<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Arunadevi Ramesh on Medium]]></title>
        <description><![CDATA[Stories by Arunadevi Ramesh on Medium]]></description>
        <link>https://medium.com/@arunadeviramesh1011?source=rss-4c4e1917cda3------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*dX9S-oLUjVxoZEhmb7qfqw.jpeg</url>
            <title>Stories by Arunadevi Ramesh on Medium</title>
            <link>https://medium.com/@arunadeviramesh1011?source=rss-4c4e1917cda3------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 01:53:32 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@arunadeviramesh1011/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Detection of COVID-19 Using Convolutional Neural Network Algorithm]]></title>
            <link>https://medium.com/@arunadeviramesh1011/detection-of-covid-19-using-convolutional-neural-network-algorithm-bce0e7cbc800?source=rss-4c4e1917cda3------2</link>
            <guid isPermaLink="false">https://medium.com/p/bce0e7cbc800</guid>
            <category><![CDATA[detection]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[covid19]]></category>
            <category><![CDATA[cnn-algorithm]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Arunadevi Ramesh]]></dc:creator>
            <pubDate>Fri, 10 Jul 2020 07:24:05 GMT</pubDate>
            <atom:updated>2020-07-10T07:50:34.733Z</atom:updated>
            <content:encoded><![CDATA[<h3>Detection of COVID-19 Using Deep Learning (Convolutional Neural Network Algorithm)</h3><p><strong>Coronavirus disease 2019 (COVID-19)</strong> is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). It was first identified in December 2019 in <strong>Wuhan</strong>, the capital of China’s Hubei province. While the World Health Organization (WHO) announced a ‘mystery pneumonia’ on December 31, 2019, and it has spread globally, resulting in the ongoing 2019–20 coronavirus pandemic. Since then, the virus has been identified (SARS-CoV-2), the disease named (COVID-19), and there has been global spread.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Vka4LFDhCl42KAAkZzsxIQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZTXkybhlr3WOsHfSiGTDNA.jpeg" /><figcaption>First Image COVID-19 PERSON X-ray and Second one Normal X-ray</figcaption></figure><p>This idea is to overcome these circumstances using the Deep Learning technique for better and efficient treatment. Analysis of COVID-19 using deep learning includes lungs x-rays of patients. The problem is a binary classification problem where we classify Normal and COVID-19 cases.There are several pros and cons of using Deep Learning to tackle such kinds of situations:</p><ol><li>Pros: less expensive</li><li>Cons: Practically we need 100% accuracy as we can’t wrongly identify the patients as it might lead to further spread of disease which is highly discouraged.</li></ol><p><strong>COLLECTING THE DATASET</strong></p><p>Dataset consists of<strong> </strong>COVID-19 X-ray<strong> </strong>scan images. Dataset is taken from kaggle repository. Here is a <a href="https://www.kaggle.com/khoongweihao/covid19-xray-dataset-train-test-sets?">Dataset Link</a> for the COVID-19 Lung X-ray dataset which consists of 186 images — 93 images of each.</p><p><strong>MODEL ARCHITECTURE DEPLOYMENT</strong></p><p>Our model tend to have<em> </em><strong><em>3 hidden layers,</em></strong> you can experiment with more or fewer layers that is up to you. Our Proposed Model going to have <em>(150,150,3) </em>input neurons that are we are resizing our data to 150*150 <strong><em>with 3 channels</em> </strong>as it is considered to be the ideal size and therefore our model can grasp even minutiae and necessary features from the image.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/728/0*lAWVQi7AqhaVVeS5.png" /></figure><p>I am going to flatten our features and will use sigmoid as activation function as we are having<em> binary classification problem</em>, and thus our output will only contain one cell, RMSprop <strong><em>as optimizer </em></strong>works pretty well with<strong><em> sigmoid </em></strong>hence compiling the model with them in addition to <strong><em>cross binary entropy.</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/499/0*9r3Vxq1armPzfFqv.png" /></figure><p><strong>DATA AUGMENTATION</strong></p><p>Data augmentation is a way we can reduce the overfitting on models,where we increase the amount of training<strong> </strong>data using information only in our training data. The field of data augmentation is not new, and in fact, various data augmentation techniques have been applied to specific problems.</p><p>And here goes the code</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/868/0*4niCU9S2fwXutlPe.png" /></figure><p>So, we completed all the steps of construction and its time to train our model.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/903/0*E_vyzOZpKHBatvMY.png" /></figure><p>Now lets test a random image.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/723/0*Dm-fm4M0WZi9WCq_.png" /></figure><p>Our proposed methodology correctly predicted the image of the COVID !! The results we are getting is good but still, the accuracy can be improved in order to fulfill our intention.</p><h3>Source Code</h3><p>The whole source code can be found at my Github Repository:- <a href="https://github.com/ArunadeviRamesh/Detection-of-COVID-Deep-Learning-99.3-Accuracy-">Covid-19 detection source code</a></p><p>As always, thank you so much for reading, and please share this article if you found it useful!</p><p><strong><em>Stay home, Stay safe!!</em></strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bce0e7cbc800" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Content Based Image Retrieval Using Autoencoders ( vgg19 and convolutional neural network)]]></title>
