<div class="sc-cmRAlD dkqmWS"><div class="sc-UEtKG dGqiYy sc-flttKd cguEtd"><div class="sc-fqwslf gsqkEc"><div class="sc-cBQMlg kAHhUk"><h2 class="sc-dcKlJK sc-cVttbi gqEuPW ksnHgj">About Dataset</h2></div></div></div><div class="sc-jgvlka jFuPjz"><div class="sc-gzqKSP tNtjD"><div style="min-height: 80px;"><div class="sc-etVRix jqYJaa sc-bMmLMY ZURWJ"><h4>Introduction</h4>

<p>White blood cells called leukocytes include two groups of phagocytes and <code>lymphocytes</code>. While phagocytes comprise cells of the innate immune system and function rapidly after infection, lymphocytes mediate the acquired immune response. Phagocytes, themselves, can be divided into granulocytes (<code>neutrophils</code>, <code>basophils</code>, and <code>eosinophils</code>) and <code>monocytes</code>. In <strong>Table 1.</strong> and <strong>Fig 1.</strong>, you can see  the characteristics and images of the five categories of white blood cells.</p>

<p><strong>Fig 1.</strong> Five types of white blood cells in the normal peripheral blood.</p>

<p><img alt="Fig 1" src="https://media.springernature.com/full/springer-static/image/art%3A10.1038%2Fs41598-021-04426-x/MediaObjects/41598_2021_4426_Fig1_HTML.png"></p>

<p><strong>Table 1.</strong> Characteristics of white blood cells.</p>

<table>

  <tbody><tr>

    <th>WBCs</th>

    <th>% In blood</th>

    <th>Nucleus</th>

    <th>Cytoplasm</th>

    <th>Size (μm)</th>

  </tr>

  <tr>

    <td>Neutrophils</td>

    <td>60%</td>

    <td>It's divided into 2 to 5 segments and stains dark purple (multi-lobed)</td>

    <td>It's pale pink to tan with pink-purple granules</td>

    <td>12–16</td>

  </tr>

  <tr>

    <td>Eosinophils</td>

    <td>3%</td>

    <td>It has 2 lobes that each stains purple, and is difficult to be seen</td>

    <td>It's pale pink-tan but contains large purple/blue-black granules which obscure the cell nucleus</td>

    <td>14–16</td>

  </tr>

  <tr>

    <td>Basophils</td>

    <td>1%</td>

    <td>It's divided into 2 to 5 segments and stains dark purple (multi-lobed)</td>

    <td>It's pale pink to tan with pink-purple granules</td>

    <td>14–16</td>

  </tr>

  <tr>

    <td>Monocytes</td>

    <td>6%</td>

    <td>It's singular and is kidney shaped (convoluted shape), bean shaped or horseshoe shaped with deep indentation</td>

    <td>It stains a blue-gray color and is "ground glass" with tiny granules, Vacuoles are sometimes present in it</td>

    <td>14–20</td>

  </tr>

  <tr>

    <td>Lymphocytes</td>

    <td>30%</td>

    <td>It's large, round or oval, and is dark staining</td>

    <td>It is not present or very small, and is pale blue in color, and occasionally has purple-reddish granules</td>

    <td>8–15</td>

  </tr>

</tbody></table>

<h4>Data notebooks</h4>

<p>Two notebooks have been created to solve the classification of white blood cells problem using the transfer learning method on two pre-trained networks Resnet50 and InceptionV3.<br>

<a rel="noreferrer nofollow" aria-label="Github Notebook (opens in a new tab)" target="_blank" href="https://github.com/masoudnick/White-Blood-Cells-Classification/blob/main/White_Blood_Cells_Classification.ipynb">Github Notebook</a><br>

<a aria-label="Kaggle Notebook (opens in a new tab)" target="_blank" href="https://www.kaggle.com/code/masoudnickparvar/resnet50-white-blood-cells-classification">Kaggle Notebook</a></p>

<h4>Data weakness</h4>

<p>One of the weak points of this dataset is the lack of balance in the number of images of 5 classes, which can be solved with data augmentation technique.</p>

<h4>Data availability</h4>

<p>For more information about:</p>

<ol>

<li>How to collect dataset?</li>

<li>Which pre-processing performed on this dataset?</li>

<li>How to cite to this dataset in research?<br>

Visit the main site of Raabin-WBC dataset: <a rel="noreferrer nofollow" aria-label="Site (opens in a new tab)" target="_blank" href="https://www.raabindata.com/free-data">Site</a></li>

</ol></div></div></div>