https://zeynelsamak.github.io/news/

1<!DOCTYPE html>
2<html lang="en">
3
4  <!-- Head -->
5  <head>
6<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">    <!-- Metadata, OpenGraph and Schema.org -->
7    
8
9    <!-- Standard metadata -->
10    <meta charset="utf-8">
11    <meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
12    <meta http-equiv="X-UA-Compatible" content="IE=edge">
13    <title>news | Zeynel Abidin Samak</title>
14    <meta name="author" content="Zeynel Abidin Samak">
15    <meta name="description" content="Zeynel Abidin Samak, PhD, ViLab, University of Bristol
16">
17    <meta name="keywords" content="zeynel,samak,abidin, deep,learning,stroke,thrombectomy,medical">
18
19
20    <!-- Bootstrap & MDB -->
21    <link href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" rel="stylesheet" integrity="sha256-DF7Zhf293AJxJNTmh5zhoYYIMs2oXitRfBjY+9L//AY=" crossorigin="anonymous">
22    <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/css/mdb.min.css" integrity="sha256-jpjYvU3G3N6nrrBwXJoVEYI/0zw8htfFnhT9ljN3JJw=" crossorigin="anonymous">
23
24    <!-- Fonts & Icons -->
25    <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/@fortawesome/[email protected]/css/all.min.css" integrity="sha256-mUZM63G8m73Mcidfrv5E+Y61y7a12O5mW4ezU3bxqW4=" crossorigin="anonymous">
26    <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/css/academicons.min.css" integrity="sha256-i1+4qU2G2860dGGIOJscdC30s9beBXjFfzjWLjBRsBg=" crossorigin="anonymous">
27    <link rel="stylesheet" type="text/css" href="https://fonts.googleapis.com/css?family=Roboto:300,400,500,700|Roboto+Slab:100,300,400,500,700|Material+Icons">
28
29    <!-- Code Syntax Highlighting -->
30    <link rel="stylesheet" href="https://cdn.jsdelivr.net/gh/jwarby/jekyll-pygments-themes@master/github.css" media="" id="highlight_theme_light">
31
32    <!-- Styles -->
33    
34    <link rel="stylesheet" href="/assets/css/main.css">
35    <link rel="canonical" href="https://zeynelsamak.github.io//news/">
36    
37    <!-- Dark Mode -->
38    
39    <link rel="stylesheet" href="https://cdn.jsdelivr.net/gh/jwarby/jekyll-pygments-themes@master/native.css" media="none" id="highlight_theme_dark">
40
41    