            <link>https://medium.com/swlh/content-based-image-retrieval-using-autoencoders-vgg19-and-convolutional-neural-network-354a1773de02?source=rss-4c4e1917cda3------2</link>
            <guid isPermaLink="false">https://medium.com/p/354a1773de02</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[vgg19]]></category>
            <category><![CDATA[autoencoder]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[image-retrieval]]></category>
            <dc:creator><![CDATA[Arunadevi Ramesh]]></dc:creator>
            <pubDate>Fri, 10 Jul 2020 05:37:08 GMT</pubDate>
            <atom:updated>2020-07-10T07:57:13.618Z</atom:updated>
            <content:encoded><![CDATA[<p>In this article, Autoencoders can be used for finding similar images in an unlabeled image dataset. Autoencoders are neural networks comprised of both an encoder and a decoder. The goal is to compress your input data with the encoder, then decompress the encoded data with the decoder such that the output is a good reconstruction of your original input data.</p><p>This method work with a small training database of 5 common classes : tom, jerry, building, human faces and some food items. Content based image retrieval (CBIR) systems enable to find similar images to a query image among an image dataset. This article uses the deep learning framework to perform image retrieval unlabeled image dataset. This CBIR system will be based on a convolutional autoencoder. It is a class of unsupervised deep learning algorithms.we used three methods such as convAE, simpleAE and vgg16. Basically we first extract features from an image database and store it. Then compute the features associated with a query image. Finally Retrieve images with the closest features.</p><p>Usage of high RAM is only disadvantages in this model. But still, this model can return good accuracy.</p><p><strong>COLLECTING THE DATASET</strong></p><p>Dataset consists of 5 common item classes: tom, jerry, building, human faces and some food items.Here is a <a href="https://github.com/ArunadeviRamesh/Content-Based-Image-Retrieval-using-Autoencoders-Unsupervised-deep-learning-algorithms/tree/master/Dataset">Dataset </a>Link for the image retrieval dataset which consists of 57 images for training (Fig 1) and 7 images for testing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TR0ujYFmdzxonF7TeuzY9A.png" /><figcaption>Fig 1: Small Training Images</figcaption></figure><p><strong>MODEL ARCHITECTURE DEPLOYMENT</strong></p><p>We tend to have<em> </em><strong><em>3 hidden layers for encoded,</em></strong> you can experiment with more or fewer layers that is up to you. I am going to follow the traditional approach of increasing the neurons as we go deep inside the layer; as it helps to learn more features from the image which returns us better certainty.We are going to have <em>(100,100,3) </em>input neurons that are we are resizing our data to 100*100 <strong><em>with 3 channels</em> </strong>as it is considered to be the ideal size and therefore our model can grasp even minutiae and necessary features from the image.Next three hidden layers for decoded using encoded images.(Fig 2)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/802/1*_psJwy1GMqMAOX8wSbGSEQ.png" /><figcaption>Fig 2: Model Architecture Deployment</figcaption></figure><p><strong>K-NEAREST-NEIGHBOUR MODEL</strong></p><p>After training our autoencoder model using vgg19 or convolutional neural networks, Fit the training model using K-Nearest Neighbour model (Fig 3) because KNN is used to display top kth similar images.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/823/1*o0tjJiww2y48LX2E2fOiVw.png" /><figcaption>Fig 3: K-Nearest-Neighbour Model</figcaption></figure><p><strong>SIMILAR IMAGE RETRIEVAL FOR TESTING IMAGES</strong></p><p>After we trained our model, Now lets test a random image. A top two similar test image retrieval from the training images (Fig 4).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/301/1*Fa_NFuiODzt1oqJ7KN6erw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/301/1*riRnCh5RsdvDJl5qXmYfvQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/301/1*2B2DgSVMqNfNYSKac3AkgA.png" /><figcaption>Fig 4: Test image retrieval from the training set</figcaption></figure><h3>Source Code</h3><p>The whole source code can be found at my Github Repository:- <a href="https://github.com/ArunadeviRamesh/Content-Based-Image-Retrieval-using-Autoencoders-Unsupervised-deep-learning-algorithms">Image Retrieval using autoencoders source code</a></p><p>As always, thank you so much for reading, and please share this article if you found it useful!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=354a1773de02" width="1" height="1" alt=""><hr><p><a href="https://medium.com/swlh/content-based-image-retrieval-using-autoencoders-vgg19-and-convolutional-neural-network-354a1773de02">Content Based Image Retrieval Using Autoencoders ( vgg19 and convolutional neural network)</a> was originally published in <a href="https://medium.com/swlh">The Startup</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Predicting COVID-19 Death Using Linear Regression]]></title>
