You are on page 1of 15

DETECTION OF

ABNORMALITIES IN
BRAIN USING EXTREME
LEARNING MACHINE

BATCH NO : 7
VYSHNAVI KHANDE 160031497
AVUTHU LIKITHA 160030088
TULLIMALLI HARSHA SREE 160031385

GUIDED BY
DR.SANDEEP KUMAR SATAPATHY
INTRODUCTION
 As the abnormalities in brain differ, the
extreme learning machine technique gives
the better results with a maximum accuracy
rate.
 There are many methods where the dataset
is already provided and the outcomes are
predicted. But finding abnormalities through
the pixel image is being done in this project
OBJECTIVE
 The main objective of our project is to
process the MRI scan images and find the
abnormalities in brain using extreme learning
machine technique.
METHODOLOGY
 Image Processing
 Extreme Learning Machine
IMAGE PROCESSING
 Image processing is the analysis and
manipulation of digital images.
 It is a method to perform some operations on
the image in order to get an enhanced image
or extract some useful image through it.
 It takes image as input and output may be
characteristics of image of features
associated with image.
PACKAGES USED
 reshape2
 png
EXTREME LEARNING MACHINE
 Extreme learning machines are feed forward
neural network for classification, regression
and clustering with a single or multiple layer
of hidden nodes
 These hidden nodes can be randomly
assigned and never be updated (i.e. they
are random projection but with nonlinear
transforms), or can be inherited from their
ancestors without being changed
VARIANTS OF ELM
 Incremental ELM
 Error minimized ELM
 Evolutionary ELM
DATASET
 The data set is extracted using the MRI scan
images and applying image processing to it so
that the pixel values are extracted.
 These pixels are the values in the dataset.
DATA SET IMPLEMENTATION
download.file("http://www.med.harvard.edu/AANLIB/
cases/caseNN2/mr1/016.png",destfile = "hello.png")
install.packages("png")
library(png)
install.packages("reshape2")
library(reshape2)
x <- readPNG("hello.png")
dim(x)
x <- array(sapply(1:3, function(i) sample(0:9, 100,
replace=T)+i*10))
x
as.vector(x)
write.table(x,file="Sample4.csv",sep=",")
BASIC IMPLEMENTATION
 The implementation is done using the
package called elm which is the main
backbone to perform extreme learning
machine technique.
 Training and prediction functions are
provided for the extreme learning machine
algorithms. The ELM uses a single layer feed
forward neural network with random
generated weights and no gradient based
back propagation
REPOSITORY
 The repository of elm package is CRAN.
 It trains the extreme learning machine with
random values.
 The data is classified and binary values are
obtained.
OUTCOME
 The outcome from image processing of MRI
scan images i.e; the values in the dataset are
classified using extreme learning machine.
REFERENCES
 Huang, G., Huang, G.B., Song, S. and You,
K., 2015. Trends in extreme learning
machines: A review. Neural Networks, 61,
pp.32-48.
 Lama, R.K., Gwak, J., Park, J.S. and Lee,
S.W., 2017. Diagnosis of Alzheimer’s disease
based on structural MRI images using a
regularized extreme learning machine and
PCA features. Journal of healthcare
engineering, 2017.

You might also like