16

Nucleus and cytoplasm segmentation in microscopic images using K-

means clustering and region growing

Segmentation of leukocytes acts as the foundation for all automated

image-based hematological disease recognition systems. Most of the time,
hematologists are interested in evaluation of white blood cells only. Digital
image processing techniques can help them in their analysis and diagnosis.
Acquiring important information by peripheral blood cell differential counting
and, therefore, the microscopic review for correct patient diagnosis is an
exhaustive effort and requires extremely trained or qualified experts or
professionals. Automatic detection of WBCs in microscopic images as a
substitute for manually locating, identifying, and counting different classes of
cells is an important topic in the domain of cancer diagnosis. The main
objective of this paper is to detect leukocytes from a blood smear microscopic
image and segment them into their two dominant elements, nucleus and
cytoplasm. The segmentation is conducted using two stages of applying K-
means clustering. First, the nuclei are segmented using K-means clustering.
Then, a proposed method based on region growing is applied to separate the
connected nuclei. Next, the nuclei are subtracted from the original image.
Finally, the cytoplasm is segmented using the second stage of K-means
clustering.

The results indicate that the proposed method is able to extract the
nucleus and cytoplasm regions accurately and works well even though there is
no significant contrast between the components in the image. In this paper, a
method based on K-means clustering and region growing is proposed in order
to detect leukocytes from a blood smear microscopic image and segment its
components, the nucleus and the cytoplasm. As region growing step of the
algorithm relies on the information of edges, it will not able to separate the
connected nuclei more accurately in poor edges and it requires at least a weak
edge to exist between the nuclei. This method is efficiently used in medical
field for blood smear analysis.
 17

Acute leukemia classification by using SVM and K-Means

clustering

AML is one of the most common types of leukemia in adults, affecting

more than 10,000 adults that are diagnosed every year in the U.S. AML also
makes up 15 to 20 percent of childhood leukemia. About 500 children and
adolescents in the U.S. each year are affected by AML.The system takes as
input, Color images of stained peripheral blood smears and identifies the class
of each of the White Blood Cells (WBC). The process involves segmentation,
feature extraction and classification. Our work focuses on classification of Foil
of Bretagne (Lymphoid) and Almeida Lloyd (Myeloid). So that, physicians
can analyze, detect anomalies and ensure the diagnosis. The experiment results
showed that the performance of identification leukemia using our image
processing techniques could classify 100 sample images to lymphoid stem
cells and myeloid stem cells the method has been evaluated using K-Means
clustering. Features extracted from the segmented cytoplasm and nucleus, are
motivated by the visual cues of shape and texture. Various classifiers have
been explored on different combinations of feature sets. The results presented
here are based on trials conducted with normal cells. The highest performance
using SVM was of 92%. White blood cells (WBC) or leukocytes play a
significant role in the diagnosis of different diseases (including Leukemia),
and therefore, extracting information about that is valuable for hematologists.
Leukemia refers to a progressive, malignant disease of the blood-forming
organs.

Complete blood count process is the first step for leukemia screening.
The pathology is characterized by the uncontrolled accumulation of immature
white blood cells. The four main types of leukemia are Acute Myelogenous
Leukemia (AML); Acute Lymphoblastic Leukemia (ALL); Chronic Myeloid
Leukemia (CML); Chronic Lymphocytic Leukemia (CLL). In this paper, we
build up a decision support tool to improve classification of Acute
Myelogenous Leukemia (AML). AML is a fast-growing cancer of the blood
and bone marrow.
 18

Blue Brain is the name of the worlds first virtual brain. A Virtual machine is
one that can function as, a very appropriate application of an Artificial
Intelligence human brain. Reverse engineering is a foremost concept of
implementing the human brain and recreates it at the cellular level inside a
complete simulation. The four major motivations behind the Blue Brain
Technology are treatment of brain disfunctioning, scientific curiosity about
consciousness and human mind, a bottom up approach towards building
thinking machine and databases of all neuroscientific research results and
related past stories. There are three main steps 2to build the virtual brain are
data acquisition, simulation and visualization of results. The mission of
undertaking the Blue Brain technology is to gather all existing knowledge of
the brain to build a complete theoretical framework. The Blue Brain Project is
assumed to be the first one to explore about a true Artificial Intelligence via
the process of reverse engineering and also the effort to reverse engineering a
human brain.

