SOM Based Visualization Technique For Detection Of Cancerous Masses In Mammogram
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2011, Vol 2, Issue 9
Abstract
Breast cancer is the most common form of cancer in women. An intelligent computer-aided diagnosis system can be very helpful for radiologist in detecting and diagnosing micro calcifications patterns earlier and faster than typical screening programs. In this paper, we present a system based on gabor filter based enhancement technique and feature extraction techniques using texture based segmentation and SOM(Self Organization Map) which is a form of Artificial Neural Network(ANN) used to analyze the texture features extracted. SOM determines which texture feature has the ability to classify benign, malignant and normal cases. Watershed segmentation technique is used to classify cancerous region from the non cancerous region. We have investigated and analyzed a number of feature extraction techniques and found that a combination of ten features, such as Cor-relation, Cluster Prominence, Energy, Entropy, Homogeneity, Difference variance, Difference Entropy, Information Measure, and Normalized are calculated. These features gives the distribution of tonality information and was found to be the best combination to distinguish a benign micro calcification pattern from one that is malignant and normal. The system was developed on a Windows platform. It is an easy to use intelligent system that gives the user options to diagnose, detect, enlarge, zoom, and measure distances of areas in digital mammograms. Further Using Linear Filtering Technique and the Texture Features as Mask are convolved with the segmented image .The tumor is detected using the above method and using watershed segmentation, a fair segmentation is obtained The artificial neural network with unsupervised learning together with texture based approach leads to the accuracy and positive predictive value of each algorithm were used as the evaluation indicators. 121 records acquired from the breast cancer patients at the MIAS database. The results revealed that the accuracies of texture based unsupervised learning has 0.9534 (sensitivity 0.98716 and specificity 0.9582 which was detected thorough the ROC. The results showed that the gabor based unsupervised learning described in the present study was able to produce accurate results in the classification of breast cancer data and the classification rule identified was more acceptable and comprehensible.
Authors and Affiliations
S. Pitchumani Angayarkanni , Dr. V. Saravanan 2
Keywords
Related Articles
- EP ID EP102954
- DOI -
- Views 108
- Downloads 0
Ontological Model to Predict user Mobility
With the remarkable technological evolution of mobile devices, the use of computing resources has become possible at any time and independent of the geographical position of the user. This phenomenon has various names su...
A Two-Level Fault-Tolerance Technique for High Performance Computing Applications
Reliability is the biggest concern facing future extreme-scale, high performance computing (HPC) systems. Within the current generation of HPC systems, projections suggest that errors will occur with very high rates in f...
Optimal Compression of Medical Images
In today’s healthcare system, medical images are playing a vital role in the diagnosis. The challenges arise to the hospital management systems (HMS) are to store and communicate the large volume of medical images genera...
Analysis of k-Coverage in Wireless Sensor Networks
Recently, a concept of wireless sensor networks has attracted much attention due to its wide-range of potential applications. Wireless sensor networks also pose a number of challenging optimization problems. One of the f...
Internal Model Control of A Class of Continuous Linear Underactuated Systems
This paper presents an Internal Model Control (IMC) structure designed for a class of continuous linear underactuated systems. The study treats the case of Minimum Phase (MP) systems and those whose zero dynamics are not...