<?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 Suneet Jain on Medium]]></title>
        <description><![CDATA[Stories by Suneet Jain on Medium]]></description>
        <link>https://medium.com/@suneet.bhopal?source=rss-1426a2b7c04d------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*poAecYUbwMtde1ZwEDzRCA.jpeg</url>
            <title>Stories by Suneet Jain on Medium</title>
            <link>https://medium.com/@suneet.bhopal?source=rss-1426a2b7c04d------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 22:59:04 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@suneet.bhopal/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[Artificial Intelligence in Health Care]]></title>
            <link>https://medium.com/@suneet.bhopal/artificial-intelligence-in-health-care-8f73f6bb808c?source=rss-1426a2b7c04d------2</link>
            <guid isPermaLink="false">https://medium.com/p/8f73f6bb808c</guid>
            <category><![CDATA[health-care-reform]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Suneet Jain]]></dc:creator>
            <pubDate>Sat, 20 Nov 2021 05:37:37 GMT</pubDate>
            <atom:updated>2021-11-20T05:37:37.711Z</atom:updated>
            <content:encoded><![CDATA[<p>Hello readers! Today I am writing his blog to discuss with you the recent advances in the field of health care with Artificial Intelligence. We will be discussing how Artificial Intelligence when integrated with healthcare makes the process more convenient and increases efficiency, and later I’ll be discussing a recent research project I did on neonatal death prediction with machine learning-based classifiers.</p><p>What is Artificial Intelligence?</p><p>Before discussing the role of Artificial Intelligence in Health Care, we must understand the meaning of Artificial Intelligence, Artificial Intelligence or AI means human-like intelligence shown by computer systems or machines. AI contains the ability to solve problems and make decisions like humans with a similar or better efficiency than humans.</p><p>Machine learning is just another type of AI, it is the most commonly used form of AI and is the base for the majority of the AI systems out there. The most useful case of machine learning in healthcare is in the field of precision medicine, meaning being able to predict what kind of treatment would work best on a patient by looking at their test results and previous medical history.</p><p>History of Artificial Intelligence in Healthcare</p><p>In the 1960s and 1970s, the world’s first problem-solving system known as “Dendral” was created. It would help organic chemists identify organic compounds by using the already known knowledge of organic chemistry. This was one of the first utilization of Artificial Intelligence in the field of science and research developed by a few researchers at Stanford University.</p><p>Although “Dendral” was never directly used in healthcare it provided a base for the creation of “Mycin” in the mid-1970s. “Mycin” is considered one of the first uses of Artificial Intelligence in the field of medicine and healthcare. Its purpose was to identify bacteria that caused meningitis, other similar diseases, and then recommend antibiotics based on the patient’s body weight. Thus giving it the name “Mycin” as used in many antibiotics as the suffix “-mycin”. For example Azithromycin (used in throat infections).</p><p>Current day scenario of Artificial Intelligence in Healthcare</p><p>With the impact of COVID-19 across the globe many countries have started to see the importance of fast and efficient healthcare facilities, thus a lot of funds and budgets are being invested in research towards the use of Artificial Intelligence. According to a letter by CDC, the United States alone will invest almost $2 Billion in AI-related healthcare research over the next 5 years.</p><p>For example, in the field of dermatology (which is a field that deals mostly with images), image processing along with deep learning is being used vitally for identifying and labelling images with potential diseases. One important research is skin cancer detection from face photographs and lesion images.</p><p>Another use case for AI in health care is radiology, Machine Learning is being used to detect diseases in patients through Computerized Tomography (CT scan) and Magnetic Resonance Imaging (MRI). Which once approved for independent use, without human supervision, will reduce the workload on physicians and will independently detect and diagnose patients.</p><p>Neonatal and Infant Death Prediction with Machine Learning</p><p>In the past few months, I have worked with a renowned researcher Dr. Jasjit Suri on Neonatal (life less than 28 days) and Infant (life less than 1 year) Mortality prediction with machine learning-based AI systems. We trained 5 different machine learning-based classifiers on 15.8 million patients data from hospitals obtained with the help of the National Center for Health Statistics (NCHS). Using this research, we proved that machine learning can help predict Infant and Neonatal deaths. We also identified the risk factors that affect these deaths the most, this will help clinicians and doctors to help prevent these deaths by taking care of these risk factors. Thus, lowering Neonatal and Infant mortality rates and helping save lives.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8f73f6bb808c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data Preprocessing using Python]]></title>