On the Modeling of a Nano-Communication Network using Spiking

Neural Architecture

The human neural system is comprised of millions of neural cells

which are fully networked through synaptic connections. Each neuron receives
the input through dendrites from neighboring nodes, in order to fire an action
potential through its axon into the next neurons. In this paper, a biological
communication network is modeled and a scenario is implemented to detect
the damaged neural cells. For this purpose, the membrane potential of the
biological network nodes is monitored and evaluated using spiking neural
network algorithm. The supercomputer-based reconstructions and
simulations built by the project offer a radically new approach for
understanding the multilevel structure and function of the brain.

Computing with Spiking Neuron Networks

 19

Spiking Neuron Networks (SNNs) are often referred to as the

third generation of neural networks. Highly inspired from natural
computing in the brain and recent advances in neurosciences, they derive their
strength and interest from an accurate modeling of synaptic interactions
between neurons, taking into account the time of spike firing. SNNs
overcome the computational power of neural networks made of threshold or
sigmoidal units. Based on dynamic event-driven processing, they open up
new horizons for developing models with an exponential capacity of
memorizing and a strong ability to fast adaptation. Today, the main
challenge is to discover efficient learning rules that might take advantage
of the specific features of SNNs while keeping the nice properties (general-
purpose, easy-to-use, available simulators, etc.) of traditional connectionist
models. This paper presents the history of the spiking neuron,
summarizes the most currently-in-use models of neurons and synaptic
plasticity, the computational power of SNNs is addressed and the
problem of learning in networks of spiking neurons is tackled.

A Review The Blue Brain Technology Using Wetware Technology and

Fuzzy Logic

Today scientists are in research to create an artificial brain that can

think, response, take decision, and keep anything in memory. With the
advancement in technology, human, the ultimate source of information and
discovery should also be preserved. In other words, human is does not live for
thousands of years but the information in his mind could be saved and used for
several thousands of years. The technology helpful in this activity is Blue
Brain. The main aim is to upload human brain into machine. It can be used for
the development of the human society. The Blue Brain Project plans to reverse
engineer the human brain as a supercomputer simulation. The concept of
artificial neural network, fuzzy logic and the wetware technology is used.

Study on Blue Brain with its Application

 20

Today, scientists are in research to create an artificial brain that can

think, respond, take decision, and store anything in memory. The main aim of
this research is to upload human brain into machine. So that man can think and
take decision without any effort. After the death of the body, the virtual brain
will act as the man. So, even after the death of a person, it will not lose the
knowledge, intelligence, personalities, feelings and memories of that man that
can be used for the development of the human society. This paper present the
complete research work which explains the concept and functioning model of
blue brain and the recent research and developments in the process.
Technology has been progressing to a great extend such that even the human
brains are being created artificially through the science of artificial
intelligence. This data is used to build biologically realistic models of neurons
and networks of neurons in the cerebral cortex. The simulations are carried out
on a Blue Gene supercomputer built by IBM. IBM along with scientists at
BMI and EPFL Research University are involved in this simulation workflow.

Blue Brain

With the advancement in technology, human, the ultimate source of

information and discovery should also be preserved. The technology helpful in
this activity is Blue Brain. This journal paper consists of the information
on Blue Brain project, concepts of Blue Brain, its requirements,
strategies undertaken to build a Blue Brain, advantages and disadvantages and
many more. Typical scientists, impending research funders and scientific
journalists are still doubtful on success of mind uploading. Significant
mainstream research in related areas is being conducted in animal
brains, comparing, contrasting and simulation, developing of faster super
computers, virtual reality, brain-computer interfaces, connectors and
information extraction from dynamically working brains. Brain simulation is
unbelievably inter-disciplinary research.

Blue Brain Technology

 21

The research involves studying slices of living brain tissue using

microscopes and patch clamp electrodes. Data is collected about all the
many different neuron types. This data is used to build biologically realistic
models of neurons and networks of neurons in the cerebral cortex. The
simulations are carried out on a Blue Gene supercomputer built by IBM, hence
the name "Blue Brain". The research involves studying slices of living
brain tissue using microscopes and patch clamp electrodes. Data is
collected about all the many different neuron types. Human society is always
in need of such intelligence and such an intelligent brain to have with. But the
intelligence is lost along with the body after the death. The virtual brain is a
solution to it.

