<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=utf-8">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    Dear all, <br>
    <br>
    I updated dmritool for C++11 with better efficiency. <br>
    <br>
    I also released some codes for algorithms and demos on uniform
    single and multiple shell sampling scheme design. <br>
    If you are interested in sampling scheme design, please check the
    tutorial for the updates. <br>
    <a class="moz-txt-link-freetext" href="https://diffusionmritool.github.io/tutorial_qspacesampling.html">https://diffusionmritool.github.io/tutorial_qspacesampling.html</a><br>
    <br>
    Here are the updates in details (see the news
    <a class="moz-txt-link-freetext" href="https://diffusionmritool.github.io/news.html#dmritool-now-uses-c-11">https://diffusionmritool.github.io/news.html#dmritool-now-uses-c-11</a>):
    <br>
    <ul class="simple">
      <li>We use C++11 in DMRITool now.
        <ul>
          <li>NDArray now is more efficient with C++11.</li>
          <li>The codes have to be built with gcc (&gt;= 4.8), ITK
            (&gt;= 4.9). Please check <a class="reference internal"
              href="https://diffusionmritool.github.io/building.html"><span
                class="doc">the building instruction</span></a>.</li>
        </ul>
      </li>
      <li>Some algorithms and demos on sampling scheme design are
        released.
        <ul>
          <li>Please check the <a class="reference internal"
              href="https://diffusionmritool.github.io/tutorial_qspacesampling.html"><span
                class="doc">updated tutorial on sampling scheme design</span></a>.</li>
          <li>1-Opt: <a class="reference internal"
href="https://diffusionmritool.github.io/commands/SamplingSchemeQSpace1OptEstimation.html"><span
                class="doc">SamplingSchemeQSpace1OptEstimation</span></a>,
            a new greedy algorithm which refines a given initialization.
            It can fix a known issue of <a class="reference internal"
href="https://diffusionmritool.github.io/commands/SamplingSchemeQSpaceIMOCEstimation.html"><span
                class="doc">IMOC</span></a> when the number of samples
            is large.
            See <a class="reference external"
href="https://diffusionmritool.github.io/tutorial_qspacesampling.html#sampling-schemes-by-imoc-1-opt-cnlo">the
              tutorial</a>.</li>
          <li>CNLO: <a class="reference internal"
href="https://diffusionmritool.github.io/matlabfiles/OptimalSamplingSingleSubset.html"><span
                class="doc">OptimalSamplingSingleShellCNLO</span></a>
            for single-shell scheme design.</li>
          <li>CNLO: <a class="reference internal"
href="https://diffusionmritool.github.io/matlabfiles/OptimalSamplingMultiShellCNLO.html"><span
                class="doc">OptimalSamplingMultiShellCNLO</span></a> for
            multiple-shell scheme design.</li>
          <li>jointly use IMOC + 1-Opt + CNLO for sampling scheme
            design:
            <ul>
              <li><a class="reference external"
href="https://diffusionmritool.github.io/demos/demo_singleshell_IMOC_1Opt_CNLO.html">demo
                  for single shell scheme</a></li>
              <li><a class="reference external"
href="https://diffusionmritool.github.io/demos/demo_mutishell_IMOC_1Opt_CNLO.html">demo
                  for multiple shell scheme</a></li>
            </ul>
          </li>
        </ul>
      </li>
    </ul>
    <br>
    Thanks. <br>
    <br>
    best,<br>
    Jian Cheng<br>
    <br>
  </body>
</html>