            <link>https://medium.com/@suneet.bhopal/data-preprocessing-using-python-1bfee9268fb3?source=rss-1426a2b7c04d------2</link>
            <guid isPermaLink="false">https://medium.com/p/1bfee9268fb3</guid>
            <category><![CDATA[aiml]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[computer-science]]></category>
            <category><![CDATA[encoding]]></category>
            <category><![CDATA[data-preprocessing]]></category>
            <dc:creator><![CDATA[Suneet Jain]]></dc:creator>
            <pubDate>Sun, 15 Nov 2020 07:44:24 GMT</pubDate>
            <atom:updated>2021-07-30T15:26:57.790Z</atom:updated>
            <content:encoded><![CDATA[<p>This article will take you through the basic concepts of <strong>Data Preprocessing</strong> and implement them using <strong>python</strong>. We’ll be starting from the basics so if you have no prior knowledge about machine learning or data preprocessing, no need to worry!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Tax3Kk7aLW4MIzdmR8Ke1Q.jpeg" /></figure><p>Use the<strong> ipynb file available </strong><a href="https://github.com/devilsboy/DataPreprocessing"><strong>here</strong></a> to follow along on the implementation that I have performed below. Everything including the dataset is present in the repository.</p><p>Let’s begin!</p><h3><strong>What is Data Preprocessing?</strong></h3><p>Data Preprocessing is the process of making data suitable for use while training a machine learning model. The dataset initially provided for training might not be in a ready-to-use state, for e.g. it might not be formatted properly, or <strong>may contain missing or null values</strong>.</p><p>Solving all these problems using various methods is called Data Preprocessing, using a properly processed dataset while training will not only make life easier for you but also <strong>increase the efficiency and accuracy</strong> of your model.</p><h3>Steps in Data Preprocessing:</h3><p>In this article, We’ll be covering the following steps:</p><ul><li>Importing the libraries</li><li>Importing the dataset</li><li>Taking care of missing data</li><li>Encoding categorical data</li><li>Normalizing the data</li><li>Splitting the data into test and train</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*L6vivCP633xS1XAGXJOXlQ.jpeg" /><figcaption>Steps of data processing</figcaption></figure><blockquote><em>Want to read this story later? </em><a href="http://usejournal.com/app/signup?save_link=https://blog.usejournal.com/data-preprocessing-using-python-1bfee9268fb3"><em>Save it in Journal.</em></a></blockquote><h3>Step 1: Importing the libraries</h3><p>In the beginning, we’ll import three basic libraries which are very common in machine learning and will be used every time you train a model</p><ol><li>NumPy:- it is a library that allows us to work with arrays and as most machine learning models work on arrays NumPy makes it easier</li><li>matplotlib:- this library helps in plotting graphs and charts, which are very useful while showing the result of your model</li><li>Pandas:- pandas allows us to import our dataset and also creates a matrix of features containing the dependent and independent variable.</li></ol><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/dd13dc085cd968268e99f6df6ff02923/href">https://medium.com/media/dd13dc085cd968268e99f6df6ff02923/href</a></iframe><h3>Step 2: Importing the dataset</h3><p>The data that we’ll be using can be viewed and downloaded from <a href="https://github.com/devilsboy/DataPreprocessing/blob/main/Data.csv"><strong>here</strong></a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/361/1*tsIJxKZz-8dTM9DJgSd_Mw.png" /><figcaption>Sample dataset that we’ll be using</figcaption></figure><p>As you can see in the above image we are using a very simple dataset that contains information about customers who have purchased a particular product from a company.</p><p>It contains various information about the customers like their age, salary, country, etc. It also shows whether a particular customer has purchased the product or not.</p><p>It also contains a <strong>null value</strong> in the fifth row.</p><p>Let’s begin by importing the data.</p><p>As the given data is in CSV format, we’ll be using the <em>read_csv </em>function from the pandas<em> </em>library.</p><p>Now we’ll show the imported data. You must remember the data imported using the <em>read_csv</em> function is in a Data Frame format, we’ll later convert it into <em>NumPy arrays </em>to perform other operations and training.