Reconstruction of Neural Circuits Using Serial Block-Face Scanning

Electron Microscopy

Electron microscopy is currently the only available technique with a

spatial resolution sufficient to identify fine neuronal processes and synaptic
structures in densely packed neuropil. For large-scale volume reconstruction
of neuronal connectivity, serial blockface scanning electron microscopy
allows us to acquire thousands of serial images in an automated fashion and
reconstruct neural circuits faster by reducing the alignment task. Here we
introduce the whole reconstruction procedure of synaptic network in the rat
hippocampal CA1 area and discuss technical issues to be resolved for
improving image quality and segmentation. Compared to the serial section
transmission electron microscopy, serial block-face scanning electron
microscopy produced much reliable three dimensional data sets and
accelerated reconstruction by reducing the need of alignment and distortion
adjustment. It involves the domains like brain imaging, neuroscience,
computer science, nanotechnology, AI, biotechnology, psychology,
philosophy, and many more.

Blue Brain: Bringing a Virtual Brain to Life

 22

The Blue Brain Project is the first complex project in which a

human brain is been portrayed in such a way that it helps us to
understand its function and dysfunction through detailed simulations. The
main aim is to upload human brain into a machine. So that man can think, take
decision without any effort. After the death of the body, the virtual brain will
act as the man .So, even after the death of a person it can be used for the
development of the human society. The project's novel research strategy
exploits interdependencies in the experimental data to obtain dense maps of
the brain, without measuring every detail of its multiple levels of
organization. The goal of the Blue Brain Project is to build biologically
detailed digital reconstructions and simulations of the rodent, and ultimately
the human brain.

Practical Applications of Spiking Neural Network in Information

Processing and Learning

Historically, much of the research effort to contemplate the neural

mechanisms involved in information processing in the brain has been
spent with neuronal circuits and synaptic organization, basically
neglecting the electrophysiological properties of the neurons. This paper
present instances of a practical application using spiking neurons and temporal
coding to process information, building a spiking neural network SNN to
perform a clustering task. The input is encoded by means of receptive fields.
The delay and weight adaptation uses a multiple synapse approach. Dividing
each synapse into sub-synapses, each one with a different fixed delay. The
delay selection is then performed by a Hebbian reinforcement learning
algorithm, also keeping resemblance with biological neural networks. It was
presented a practical applications of a neural network, built with more
biologically inspired neuron, to perform a real neurological signal
transmission through it.
 23

Toward Reverse Engineering the Brain: Modeling Abstractions and

Simulation Frameworks

Biological neural systems are well known for their robust and power-
efficient operation in highly noisy environments. Biological circuits are made
up of low-precision, unreliable and massively parallel neural elements with
highly reconfigurable and plastic connections. Recent research in spiking
neural networks has demonstrated interesting principles about learning and
neural computation. Understanding and applying these principles to practical
problems is only possible if large-scale spiking neural simulators can be
constructed. This paper reviews the modeling abstractions for neural circuits
and frameworks for modeling, simulating and analyzing spiking neural
networks. This paper focus on modeling abstractions, simulation frameworks
and hardware architectures for modeling brain circuitry.

Artificial Intelligence-A Literature Survey of Blue Brain

A man becomes intelligent because of brain. It is complex than any

other circuits in the world. Understanding the human brain is one of the
greatest challenges faced by the 21stcentury scientists. The main aim of this
paper is to describe and upload human brain intomachine. IBMs Blue Gene
supercomputer allows a quantum leap in the level of detail at which the brain
can be modeled. It is like uploading a mind in a computer. Mind uploading
probably can be achieved by either of two methods: 1. Copy and Transfer. 2.
Slow and steady replacement of neurons. This paper, includes the complete
research work which explains the concept and functioning model of blue
brain and the recent research and developments in the process. Data is
collected about all the many different neuron types. This data is used to
build biologically realistic models of neurons in the cerebral cortex. The
simulations are carried out on a Blue Gene supercomputer built by IBM.
Simulating the brains biological systems into a 3D model, which would
recreate the electrochemical interactions that take place inside the brain.
 24

A New User-Adapted Search Haptic Algorithm to Navigate along

Filiform Structures

The behavior of the brain and the behavior of brain simulations depend
to a large extend on the neural topology. Neural elements are organized in a
connected, dense, complex network of thread-like (i.e., filiform)structures.
The analysis of a computer-based simulation using just the visual modality is a
highly complex task due to the complexity of the neural topology and the large
amounts of multi-variable and multi-modal data generated by computer
simulations. This paper describes the use of haptic devices to aid in the
navigation along these neural structures, helping neurobiologists in the
analysis of neural network topologies. However, haptic navigation constrained
to complex filiform networks entails problems when these structures have high
frequency features, noise and/or complex branching nodes.

