Vignan University

About: Vignan University is a education organization based out in Guntur, Andhra Pradesh, India. It is known for research contribution in the topics: Control theory & CMOS. The organization has 1138 authors who have published 1381 publications receiving 7798 citations.

Categorization of Thyroid Cancer Sonography Images Using an Amalgamation of Deep Learning Techniques

Abstract: Thyroid cancer is one of the most common types of cancer. For detecting this, it requires the development of new diagnostic tools. Its growing incidence results in increasing the burden of radiologists to detect and diagnose thyroid cancer cases based on ultrasound imaging. To reduce the analysis burden in the ultrasound images, we formed collaborative techniques that included two deep learning models, namely convolutional neural networks and transfer learning for the image-based analysis. The first model uses a six-layer convolutional architecture, whereas the second model uses a pre-trained VGGNet-16 architecture. We created a 6-CNN model with an efficient end-to-end analysis. This work has some set of existing images of thyroid ultrasonography which is an open-access resource for the scientific community. The dataset consists of malignant and benign thyroidal ultrasound pictures. By applying our model, we got an overall accuracy of 97.13 with the combination of CNN-VGGNet-16 techniques. The experimental results demonstrate that integrating image analysis information into a CNN leads to improved performance in thyroid cancer diagnosis.

Simulation-Based Comparison of Vampire Attacks on Traditional Manet Routing Protocols

Abstract: A mobile ad hoc network is wireless network without infrastructure and is self-configuring network of mobile nodes where nodes move to their places randomly, leave around or join the network. Over the many years, simulation work has been carried out to achieve a new reliable routing protocol. This paper seeks to rationalize Vampire attacks on the network layer that deals with route establishment. During route establishment and packet forwarding, Vampire attacks perpetually weaken the network by draining nodes’ energy. The Vampire attacks degrade the performance of routing protocol. All the basic traditional routing protocols those are examined here are open to Vampire attacks, which are damaging, and is time-consuming to troubleshoot. Simulation results show DSDV performs better than its counter parts.