
SLIDE 24

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Preliminary Results

  • Group 1 (ECG): CNN models separately trained on lead-I, II, III ECG

waveforms, and a unified model combining all the three leads.

  • Group 2: ML models for heart rate variability (including LR, DT and RF)
  • Group 3: ML models (including LR, DT and RF) trained on vital signs, or lab

results, or both.

  • Group 4 (combined): CNN model on ECGs combined with the ML models with

vital signs and lab results

slide-25
SLIDE 25

Comparison of models

slide-26
SLIDE 26

Results combined scores

Post operative day The combined scores are computed by taking an average of all the predicted scores from 3-Lead waveforms, vital signs and lab results.

slide-27
SLIDE 27

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Dynamic changes of patients’ embeddings in a 2-D projection space

Blue dots: A test patient’s embedding’s that change over time

Red dots: Post-op critical Green dots: Ready-to transfer to the floor

slide-28
SLIDE 28

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Post op day zero from Norwood operation, CICU

slide-29
SLIDE 29

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Post op day 1

slide-30
SLIDE 30

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Post op day 2

slide-31
SLIDE 31

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Post op day 3

slide-32
SLIDE 32

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Post op day 4

slide-33
SLIDE 33

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Post op day 8

slide-34
SLIDE 34

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Day 9 post op

slide-35
SLIDE 35

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Conclusions

  • Data, and knowledge exists in the critical care waveforms
  • Application of CNN to critical care data may enhance patient

monitoring and patient management

  • The initial work presented is undergoing extensive clinical

correlation and validation

slide-36
SLIDE 36

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Conclusions

  • Additional goals include a real time “barometer” of clinical

wellness

  • Opportunity for real time feed back for all medical decisions

being made

slide-37
SLIDE 37

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Research Team

  • Kevin Maher, MD CHOA/Emory
  • Alaa Aljiffry, MD CHOA/Emory
  • Yanbo Xu, PhD Georgia Tech
  • Jimeng Sun, PhD Georgia Tech
  • Siddharth Biswal, MS Georgia Tech
  • Shenda Hong, PhD Georgia Tech
  • IT Team

CHOA/GT

slide-38
SLIDE 38

Innovation and Collaboration:
 Advancing the Science and Treatment of Congenital Heart Disease

Thank you

slide-39
SLIDE 39

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-40
SLIDE 40

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-41
SLIDE 41

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

: A full trajectory of transfer prediction on a test patient who was transferred to step- down on Day 7.

A threshold line: Median of the prediction scores across the full predictive trajectory

slide-42
SLIDE 42

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-43
SLIDE 43

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Near Infrared Spectroscopy

slide-44
SLIDE 44

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Despite anticoagulation, thrombosis persists

slide-45
SLIDE 45

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-46
SLIDE 46

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

CNN results using ECG + BP

Critical Discharge Inpatient, stable arrest

2.Inpatient stable

  • 4. Critical

Current accuracy is 97% to distinguish these 4 groups ate the patient groups

slide-47
SLIDE 47

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

slide-48
SLIDE 48

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

A 5-second ECG waveform contains 5 * 250Hz = 1,250 numeric values. A 1-hour ECG waveform contains 3600 * 250Hz = 900,000 numeric values. A 1-day ECG waveform contains up to 24 * 3600 * 250Hz = 21M numeric values. An ICU stay could range from 5 days up to 28 days (in our study cohort). Our goal: To develop a comprehensive computational algorithm that can efficiently learn patients’ dynamic physiological status from continuous ECG waveforms.

Methods

slide-49
SLIDE 49

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Arterial wave forms

slide-50
SLIDE 50

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Methods (patient population)

  • Evaluate a cohort of patients with HLHS having

undergone the Norwood operation

  • (newborn infants, high risk open heart surgical

procedure)

  • Patients arriving from the operating room critically ill
  • The same cohort, when being transferred to the

general cardiology floor considered “healthy”

  • Evaluate continuous waveform data, lab, EMR, & vital

signs

slide-51
SLIDE 51

Children’s Healthcare of Atlanta, Emory University, Georgia Tech

Methods

We designed and trained a convolutional neural network algorithm

  • Trained on 250,000 of 30-second waveforms of 3-lead ECGs
  • critically ill (first day post op)
  • clinically well (ready for transfer to the floor)