Genetically Optimized Clustering Probability and Fuzzy Logic Based Clustering Approach in Wireless Sensor Network
Abstract:

Authors: Karishma Kothari, Piyush Sharma, Luv Sharma

File Name: Karishma_68000-18-107.pdf
 Download Abstract | Download Full Paper

Medical Image Fusion using Hybrid DCT-DWT based Approach
Abstract: During a clinical study or diagnosis, many sources of information are available to the clinician: different imaging modalities, and different tools to better study the structure or pathology. The clinician must synthesize these different pieces of information in order to perform accurate and reliable diagnosis and / or treatment. However, this synthesis can be long, tedious and remains highly dependent on the operator who takes care of it. In this context, image fusion appears as a new tool to help diagnosis; making the task easier for the doctor by providing a simpler fusion tool than mental fusion. As part of this work, the paper presents a comparative study of fusion techniques, the first one based on the discrete wavelet transform (DWT) and the second one based on discrete cosine transform (DCT) and the third one that is new based on the hybridization of DCT and DWT.

Authors: Shadma Amreen, Prof. Anuj Bhargava, Prof. Prashant Badal

File Name: Shadma_68000-18-104.pdf
 Download Abstract | Download Full Paper

Fusion based Hybrid Cooperative Spectrum Sensing in Cognitive Radio Framework
Abstract: Cognitive radio sensor network (CRSN) demands energy efficient and a cost effective cooperative spectrum sensing techniques which perform well in fading and shadowing environment. The work presented in this paper is the Fusion technique for hybrid cooperative spectrum sensing using AND fusion of three secondary users for decision making. Transmission by primary user is how many times detected at secondary users shows the probability of detection. The results of simulation shows that probability of detection improves using AND fusion for the three secondary users. Performance of proposed system is evaluated using SNR, probability of false alarm and Sensing Time.

Authors: Bhagya Shree Bhati, Dr. Mahesh Kumar Porwal

File Name: Bhagyashri_61000-18-103.pdf
 Download Abstract | Download Full Paper

Heart Disease Classification using PCA and Back-Propagation Neural Network
Abstract: Cardiovascular diseases are the leading cause of disability and premature death worldwide, and contribute substantially to the rising costs of health care. The fundamental anatomy pathological lesion is atherosclerosis, which occurs over the years and is usually advanced when symptoms appear, usually at maturity. Acute coronary and cerebrovascular events often occur suddenly and are often fatal before medical attention can be provided. It has been shown that the modification of risk factors reduces mortality and morbidity in people with cardiovascular diseases, diagnosed or not. The main objective of this research work is to develop a prototype which can determine and extract unknown knowledge (patterns and relations) related with heart disease from a past heart disease database record. This paper uses Neural Networks Algorithm technique for heart disease prediction. PCA is used to reduce number of attributes which indirectly reduces the no. of diagnosis tests which are needed to be taken by a patient. Performance of proposed approach is evaluated using confusion matrix plot.

Authors: Shivani Sawai, Aanchal Koul, Antara Rangnekar

File Name: Shivani_61000-18-108.pdf
 Download Abstract | Download Full Paper

A Diversity Recommendation Method Based on Product Marginal Utility
Abstract:

Authors: Wang Qian, Yu Jijun

File Name: Wang_61000-18-101.pdf
 Download Abstract | Download Full Paper