A presentation on Artificial Neural Networks with special reference to the domain of Medical Science Compiled by : Tonmoy Bhagawati, DC2013MTC0033 Mtech 1st Semester,DBCET Specialization : Artificial Intelligence
Jan 17, 2015
A presentation onArtificial Neural Networks with special reference to the domain
of Medical Science
Compiled by :Tonmoy Bhagawati, DC2013MTC0033
Mtech 1st Semester,DBCETSpecialization : Artificial
Intelligence
Networks : An Introduction
One efficient way of solving
complex problems
is following the lemma “divide and
conquer”
Networks are one approach for achieving
this. All networks are characterized
by the following
components: a set of nodes,
and connections
between nodes.
The connections determine
the information
flow between nodes. They
can be unidirectiona
l and bidirectional
The interactions of nodes though
the connections
lead to a global
behaviour of the network. This global
behaviour is said to be emergent.
One type of network sees the nodes as
‘artificial neurons’. These are
called Artificial
neural networks
(ANNs).
Biological Neural NetworksNatural neurons receive signals through synapses located on the dendrites or membrane of the neuron.
When the signals received are strong enough (surpass a certain threshold), the neuron is activated and emits a signal though the axon.
This signal might be sent to another synapse, and might activate other neurons.
The complexity of real neurons is highly abstracted when modeling artificial neurons.
These basically consist of inputs (like synapses), which are multiplied by weights (strength of the
respective signals), and then computed by a mathematical function which determines the
activation of the neuron.
Artificial Neural Networks
The higher a weight of an artificial neuron is, the stronger the input which is multiplied by it will be.
Weights can also be negative, so we can say that the signal is inhibited by the negative weight. Depending on the weights, the computation of the neuron will be different.
By adjusting the weights of an artificial neuron we can obtain the output we want for specific inputs. But when we have an ANN of hundreds or thousands of neurons, it would be quite complicated to find by hand all the necessary weights.
we can find algorithms which can adjust the weights of the ANN in order to obtain the desired output from the network.
Learning or Training
ANN and the Medical Science
Experience is as important for an ANN as it is for man.
Treatment planning in medicine, radiotherapy, rehabilitation, etc. is
being done using ANN.
Mortality prediction by ANN in different medical situations can be
very helpful for hospital management.
ANN has a promising future in fundamental medical and
pharmaceutical research, medical education and surgical robotics.
SpecificsClinical diagnosis: Acute myocardial
infarction (AMI) was one of the earliest applications of
ANNs
Pulmonary embolism (PE) and back pain are two other areas where comparisons have been
made between diagnostic efficiencies of human experts
and ANN.
Pathology : picture processing ability of ANN makes it very
suitable for use in classification of
histology/cytology specimens.
Microbiology :Paralysis mass spectrometry (PMS) is a specialized area of
microbiology in which the potential of ANN has been
demonstrated.
Advantages and Drawbacks
Noise-Tolerance, Fault-Tolerance against Hardware
errors, Sensible classification of unknown input, Building own internal representation
Local Minima, Slow Training Process, Choice
of suitable network topology, Preprocessing
ConclusionArtificial neural network theory is derived from many disciplines including neuroscience,
psychology, mathematics, physics, engineering, computer science, philosophy, biology and linguistics. ANNs exploit the massively parallel local processing and distributed representation properties that are believed to exist in the brain. The primary intent of ANNs is to explore and reproduce human information processing tasks such as speech, vision, knowledge processing, motor control and especially, pattern matching.Though ANN is being tested in various fields of medicine, there remains a lot of room for its improvement and validation.
An extensive amount of information is currently available to clinical specialists, ranging from details of clinical symptoms to various types of biochemical data and outputs of imaging devices. Each type of data provides information that must be evaluated and assigned to a particular pathology during the diagnostic process. To streamline the diagnostic process in daily routine and avoid misdiagnosis, artificial intelligence methods (especially computer aided diagnosis and artificial neural networks) can be employed. These adaptive learning algorithms can handle diverse types of medical data and integrate them into categorized outputs. In this presentation, we briefly reviewed and discuss the philosophy, capabilities, and limitations of artificial neural networks in medical diagnosis through selected subjects.
References1. Importance of Artificial Neural Network in Medical Diagnosis disease like acute nephritis disease and heart
disease by Irfan Y. Khan, P.H. Zope, S.R. Suralkar Dept. of Ele. & Tele. SSBT’s college of Engg. & Tech, Bambhori, Jalgaon, India
2. Artificial neural network and medicine by Zulfuquar Hossain khan, Saroj Kant Mohapatra, Prafulla kumar Khodiar and S. N. Ragu Kumar. (1997)
3. Baxt WG. Use of an artificial neural network for the diagnosis of myocardial infarction. Ann Intern Med 1991
4. Artificial Intelligence : a Modern Approach 2nd edition by Russell and Norvig 5. Artificial Neural networks by b. Yegnanarayana 6. Artificial Neural Networks for Beginners by Carlos Gershenson 7. Artificial neural networks- the hot topic in recent pharmaceutical research by Abdul Althafi 8. Artificial Intelligence and Intelligent Systems by N.P. Padhy 9. Artificial neural networks Opening the black box by Judith E. Dayhoff and James M. DeLeo 10. Application of Neural Networks in Diagnosing Cancer Disease Using Demographic Data - 2010 International
Journal of Computer Applications (0975 – 8887) Volume 1 – No. 2611. http://en.wikipedia.org/wiki/Artificial_neural_network (accessed o n 11.12.13)