A parameter optimization method for Digital Spiking Silicon Neuron

model

DSSN model is a qualitative neuronal model designed for efficient

implementation in a digital arithmetic circuit. Previous studies extended this
model to support a wide variety of neuronal classes. Parameters of the DSSN
model were hand-fitted to reproduce neuronal activity precisely. This work
studied automatic parameter fitting procedure for the DSSN model. Optimized
parameters of the model by a GPU-based implementation of the differential
evolution algorithm in order to reproduce waveforms of the ionic-conductance
models and reduce necessary circuit resources for the implementation. A wide
variety of neuronal models have been used in silicon neuronal networks
because of the trade-off between reproducibility of neuronal activities
and computational efficiency. The research involves studying slices of living
brain tissue. It also confirmed the qualitative similarity between the ionic-
conductance model and the DSSN model by measuring statistics for the
spike timing.
 25

The Effects of HCN and KLT Ion Channels on Adaptation and

Refractoriness in a Stochastic Auditory Nerve Model

An accurate model of auditory nerve fibers may assist in developing

improved cochlear implant stimulation strategies. Previous studies have shown
that the original HodgkinHuxley (HH) model may be better at describing
nodes of Ranvier in ANFs than models for other mammalian axon types.
Recent physiological investigations of ANFs have shown the presence of a
number of ion channel types not considered in the previous modeling studies,
including low-threshold potassium (KLT) channels and hyper polarization-
activated cation (HCN) channels. This study investigates inclusion of these ion
channel types in a stochastic HH model of a single node of Ranvier. A single
node of Ranvier in a mammalian ANF was modeled. The node consisted of
four voltage-gated ion channel types, which are: fast sodium channels,
delayed-rectifier potassium channels, KLT channels and HCN channels.
Thereby eliminating the chances of any brain malfunctions such as psychiatric
disorders like depression and autism, which are possible in the normal brain.
IBM names this project as the Blue Brain project.

Blue Brain - The Future Generation

The research involves studying slices of living brain issue using

microscopes and patch clamp electrodes. Data is collected about all the many
different neuron types. This data is used to build biologically realistic models
of neurons and networks of neurons in the cerebral cortex. A basic unit of the
cerebral cortex is the cortical column. Each column can be mapped to one
function, e.g. in rats one column is devoted to each whisker. A rat cortical
column has about 10,000 neurons and is about the size of pinhead.
Techniques are being developed for multiscale simulation whereby active
parts of the brain are simulated in great detail while quiescent parts are not
so detailed. Every two weeks a column model is run. The simulations
reproduce observations that are seen in living neurons.
 26

A Review on Blue Brain: The future Artificial Brain

Blue brain is a concept which allows to copy or to transfer all the

contents of a human brain into a virtual brain that resides inside a Super
computer. The Super computer used in this is Blue Gene as of the
current information revealed. It is like uploading a mind in a computer.
Research investigating artificial brains plays three important roles in
science. 1) A study called cognitive neuroscience, which helps neuroscientists
to make an on-going attempt to understand how human brain works, 2) A
thought experiment in the philosophy of artificial intelligence (AI), which
demonstrates that it is actually possible to create a machine that has all the
capabilities of a human being in theory, 3) A serious long term project to
create machines capable of general intelligent action or Artificial General
Intelligence. Mind uploading can probably be achieved by either of two
methods, copy and Transfer or slow and steady replacement of neurons.

Blue Brain: An approach towards creation of virtual brain

Blue Brain is an attempt to reverse engineer the human brain and

recreate it at the cellular level inside a computer simulation. It is used
to study the brains architectural and functional principles and the
development of blue brain involves the merging of neural networks, computer
simulation and cognitive science. This paper reviews the blue brains
construction. The steps involved in its creation. The experiment was
conducted both on a PC and on the Celoxica board. With the advancement in
technology, human, the ultimate source of information and discovery should
also be preserved. Human dont live for thousands of years but the information
in its mind could be saved and used for several thousands of years. The
technology of Blue Brain is of great help in the preservation. This paper
discusses about the hardware and supercomputers being used to build a Blue
Brain.
 27

Large-Scale Brain Networks in Cognition: Emerging Methods and

Principles

An understanding of how the human brain produces cognition

ultimately depends on knowledge of large scale brain organization. Although
it has long been assumed that cognitive functions are attributable to the
isolated operations of single brain areas. This paper review the current
research on structural and functional brain organization, and argue that the
emerging science of large-scale brain networks provides a coherent framework
for understanding of cognition. Critically, this framework allows a principled
exploration of how cognitive functions emerge from, and are constrained by,
core structural and functional networks of the brain. This paper reviews the
use of basic terminology from network theory which provides a standardized
framework for comparison and contrast of current concepts of large-scale
brain networks emerging from the fields of neuro-anatomy and cognitive
electrophysiology.

