Minimization of the death proportions of lung cancer there is a requirement of a computer-aided diagnosis (CAD) system for early-stage detection and classification of lung nodules. This paper has proposed two novel approaches based on deep learning techniques improvement of the classification accuracy of lung nodules in computed tomography (CT) scans. Our first approach uses Two-dimensional Convolution Neural Network architecture for automatic feature extraction and classification of lung candidates as cancer nodules or noncancerous nodules. In the second approach, we have used the progressive scaling technique in which we have started with small size images(n x n) and train the model than we gradually increase the size of image(In the ratio of 2n x 2n) and train again this approach give promising result to increase classifier result. We have measured our approaches on the Lung Image Database Consortium image collection (LIDC/IDRI) dataset expected by the LUNA16 challenge. After experimenting with models we have created an android mobile application using the TensorFlowlite framework which is taking lung CT scan as an input and produce benign or malignant probability in the given scan. Experimental results proved that our deep learning architecture with a combination of cost-sensitive loss function and augmentation of minority class has produced an accuracy of 96.9%, the sensitivity of 96.2%, and specificity of 97.2 %. In the second approach of progressive resizing, we have graduated with an increase in the testing accuracy.
In recent years, Machine Learning techniques that are based on Deep Learning networks that show a great promise in research communities.Successful methods for deep learning involve Artificial Neural Networks and Machine Learning. Deep Learning solves severa problems in bioinformatics. Protein Structure Prediction is one of the most important fields that can be solving using Deep Learning approaches.These protein are categorized on basis of occurrence of amino acid patterns occur to extract the feature. In these paper aimed to review work based on protein structure prediction solve using Deep Learning Networks. Objective is to review motivate and facilitatethese deep learn the network for predicting protein sequences using Deep Learning.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.