            <link>https://medium.com/swlh/predicting-covid-19-death-using-linear-regression-752b4dccedab?source=rss-4c4e1917cda3------2</link>
            <guid isPermaLink="false">https://medium.com/p/752b4dccedab</guid>
            <category><![CDATA[covid19]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[deep-neural-networks]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[linear-regression]]></category>
            <dc:creator><![CDATA[Arunadevi Ramesh]]></dc:creator>
            <pubDate>Thu, 25 Jun 2020 10:19:32 GMT</pubDate>
            <atom:updated>2020-07-06T20:12:34.363Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Predicting COVID-19 Death Using Machine Learning</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*2oiKtKciEPcDblTSrR4udA.jpeg" /></figure><p><strong>Coronavirus disease 2019 (COVID-19)</strong> is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). It was first identified in December 2019 in <strong>Wuhan</strong>, the capital of China’s Hubei province. While the World Health Organization (WHO) announced a ‘mystery pneumonia’ on December 31, 2019, and it has spread globally, resulting in the ongoing 2019–20 coronavirus pandemic. Since then, the virus has been identified (SARS-CoV-2), the disease named (COVID-19), and there has been global spread.</p><p>The severity of this disorder could be well established by the fact that while the first seven cases were reported on January 20, 2020, in the United States of America, by April 2020, it crossed over 300,000 cases. By May 9, 2020, more than 3.93 million COVID-19 cases were reported in 187 countries (and territories), resulting in about 274,000 deaths (WHO). Several countries, including India, have gone into a state of lockdown to prevent the spread of this deadly virus.</p><p>In this particular scenario, one primary thing needs to be done. Now We need Future death count prediction across world. Because to prevent the massive amount of death count across world. The open source global world COVID-19 dataset is collected from the WHO COVID-19 repository as a CSV file. Parameters have chosen three independent variables such as total_cases, recovery and death and one dependent variables( After one week Death count for COVID-19 across world).This one gave Highest accuracy and low error rate.</p><p>Classification algorithm is only used for discrete target values whereas regression models are used for continuous target values. Why Linear Regression is best to predict future death count across world? Because the Same Methodology is used some classification algorithm such as Logistic, Random Forest and Light GBM. Mean Absolute Error (MAE) is very important to choose best algorithm to predict the death count. But It gave higher accuracy, but high error rate using classification algorithms whereas It gave higher accuracy and also low error rate using Regression algorithm (Mean absolute Error is 30298 for Logistic Regression and 4411 for Linear Regression). Finally Regression Models is the best algorithm to predict the after one week cumulative death count across world.</p><p>First India dataset for COVID-19 is taken. In our proposed methodology, prediction of future death count for five states like Maharashtra, Madhya Pradesh, Gujarat, Delhi and Tamil Nadu in India. cumulative death count on June-29 is predicted based on cumulative confirmed cases, Recovery and cumulative death count on June-13. India will reach nearly 13949 cumulative death count on June 29, Tamil Nadu will reach nearly 929, Gujarat will reach nearly 1660, Maharashtra will reach nearly 6767, Madhya Pradesh will reach nearly 524 and Delhi will reach nearly 2425 cumulative death count on June -29. All those experimental results were predicted using Linear Regression algorithm based on our proposed Methodology.</p><p>Early forecast of the tracking system can help to take necessary actions. This article proposed to utilize the various data mining and machine learning models for COVID-19 epidemic analysis using world dataset from who website. <strong>The results show that <em>Linear Regression (LR) yielded a good accuracy compare to other approaches in forecasting the COVID-19 pandemic virus.</em></strong> Evaluating this experiment results using various error metrics such as Mean Absolute Error (MAE), R2_score, Mean Squared Error (MSE) and Root Mean Squared Error (RMSE).</p><p>Our proposed approach will predict cumulative number of death rate in various countries using the number of confirmed cases /active cases in each day. Developing a Graphical User Interface for our study to predict the number of death cases which will be helpful for Government and Health care sector (hospitals and pharmacy) to discover vaccinations at the earliest. The study can further be extended with multiple features to utilize other deep learning models.</p><p>As always, thank you so much for reading, and please share this article if you found it useful!</p><p><strong><em>Stay home, Stay safe!</em></strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=752b4dccedab" width="1" height="1" alt=""><hr><p><a href="https://medium.com/swlh/predicting-covid-19-death-using-linear-regression-752b4dccedab">Predicting COVID-19 Death Using Linear Regression</a> was originally published in <a href="https://medium.com/swlh">The Startup</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>