41<script src="/assets/js/theme.js"></script>

41
42    

42<script src="/assets/js/dark_mode.js"></script>

42
43    
44
45  </head>
46
47  <!-- Body -->
48  <body class="fixed-top-nav sticky-bottom-footer">
49
50    <!-- Header -->
51    <header>
52
53      <!-- Nav Bar -->
54      <nav id="navbar" class="navbar navbar-light navbar-expand-sm fixed-top">
55        <div class="container">
56          <a class="navbar-brand title font-weight-lighter" href="/"><span class="font-weight-bold">Zeynel </span>Abidin Samak</a>
57          <!-- Navbar Toggle -->
58          <button class="navbar-toggler collapsed ml-auto" type="button" data-toggle="collapse" data-target="#navbarNav" aria-controls="navbarNav" aria-expanded="false" aria-label="Toggle navigation">
59            <span class="sr-only">Toggle navigation</span>
60            <span class="icon-bar top-bar"></span>
61            <span class="icon-bar middle-bar"></span>
62            <span class="icon-bar bottom-bar"></span>
63          </button>
64
65          <div class="collapse navbar-collapse text-right" id="navbarNav">
66            <ul class="navbar-nav ml-auto flex-nowrap">
67
68              <!-- About -->
69              <li class="nav-item ">
70                <a class="nav-link" href="/"><strong> </strong></a>
71              </li>
72              
73
74              <!-- Other pages -->
75              <li class="nav-item ">
76                <a class="nav-link" href="/publications/">publications</a>
77              </li>
78              <li class="nav-item ">
79                <a class="nav-link" href="/teaching/">teaching</a>
80              </li>
81
82              <!-- Toogle theme mode -->
83              <li class="toggle-container">
84                <button id="light-toggle" title="Change theme">
85                  <i class="fas fa-moon"></i>
86                  <i class="fas fa-sun"></i>
87                </button>
88              </li>
89            </ul>
90          </div>
91        </div>
92      </nav>
93
94      <!-- Scrolling Progress Bar -->
95      <progress id="progress" value="0">
96        <div class="progress-container">
97          <span class="progress-bar"></span>
98        </div>
99      </progress>
100    </header>
101
102
103    <!-- Content -->
104    <div class="container mt-5">
105      <!-- page.html -->
106        <div class="post">
107
108          <header class="post-header">
109            <h1 class="post-title">news</h1>
110            <p class="post-description"></p>
111          </header>
112
113          <article>
114            
115          <div class="news">
116            <!-- <p>AAA  false,5,true AAA</p> -->
117            <div class="table-responsive">
118              <table class="table table-sm table-borderless">
119              
120                <tr>
121                  <th scope="row">Feb 18, 2025</th>
122                  <td>
123                    Our paper <a href="/projects/stroke_review/">“Automatic prediction of stroke treatment outcomes: latest advances and perspectives”</a> is accepted in  <a href="https://link.springer.com/journal/13534" rel="external nofollow noopener" target="_blank">Biomedical Engineering Letters</a>
124
125<!-- ---
126layout: post
127title:  Our paper "FeMA; Feature matching auto-encoder for predicting ischaemic stroke evolution and treatment outcome" is accepted in Computerized Medical Imaging and Graphics (CMIG) 2022
128date: 2022-06-03 13:00:00-0400
129inline: false
130related_posts: false
131---
132
133### Abstract ###
134Although, predicting ischaemic stroke evolution and treatment outcome provide important information one step towards individual treatment planning, classifying functional outcome and modelling the brain tissue evolution remains a challenge due to data complexity and visually subtle changes in the brain. We propose a novel deep learning approach, Feature Matching Auto-encoder (FeMA) that consists of two stages, predicting ischaemic stroke evolution at one week without voxel-wise annotation and predicting ischaemic stroke treatment outcome at 90 days from a baseline scan. In the first stage, we introduce feature similarity and consistency objective, and in the second stage, we show that adding stroke evolution information increase the performance of functional outcome prediction. Comparative experiments demonstrate that our proposed method is more effective to extract representative follow-up features and achieves the best results for functional outcome of stroke treatment. -->
135
136                  </td>
137                </tr>
138                <tr>
139                  <th scope="row">Nov 2, 2024</th>
140                  <td>
141                    Our paper <a href="https://link.springer.com/article/10.1007/s11517-024-03243-4" rel="external nofollow noopener" target="_blank">“Multi-type stroke lesion segmentation: comparison of single-stage and hierarchical approach “</a> is accepted in  <a href="https://link.springer.com/journal/11517" rel="external nofollow noopener" target="_blank">Medical &amp; Biological Engineering &amp; Computing</a>
142
143
144                  </td>
145                </tr>
146                <tr>
147                  <th scope="row">Mar 31, 2023</th>
148                  <td>
149                    I have succesfully defended my PhD thesis. Thanks to my reviewers <a href="https://reyesaldasoro.github.io/" rel="external nofollow noopener" target="_blank">Carlos Reyes-Aldasoro</a> and <a href="https://amachim.blogs.bristol.ac.uk/" rel="external nofollow noopener" target="_blank">Alin Achim</a> for their valuable feedback and time spent reading the thesis.
150
151                  </td>
152                </tr>
153                <tr>
154                  <th scope="row">Jan 19, 2023</th>
155                  <td>
156                    Our paper <a href="/projects/transop/">“TranSOP: Transformer-based Multimodal Classification for Stroke Treatment Outcome Prediction”</a> is accepted in ISBI 2023 conference
157
158                  </td>
159                </tr>
160                <tr>
161                  <th scope="row">Jun 3, 2022</th>
162                  <td>
163                    Our paper <a href="/projects/fema/">“FeMA: Feature matching auto-encoder for predicting ischaemic stroke evolution and treatment outcome”</a> is accepted in Computerized Medical Imaging and Graphics (CMIG) 2022
164
165<!-- ---
166layout: post
167title:  Our paper "FeMA; Feature matching auto-encoder for predicting ischaemic stroke evolution and treatment outcome" is accepted in Computerized Medical Imaging and Graphics (CMIG) 2022
168date: 2022-06-03 13:00:00-0400
169inline: false
170related_posts: false
171---
172
173### Abstract ###
174Although, predicting ischaemic stroke evolution and treatment outcome provide important information one step towards individual treatment planning, classifying functional outcome and modelling the brain tissue evolution remains a challenge due to data complexity and visually subtle changes in the brain. We propose a novel deep learning approach, Feature Matching Auto-encoder (FeMA) that consists of two stages, predicting ischaemic stroke evolution at one week without voxel-wise annotation and predicting ischaemic stroke treatment outcome at 90 days from a baseline scan. In the first stage, we introduce feature similarity and consistency objective, and in the second stage, we show that adding stroke evolution information increase the performance of functional outcome prediction. Comparative experiments demonstrate that our proposed method is more effective to extract representative follow-up features and achieves the best results for functional outcome of stroke treatment. -->
175
176                  </td>
177                </tr>
178                <tr>
179                  <th scope="row">Sep 29, 2021</th>
180                  <td>
181                    Our team <em>xyz.ai</em> won the <a href="https://sbsgm.saglik.gov.tr/TR-85717/teknofestte-saglik-var.html" rel="external nofollow noopener" target="_blank">3rd</a>