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/8af46e4bfe78be11564e1cfeeb4c79d9/href">https://medium.com/media/8af46e4bfe78be11564e1cfeeb4c79d9/href</a></iframe><p>In any dataset used for machine learning, there are two types of variables:</p><ul><li>Independent variable</li><li>Dependent variable</li></ul><p>The <strong>independent variable</strong> is the columns that we are going to use to predict the <strong>dependent variable</strong>, or<strong> </strong>in other words, the independent variable <strong>affects</strong> the dependent variable</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/366/0*PLqUB4waP0M4RIuA" /><figcaption>Independent and Dependent variable</figcaption></figure><p>In our dataset, the <em>country</em>, <em>age</em>, and <em>salar</em>y column are the <strong>independent variable</strong> and will be used to predict the <em>purchased</em> column which is the <strong>dependent variable.</strong></p><h3>Step 3: Handling the missing values</h3><p>As you can see in our dataset we have two missing values one in the <em>Salary </em>column in the 5th Row and another in the <em>Age </em>column of the 7th row.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/384/1*vPSKrycyj4pVFM8aGI12UA.jpeg" /><figcaption>Missing Values</figcaption></figure><p>Now there are multiple ways to handle missing values, one of them is to ignore them and delete the entire entry/row, this is commonly done in datasets containing a very large number of entries, where the missing values only constitute 0.1% of the total data. Thus they affect the model negligibly and can be removed.</p><p>But in our case, the dataset is very small and we cannot just ignore those rows. So we use another method, in which we take the mean of the entire column containing the missing values(in our case the <em>age </em>or <em>salary</em> column) and replace the missing values with that mean.</p><p>To perform this process we will use <strong>SimpleImputer </strong>class<strong> </strong>from the <strong>ScikitLearn </strong>library</p><p>Code for the Python implementation is given below:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/491c3cb61aaf5f87625397174f8a3c1f/href">https://medium.com/media/491c3cb61aaf5f87625397174f8a3c1f/href</a></iframe><p>Here the <em>“missing_values = np.nan” </em>means that we are replacing missing values and “strategy = ‘mean’ ” means that we are replacing the missing value with the mean of that column.</p><p>You can see that we have only selected the column with numerical data, as the mean can only be calculated on numerical data.</p><p>After running the above code you’ll get the following output:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*Q-tp3pJd65TuFnQNPQPerQ.png" /><figcaption>Output after replacing missing values with mean</figcaption></figure><p>As you can observe all the missing values have been replaced by the mean of the column.</p><h3>Step 4: Encoding categorical data</h3><p>In our case, we have two categorical columns, the <em>country</em> column, and the <em>purchased</em> column.</p><ul><li><strong>OneHot Encoding</strong></li></ul><p>In the <em>country</em> column, we have three different categories: France, Germany, Spain. We can simply label France as 0, Germany as 1, and Spain as 2 but doing this might lead our machine learning model to interpret that there is some correlation between these numbers and the outcome.</p><p>So to avoid this, we apply <strong>OneHot Encoding</strong></p><p>OneHot Encoding consists of turning the country column into three separate columns, each column consists of 0s and 1s. Therefore each country will have a unique vector/code and no correlation between the vectors and outcome can be formed.</p><p>You’ll understand more about it when we implement it below:</p><p>To perform this encoding we use <strong>OneHotEncoder </strong>and <strong>ColumnTransformer </strong>class from the same <strong>ScikitLearn </strong>library.</p><p>The ColumnTransformer class allows us to select the column to apply encoding on and leave the other columns untouched.</p><p>Note: The new columns created will be added in the front of the data frame and the original column will be deleted.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/d09270317057e63b3c28fd47719624ed/href">https://medium.com/media/d09270317057e63b3c28fd47719624ed/href</a></iframe><p>After performing the above implementation you’ll get the following output:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/327/1*K4dIRtg4SHAIkFkovlemOA.png" /><figcaption>New columns created after OneHot encoding</figcaption></figure><p>Now we can see that each country has got a unique vector or code, for example, France is 1 0 0, Spain 0 0 1, and Germany 0 1 0.</p><ul><li><strong>Label Encoding</strong></li></ul><p>In the last column, i.e. the purchased column, the data is in binary form meaning that there are only two outcomes either Yes or No. Therefore here we need to perform Label Encoding.</p><p>In this case, we use <strong>LabelEncoder</strong> class from the same <strong>ScikitLearn </strong>library.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/fe78dcfd93065f18b803ca58e97065c9/href">https://medium.com/media/fe78dcfd93065f18b803ca58e97065c9/href</a></iframe><p>We use ‘<em>data.iloc[:,-1]</em>’ to select the index of the column we are transforming.</p><p>After performing this our data will look something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/311/1*yT_kyFqHGPISrSZCVi3faA.png" /><figcaption>Label Encoding</figcaption></figure><p>As you can see the <em>purchased</em> column has been successfully transformed.</p><p>Now we have completed the encoding of all the categorical data in our dataset and can move to the next step.