International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056 Volume: 04 Issue: 3 | Mar -2017
www.irjet.net
p-ISSN: 2395-0072
IMPLEMENTATION OF BREAST TUMOR DETECTION USING NEAR INFRARED SENSOR AND LSVM ALGORITHM R.Anitha1, G.Manisha2, B.Pradeepa3, M.Abirami4 1,2,3Student,Dept
of Computer Science and Engineering,Panimalar Institute of Technology,Tamilnadu,India.
4Assistant
Professor,Dept of Computer Science and Engineering,Panimalar Institute of Technology,Tamilnadu,India.
--------------------------------------------------------------------------------------------------------------------------------------Abstract: The breast cancer can be detected by the
over ultrasound, MRI, and mammogram. However, it is
execution of automated tumor detection techniques on
not necessary to image a cancer to detect it. The
medical images. Some of the medical imaging techniques
biochemical spectroscopies of MRS and PET are used in
such as MRI, CT ,micro wave cannot detect tumor of size
cancer detection. PET includes biochemical sensitivity
below 3mm.This can be detected by using Near Infrared
principally through the uptake of radio-labeled glucose,
(NIR) imaging techniques and Lagrangian Support Vector
and MRS employs the detection of cancer-specific
Machine (LSVM) Algorithm. This is an non-invasive
chemicals, especially phosphoryl choline. MRI also
method of detecting tumors. The breast is enclosed within
utilizes the pharmacokinetics of gadolinium contrast
a skull which is very rigid, so any growth inside such a
agents for physiological parameters, blood flow, and
restricted space can cause problems. A breast tumor is a
permeability.
collection of abnormal cells in the breast. Breast tumors
absorption
can
non-cancerous
concentration available in the NIR window and many
(benign).Therefore, the Short Time Fourier Transform
extrinsic labeled organic compounds. NIR extrinsic labels
(STFT) filter is used.
are available for spectroscopic identification of cancer.
be
cancerous
(malignant)
or
NIR signals
employs of
the
blood,
optical water,
intrinsic
and
lipid
Spectroscopic methods will have a high signal-to-noise Keywords- Head phantom, Photo Detector, tumor
ratio
cells, NIR sensor, LSVM, DFT,STFT filter.
milliseconds, and provide detection of small amounts of
and
a
moderate
response
time,
i.e.,
1-3
cancer-detecting biochemicals or small changes therein. 1.INTRODUCTION 2.MODULE DESCRIPTION The optical breast cancer detection mainly
2.1.NIR SENSOR
focuses on spectroscopic nature of optical cancer detection and experimental results obtained using each
Near optical led is used as a tool for monitoring
technology. Most current clinical methodologies utilize
important hemodynamic parameters in tissue non
radar technology for cancer diagnosis,. An alternate
invansively .NIR can typically penetrate much further
approach is to examine the Biochemical and metabolic
into a sample than mid infrared radiation. Led offers
changes induced by cancer, either globally or locally.
great life time and spectral stability and reduced power
Although NIR techniques can be used to image thick
requirement. NIR optical LED diode have ceased
tissue, given the diffuse nature of light propagation, sub-
production of their 80 nm device and released new
millimeter spatial resolution is not attainable. Thus, the
generation of high efficiency 850 nm technology. Since
resolution of NIR is insufficient to expect increased
optical LED is used for security purpose, a LED with
sensitivity/specificity by morphological measurements
eye safety under IEC 60825-I standard is used.
Š 2017, IRJET
|
Impact Factor value: 5.181
|
ISO 9001:2008 Certified Journal
|
Page 774