33
Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease Eugene Agichtein* , Elizabeth Buffalo , Dmitry Lagun, Allan Levey, Cecelia Manzanares, JongHo Shin, Stuart Zola Intellig ent Informat ion Emory Univers ity

Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

  • Upload
    fay

  • View
    40

  • Download
    0

Embed Size (px)

DESCRIPTION

Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease. Eugene Agichtein* , Elizabeth Buffalo , Dmitry Lagun , Allan Levey, Cecelia Manzanares, JongHo Shin, Stuart Zola . Emory University. Intelligent Information Access Lab. - PowerPoint PPT Presentation

Citation preview

Page 1: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

Automated Web-Based Behavioral Test for Early Detection of

Alzheimer’s Disease

Eugene Agichtein*, Elizabeth Buffalo, Dmitry Lagun, Allan Levey, Cecelia

Manzanares, JongHo Shin, Stuart Zola

Intelligent Information

Access Lab

Emory University

Page 2: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

2

Emory IR Lab: Research Directions

• Modeling collaborative content creation for information organization and indexing.

• Mining search behavior data to improve information finding.

• Medical applications of Search, NLP, behavior modeling.

Page 3: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

3

Mild Cognitive Impairment (MCI) and Alzheimer’s Disease

• Alzheimer’s disease (AD) affects more than 5M Americans, expected to grow in the coming decade

• Memory impairment (aMCI) indicates onset of AD (affects hippocampus first)

• Visual Paired Comparison (VPC) task: promising for early diagnosis of both MCI and AD before it is detectableby other means

Page 4: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

4

VPC: Familiarization Phase

Page 5: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

5

VPC: Delay Phase

Delay

Page 6: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

6

VPC: Test Phase

Page 7: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

7

VPC Task: Eye Tracking Equipment

Page 8: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

8

Subjects with Normal Visual Recognition Memory > 66% of time on Novel Images

Page 9: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

9

VPC: Low Performance Indicates Increased Risk for Alzheimer’s Disease

1. Detects onset earlier than ever before possible

2. Sets stage for intervention

Eugene Agichtein, Emory University

Page 10: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

10

Behavioral Performance on the VPC test is a Predictor of Cognitive Decline

Eugene Agichtein, Emory University

[Zola et al., AAIC 2012]

Scores on the VPC task accurately predicted, up to three years prior to a change in clinical diagnosis, MCI patients who would progress to AD, and Normal subjects who would progress to MCI

Page 11: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

11

VPC: Gaze Movement AnalysisLagun et al., Journal of Neuroscience Methods, 2011

Visual examination behavior in the VPC test phase. In this representative example, the familiar image is on the left (A), and the novel image is on the right (B), for a normal control subject. The detected gaze positions are indicated by blue circles, with the connecting lines indicating the ordering of the gaze positions.

Page 12: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

12

Technical Contribution: Eye Movement AnalysisLagun et al., Journal of Neuroscience Methods, 2011

Page 13: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

13

Significant Performance Improvements

Method Features Accuracy Sensitivity Specificity AUC

Baseline NP 0.667 0.6 0.734 0.667

LR NP+SO+RF+FD 0.71 0.712 0.707 0.71

SVM NP+SO+RF+FD 0.869* (+30%) 0.967* (+61%) 0.772* (+5%) 0.869* (+30%)

Lagun et al., Journal of Neuroscience Methods, 2011

Page 14: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

14

Our Big Idea: Web-based VPC task (VPW)with E. Buffalo, D. Lagun, S. Zola

• Web-based version of VPC without an eye tracker

• Can be administered anywhere in the world on any modern computer.

• Can adapt classification algorithms to automatically interpret the viewing data collected with VPW

Page 15: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

15

VPC-W Architecture

Page 16: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

16

VPC-W: basic prototype demo

Delay

ViewPort

position

Familiarization (identical images)

Test (novel image on left)

Page 17: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

17

Experiment Overview

• Step 1: Optimize VPC-W on (presumably) Normal Control (NC) subjects

• Step 2: Analyze VPC-W subject behavior with both gaze tracking and viewport tracking simultaneously

• Step 3: Validate VPC-W prediction on discriminating Impaired (MCI/AD) vs. NC

Page 18: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

18

VPC-W: Novelty Preference Preserved

Delay(seconds)

Mean novelty preference, VPC

(N=30)

Mean novelty preference, VPC-W

(N=34)10 67% 65%60 68% 69%

Self-reported elderly NC subjects tested with VPC-W over the internet exhibit similar novelty preference to that of VPC.

Single-factor ANOVA reveals no significant difference between VPC and VPC-W subjects

Page 19: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

19

VPC vs. VPC-W: Similar Areas of Interest

VPC ranking VPC-W ranking

Quantifying viewing similarity: Coarse measure: divide into 9 regions (3x3), rank by VPC and VPW viewing time. The Spearman rank correlation varies between 0.56 and 0.72 for different stimuli.

