Heterogeneous Collection of

apricotpigletΤεχνίτη Νοημοσύνη και Ρομποτική

19 Οκτ 2013 (πριν από 4 χρόνια και 20 μέρες)

65 εμφανίσεις

Knowledge Systems Lab

JN
10/19/2013

Heterogeneous Collection of
Learning Systems for Confident
Pattern Recognition

Joshua R. New

Knowledge Systems Laboratory

Jacksonville State University

Knowledge Systems Lab

JN
10/19/2013

Outline


Motivation


Simplified Fuzzy ARTMAP (SFAM)


Interactive Learning Interface


System Demonstration


Conclusions and Future Work

Knowledge Systems Lab

JN
10/19/2013

Motivation

Knowledge Systems Lab

JN
10/19/2013

Motivation


Doctors and radiologists spend several
hours daily analyzing patient images (ie.
MRI scans of the brain)


The patterns being searched for in the
image are standard and well
-
known to
doctors


Why not have the doctor teach the
computer to find these patterns in the
images?

Knowledge Systems Lab

JN
10/19/2013

Motivation


Doctors and radiologists who use supervised
AI systems for image segmentation:


Usually can not interactively refine the computer’s
segmentation performance


Must be able to precisely select regions/pixels of
the image to train the computer


Often do not use an interface that facilitates
accomplishment of their task


Can easily lose where they are looking in the
image when using magnification

Knowledge Systems Lab

JN
10/19/2013

Simplified Fuzzy ARTMAP (SFAM)

Knowledge Systems Lab

JN
10/19/2013

SFAM


In order to “teach the computer” to find
tumors in neuro
-
images, a supervised
machine learning system must be used


Simplified Fuzzy ARTMAP (SFAM) is a neural
network that was created by Grossberg in
1987 and uses a mathematical model of the
way the human brain learns and encodes
information


This AI system was utilized because it allows
very fast learning for interactive training (ie.
seconds instead of days to weeks)

Knowledge Systems Lab

JN
10/19/2013

SFAM


SFAM is a computer
-
based system
capable of online, incremental learning


Two “vectors” are sent to this system for
learning:


Input feature vector gives the data is
available from which to learn


Supervisory signal indicates whether that
vector is an example or counterexample

Knowledge Systems Lab

JN
10/19/2013

SFAM


Data from which to learn


Feature vector from slice pixel values from shunted and
single
-
opponency images (Whole Brain Atlas)

Knowledge Systems Lab

JN
10/19/2013

SFAM


Vector
-
based graphic visualization of learning

Array of
Pixel Values


x

y

Category 1
-

2 members

Category 2
-

1 member

Category 4
-

3 members

0.35

0.90

Knowledge Systems Lab

JN
10/19/2013

SFAM


Only one tunable parameter


vigilance


Vigilance can be set from 0 to 1 and corresponds to the
generality by which things are classified

(ie. vig=0.3=>human, vig=0.6=>male, 0.9=>Joshua New)

0.675

0.75

0.825

Knowledge Systems Lab

JN
10/19/2013

SFAM


SFAM is sensitive to the order of the inputs

x

y

Category 1
-

2 members

Category 2
-

1 member

Category 4
-

3 members

Vector 3

Vector 1

Vector 2

Knowledge Systems Lab

JN
10/19/2013

SFAM


Voting scheme of 5 Heterogeneous
SFAM networks to overcome vigilance
and input order dependence


3 networks: random input order, set vigilance


2 networks: 3
rd

network order, vigilance
±

10%

Knowledge Systems Lab

JN
10/19/2013

SFAM

Knowledge Systems Lab

JN
10/19/2013

SFAM

Threshold results

Overlay results

Trans
-
slice results

Knowledge Systems Lab

JN
10/19/2013

Interactive Learning Interface


Screenshot of Segmentation & Features

Knowledge Systems Lab

JN
10/19/2013

System Demonstration

Knowledge Systems Lab

JN
10/19/2013

Conclusions


Doctors and radiologists can teach the
computer to recognize abnormal brain tissue


They can refine the learning systems results
interactively


They can precisely select targets/non
-
targets


They can zoom for precision while
maintaining context of the entire image


The interface developed facilitates task
performance through display of segmentation
results and interactive training

Knowledge Systems Lab

JN
10/19/2013

Future Work


Quantity of health
-
care can be increased by
utilizing these trained “agents” to allow
radiologists to only view the required images
and directing their attention for the ones that
are viewed


Quality of health care can be increased by
using the agents to classify an entire
database of images to highlight possibly
overlooked or misdiagnosed cases