181 place in <a href="https://teknofest.org/en/competitions/competition/34" rel="external nofollow noopener" target="_blank">Artificial Intelligence in Healthcare Competition</a> that is organised by <a href="https://teknofest.org/" rel="external nofollow noopener" target="_blank">TEKNOFEST</a>
182
183
184                  </td>
185                </tr>
186                <tr>
187                  <th scope="row">May 4, 2020</th>
188                  <td>
189                    Our paper <a href="/projects/evtoutcome/">“Prediction of Thrombectomy Functional Outcomes using Multimodal Data”</a> is accepted in MIUA 2020 conference
190
191<!-- ---
192layout: post
193title:  Our paper "Prediction of Thrombectomy Functional Outcomes using Multimodal Data" is accepted in MIUA 2020
194date: 2020-05-04 16:11:00-0400
195inline: false
196related_posts: false
197---
198
199### Abstract ###
200Recent randomised clinical trials have shown that while patients with ischaemic stroke benefit from endovascular thrombectomy, determining whether a patient's circumstance is favourable for the treatment remains a challenge. We propose a novel deep learning approach to directly exploit multimodal data (clinical metadata information and medical imaging, as well as clinical scores obtained from images) to estimate the success of ischaemic stroke treatment. We
201incorporate an attention mechanism in our architecture to model global feature inter-dependencies, both channel-wise and spatially. We perform comparative experiments using unimodal and multimodal data, and achieve 0.75 AUC in dichotomised mRS scores and 0.35 classification accuracy in individual mRS scores. -->
202
203                  </td>
204                </tr>
205              </table>
206            </div>
207          </div>
208
209
210          </article>
211
212        </div>
213
214    </div>
215
216    <!-- Footer -->    <footer class="sticky-bottom mt-5">
217      <div class="container">
218        © Copyright 2025 Zeynel Abidin Samak. Powered by <a href="https://jekyllrb.com/" target="_blank" rel="external nofollow noopener">Jekyll</a> with <a href="https://github.com/alshedivat/al-folio" rel="external nofollow noopener" target="_blank">al-folio</a> theme. Hosted by <a href="https://pages.github.com/" target="_blank" rel="external nofollow noopener">GitHub Pages</a>.
219
220      </div>
221    </footer>
222
223    <!-- JavaScripts -->
224    <!-- jQuery -->
225  

225<script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/jquery.min.js" integrity="sha256-/xUj+3OJU5yExlq6GSYGSHk7tPXikynS7ogEvDej/m4=" crossorigin="anonymous"></script>

225
226
227    <!-- Bootsrap & MDB scripts -->
228  

228<script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.bundle.min.js" integrity="sha256-fgLAgv7fyCGopR/gBNq2iW3ZKIdqIcyshnUULC4vex8=" crossorigin="anonymous"></script>

228
229  

229<script src="https://cdn.jsdelivr.net/npm/[email protected]/js/mdb.min.js" integrity="sha256-NdbiivsvWt7VYCt6hYNT3h/th9vSTL4EDWeGs5SN3DA=" crossorigin="anonymous"></script>

229
230
231    <!-- Masonry & imagesLoaded -->
232  

232<script defer src="https://cdn.jsdelivr.net/npm/[email protected]/dist/masonry.pkgd.min.js" integrity="sha256-Nn1q/fx0H7SNLZMQ5Hw5JLaTRZp0yILA/FRexe19VdI=" crossorigin="anonymous"></script>