VPC VPC-W

Areas of attention: heat map for VPW (viewport-based) is concentrated in similar areas to VPC (unrestricted eye-tracking) .

Page 20: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

20

Actual Gaze vs. Viewport Position

Attention w.r.t. ViewPort

Page 21: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

21

Eye-Cursor Time Lag Analysis

XY: minimum at -75.00 ms 199.8578X:minimum at -90.00 ms 161.8480Y:minimum at -35.00 ms 116.3665

Page 22: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

22

Viewport Movement ~ Eye Movement

Normal elderly subject (NP=88%, novel image is on left). Impaired elderly subject (NP=49%, novel image is on left).

Page 23: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

23

Exploiting Viewport Movement Data

Novelty Preference

fixation duration distribution

+

Page 24: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

24

VPC-W Results: Detecting MCI

21 Subjects (11 NC, 10 aMCI), recruited @Emory ADRC:

Accuracy on the pilot data comparable to best reported values for manually administered cognitive assessment test (MC-FAQ, reported accuracy, specificity, and sensitivity of 0.83, 0.9, and 0.89 respectively) (Steenland et al., 2009).

Classification method

5-fold CV 10-fold CV leave-1-out

Acc. Sens. Spec. AUC Acc. Sens. Spec. AUC Acc. Sens. Spec. AUC

Baseline: NP>=0.58 0.81 0.80 0.82 0.81 0.81 0.80 0.82 0.81 0.81 0.80 0.82 0.81

SVM (VPC-W) 0.81 0.80 0.83 0.81 0.85 0.80 0.9 0.86 0.86 0.80 0.91 0.86Accuracy, Sensitivity, Specificity, and AUC (area under the ROC curve) for automatically classifying patients tested with VPC-W using 5-fold, 10-fold, and leave-one-out cross validation.

Page 25: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

25

Current Work

• Analysis:– Applying deep learning and “motif” analysis for more

accurate analysis of trajectory– Incorporating visual saliency signals

• Data collection:– Longitudinal tracking of subjects– “Test/Retest”: effects of repeated testing– Sensitivity analysis: for possible use in drug trials– Wide range of “normative” data using Mturk worker pool

Page 26: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

26

Future Directions and Collaboration Possibilities

• Can we apply similar or the same techniques for cost-effective and accessible detection of:– Autism (previous work on difference in gaze patterns)– ADHD– Stroke/Brain trauma– Other possibilities?

• What can we learn about the searcher from their natural search and browsing behavior?– Image search and examination preferences (anorexia)– Correlate behavior with biomarkers (Health 101 cohort)

Page 27: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

27

VPC-W Summary

VPC-W, administered over the internet, elicits viewing behavior in normal elderly subjects similar to eye tracking-based VPC task in the clinic.

Preliminary results show automatic identification of amnestic MCI subjects with accuracy comparable to best manually administered tests.

VPC-W and associated classification algorithms could facilitate cost-effective and widely accessible screening for memory loss with a simple log on to a computer.

Other potential applications: online detection and monitoring of ADD, ADHD, Autism and other neurological disorders.

This project has the potential to dramatically enhance the current practice of Alzheimer’s clinical and translational research.

Page 28: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

28

Eye Tracking for Interpreting Search Behavior

• Eye tracking gives information about searcher interests:– Eye position– Pupil diameter– Saccades and fixations

Reading

Search

Camera

Page 29: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

29

We Will Put an Eye Tracker on Every Table! - E. Agichtein, 2010

• Problem: eye tracking equipment is expensive and not widely available.

• Solution: infer searcher gaze position from searcher interactions.

Page 30: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

30

Inferring Gaze from Mouse Movements

Actual Eye-Mouse Coordination Predicted

No Coordination (35%)

Bookmarking (30%)

Eye follows mouse (35%)

Guo & Agichtein, CHI WIP 2010

Page 31: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

31

Post-click Page Examination Patterns

• Two basic patterns: “Reading” and “Scanning”– “Reading”: consuming or

verifying when (seemingly) relevant information is found

– “Scanning”: not yet found the relevant information, still in the process of visually searching

Page 32: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

32

Cursor Heapmaps (Reading vs. Scanning)[Task: “verizon helpline number”]

Relevant (dwell time: 30s) Not Relevant (dwell time: 30s)

Move cursor horizontally “reading”

Passively move cursor “scanning”

Page 33: Automated Web-Based Behavioral Test for Early Detection of Alzheimer’s Disease

33

Typical Viewing Behavior (Complex Patterns) [Task: “number of dead pixels to replace a Mac”]

Relevant (dwell time: 70s) Not Relevant (dwell time: 80s)

Actively move the cursor with pauses “reading” dominant

Keep the cursor still and scroll “scanning” dominant