</p><h3>Step 5: Normalizing the dataset</h3><p>Feature scaling is bringing all of the features on the dataset to the same scale, this is necessary while training a machine learning model because in some cases the <strong>dominant features become so dominant that the other ordinary features are not even considered by the model</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/281/1*vXYUFARONQOxwfLG4DKx1Q.jpeg" /></figure><p>When we normalize the dataset it brings the value of all the features between 0 and 1 so that all the columns are in the same range, and thus there is no dominant feature.</p><p>Now to normalize the dataset we use <strong>MinMaxScaler</strong> class from the same <strong>ScikitLearn </strong>library.</p><p>The implementation of MinMaxScaler is very simple:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/23903773b0a4877299175beb2272ac66/href">https://medium.com/media/23903773b0a4877299175beb2272ac66/href</a></iframe><p>After running the above code our data set will look something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/404/1*-VCOrnhNk6LR0JWjPpZs-A.png" /><figcaption>Dataset after scaling</figcaption></figure><p>As you can see in the above image all the values in the dataset are now between 0 and 1, so there are no dominant features, and all features will be considered equally.</p><p>Note: Feature scaling is not always necessary and only required in some machine learning models.</p><h3>Step 6: Splitting the dataset</h3><p>Before we begin training our model there is one final step to go, which is splitting of the testing and training dataset. In machine learning, a <strong>larger part </strong>of the dataset is used to train the model, and a small part is used to test the trained model for finding out the accuracy and the efficiency of the model.</p><p>Now before we begin splitting the dataset we need to separate the dependent and independent variables which we have already discussed above in the article.</p><p>The last (<em>purchased</em>) column is the dependent variable and the rest are independent variables, so we’ll store the dependent variable in ‘y’ and the independent variables in ‘X’.</p><p>Another important part we need to remember is that while training the model accepts data as arrays so it is necessary that <strong>we convert the data to arrays</strong>. We do that while separating the dependent and independent variables by adding <strong><em>.values</em></strong><em> </em>while storing data in ‘X’ and ‘y’.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/58dcc8abcd283f8bf5c654ad97741050/href">https://medium.com/media/58dcc8abcd283f8bf5c654ad97741050/href</a></iframe><p>After running the above code our data will look something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/607/1*UoGFeITH9Zf8dFJqBIxW1w.png" /><figcaption>X and y</figcaption></figure><p>Now let’s split the dataset between Testing data and Training data.</p><p>To do this we’ll be using the <strong>train_test_split </strong>class from the same <strong>ScikitLearn </strong>library.</p><p>Deciding the ratio between testing data and training data is up to us and depends on what we are trying to achieve with our model. In our case, we are going to go with an 80-20% split between the train-test data. So 80% training and 20% testing data.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/7333813fb75348a0d5c5ac33676fbee3/href">https://medium.com/media/7333813fb75348a0d5c5ac33676fbee3/href</a></iframe><p>Here the <em>test_size = 0.2 </em>signifies that we have selected 20% of data as testing data, you can change that according to your choice.</p><p>After this, the <em>X_train</em> and <em>X_test</em> variables will have their respective data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*v7_5CXLKSaBcV6B-iBDrmg.png" /></figure><p>Now our data is finally ready for training!!</p><p>I hope this article was helpful in understanding data preprocessing. Be sure to drop your feedback and suggestion below.</p><p>Thank You for reading!</p><p><strong>PS: Once again you can view the code and dataset I have used on this </strong><a href="https://github.com/devilsboy/DataPreprocessing"><strong>GitHub repository</strong></a><strong>.</strong></p><h4>More from <a href="http://usejournal.com/">Journal</a></h4><p>There are many Black creators doing incredible work in Tech. This collection of resources shines a light on some of us:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fusejournal.com%2Fapp%2Fembed%2Fjournal%3Aspace%3Aproject%2F5c84f6f0-715c-48b8-9387-9e74f33b7a7b&amp;display_name=Journal&amp;url=https%3A%2F%2Fusejournal.com%2Fapp%2Fspace%2Fjournal%3Aspace%3Aproject%2F5c84f6f0-715c-48b8-9387-9e74f33b7a7b&amp;image=https%3A%2F%2Fimperio.usejournal.com%2Fimage%2Fjournal%3Aspace%3Aproject%2F5c84f6f0-715c-48b8-9387-9e74f33b7a7b%2F1602668529%3Fclient%3Dfacebook&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=usejournal" width="900" height="710" frameborder="0" scrolling="no"><a href="https://medium.com/media/ca84c63a615be91123438987244ec06c/href">https://medium.com/media/ca84c63a615be91123438987244ec06c/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1bfee9268fb3" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>