A Survey Paper on Blue Brain Technology

The initial goal of the project was the simulation of a rat neocortical
column, which can be considered the smallest functional unit of the neocortex.
It involves collecting brain portions, taking them under a microscope,
and gauging the shape and electrical behavior of neurons individually.
This method of studying and cataloguing neurons is very familiar and
worldwide. Data is collected about all the many different neuron t ypes.
The neurons are captured by their shape, electrical and physiological activity,
site within the cerebral cortex, and their population density. These
observations are translated into precise algorithms which describe the
process, function, an positioning methods of neurons. Then, the algorithms
are used to generate biologically-real looking virtual neurons ready for
simulation.
 28

From Hodgkin Huxley Neurons to Visual Attention Networks in the

Brain

A Multi-Agent System (MAS) has been developed to simulate the

Visual Attention Network Model of the brain. Tasks are assigned to the
agents according to the Attention Network Theory developed by
neuroscientists. A sequential model based on simple objects has been
constructed, aiming to show the relations and the workflow between the
different visual attention networks. Each agent is being used as an
analogy to a role or function of the visual attention systems in the
brain. A numerical solution of Hodgkin Huxley equations is presented
in this paper to describe the behavior of a neuron. The solution is
illustrated by a graphical chart interface to finely tune the behavior of
the neuron visually. The neurons are characterized by the cellular membrane
with a capacitive element and a series of varying resistive elements.

Large-Scale Synthesis of Functional Spiking Neural Circuits

This paper reviews the theoretical and software tools used to construct
brain models that are capable of performing cognitive tasks. This tool set
allow to configure 2.5 million simple nonlinear components (neurons) with 60
billion connections between them (synapses) such that the resulting model can
perform eight different perceptual, motor, and cognitive tasks. This paper
describe the methodology and tools have developed for building large-scale
systems from simulated spiking neurons. To reverse-engineer the brain in this
way, a method is needed that shows how large numbers of simple components,
each of which receives thousands of inputs from other components, can be
organized to perform the desired computations. It is achieved through the
neural engineering framework (NEF), a mathematical theory that provides
methods for systematically generating biologically plausible spiking networks
to implement nonlinear and linear dynamical systems. It is also presented a
practical applications of a neural network.
 29

Time-Variant Modeling of Brain Processes

This paper gives a brief overview of current modeling strategies in

brain research. Also, the spatial scales ranging from single neuron to large
scale activity of and between brain regions are considered. The models are
mainly time-invariant. Three time variant modeling strategies, which enable a
model-based signal analysis, are described and applied to large scale signals.
The first is derived from adaptive filter theory and covers linear system and
linear as well as nonlinear process models. The second is based on modeled
brain source signals, i.e., the inverse problem must be solved. The third
strategy consists of a generalization of Dynamic Causal Modeling which is
frequently used for analysis of directed interactions between brain structures.

Supervised Learning in Spiking Neural Networks for Precise Temporal

Encoding

Precise spike timing as a means to encode information in neural

networks is biologically supported, and is advantageous over frequency-based
codes by processing input features on a much shorter time-scale. For these
reasons, much recent attention has been focused on the development of
supervised learning rules for spiking neural networks that utilise a temporal
coding scheme. However, despite significant progress in this area, there still
lack rules that have a theoretical basis, and yet can be considered biologically
relevant. The LIF neuron is a commonly used spiking neuron model, owing to
its relative simplicity and analytical tractability, and represents a special case
of the more general Spike Response Model. For these reasons, it is analysed
by considering a single postsynaptic neuron i with a membrane potential ui at
time t, defined by the simplified Spike Response Model. This paper addresses
the scarcity of existing learning rules for networks of spiking neurons that
have a theoretical basis, and which allow for the learning of multiple and
precisely-timed output spikes. It is based on exponentially filtered output spike
trains, to be a highly efficient, spike-based neural classifier.