232
233  

233<script defer src="https://cdn.jsdelivr.net/npm/imagesloaded@4/imagesloaded.pkgd.min.js"></script>

233
234  

234<script defer src="/assets/js/masonry.js" type="text/javascript"></script>

234
235    
236  <!-- Medium Zoom JS -->
237  

237<script defer src="https://cdn.jsdelivr.net/npm/[email protected]/dist/medium-zoom.min.js" integrity="sha256-7PhEpEWEW0XXQ0k6kQrPKwuoIomz8R8IYyuU1Qew4P8=" crossorigin="anonymous"></script>

237
238  

238<script defer src="/assets/js/zoom.js"></script>

238<!-- Load Common JS -->
239  

239<script defer src="/assets/js/common.js"></script>

239
240  

240<script defer src="/assets/js/copy_code.js" type="text/javascript"></script>

240
241
242    
243  

243<script async src="https://d1bxh8uas1mnw7.cloudfront.net/assets/embed.js"></script>

243
244  

244<script async src="https://badge.dimensions.ai/badge.js"></script>

244
245
246    <!-- MathJax -->
247  

247<script type="text/javascript">
248    window.MathJax = {
249      tex: {
250        tags: 'ams'
251      }
252    };
253  </script>

253
254  

254<script defer type="text/javascript" id="MathJax-script" src="https://cdn.jsdelivr.net/npm/[email protected]/es5/tex-mml-chtml.js"></script>

254
255  

255<script defer src="https://polyfill.io/v3/polyfill.min.js?features=es6"></script>

255
256
257    
258    
259
260<!-- Scrolling Progress Bar -->

261<script type="text/javascript">
262  /*
263   * This JavaScript code has been adapted from the article 
264   * https://css-tricks.com/reading-position-indicator/ authored by Pankaj Parashar, 
265   * published on the website https://css-tricks.com on the 7th of May, 2014.
266   * Couple of changes were made to the original code to make it compatible 
267   * with the `al-foio` theme.
268   */
269  const progressBar = $("#progress");
270  /*
271   * We set up the bar after all elements are done loading.
272   * In some cases, if the images in the page are larger than the intended
273   * size they'll have on the page, they'll be resized via CSS to accomodate
274   * the desired size. This mistake, however, breaks the computations as the
275   * scroll size is computed as soon as the elements finish loading.
276   * To account for this, a minimal delay was introduced before computing the
277   * values.
278   */
279  window.onload = function () {
280    setTimeout(progressBarSetup, 50);
281  };
282  /*
283   * We set up the bar according to the browser.
284   * If the browser supports the progress element we use that.
285   * Otherwise, we resize the bar thru CSS styling
286   */
287  function progressBarSetup() {
288    if ("max" in document.createElement("progress")) {
289      initializeProgressElement();
290      $(document).on("scroll", function() {
291        progressBar.attr({ value: getCurrentScrollPosition() });
292      });
293      $(window).on("resize", initializeProgressElement);
294    } else {
295      resizeProgressBar();
296      $(document).on("scroll", resizeProgressBar);
297      $(window).on("resize", resizeProgressBar);
298    }
299  }
300  /*
301   * The vertical scroll position is the same as the number of pixels that
302   * are hidden from view above the scrollable area. Thus, a value > 0 is
303   * how much the user has scrolled from the top
304   */
305  function getCurrentScrollPosition() {
306    return $(window).scrollTop();
307  }
308
309  function initializeProgressElement() {
310    let navbarHeight = $("#navbar").outerHeight(true);
311    $("body").css({ "padding-top": navbarHeight });
312    $("progress-container").css({ "padding-top": navbarHeight });
313    progressBar.css({ top: navbarHeight });
314    progressBar.attr({
315      max: getDistanceToScroll(),
316      value: getCurrentScrollPosition(),
317    });
318  }
319  /*
320   * The offset between the html document height and the browser viewport
321   * height will be greater than zero if vertical scroll is possible.
322   * This is the distance the user can scroll
323   */
324  function getDistanceToScroll() {
325    return $(document).height() - $(window).height();
326  }
327
328  function resizeProgressBar() {
329    progressBar.css({ width: getWidthPercentage() + "%" });
330  }
331  // The scroll ratio equals the percentage to resize the bar
332  function getWidthPercentage() {
333    return (getCurrentScrollPosition() / getDistanceToScroll()) * 100;
334  }
335</script>

335
336
337  </body>
338</html>