<div dir="ltr"><div><font face="Lucida Grande, Verdana, Geneva, Arial, sans-serif"><span style="font-size:12px;line-height:22px">I don't know if it is relevant but the field SliceLocation is inversed between the two headers. GDCM is reorganizing the slice itself and can modify the origin and the orientation itself:</span></font></div>
<div><ol type="1" style="line-height:22px;font-size:12px;font-family:'Lucida Grande',Verdana,Geneva,Arial,sans-serif"><li style="margin-left:15px">Extract <a href="http://www.itk.org/Doxygen/html/classitk_1_1Image.html" title="Templated n-dimensional image class. " target="_blank" style="color:rgb(70,101,162);font-weight:bold">Image</a> Orientation & <a href="http://www.itk.org/Doxygen/html/classitk_1_1Image.html" title="Templated n-dimensional image class. " target="_blank" style="color:rgb(70,101,162);font-weight:bold">Image</a> Position from DICOM images, and then calculate the ordering based on the 3D coordinate of the slice</li>
</ol><div><font face="Lucida Grande, Verdana, Geneva, Arial, sans-serif"><span style="font-size:12px;line-height:22px">Regarding the spacing, it seems that the value given in Matlab and by the reader are the same [.9766 .9766]. However, you would need to give probably the adjacent slice to know about the z spacing which is not appearing inside the header. You probably did it, but what is the z spacing when checking the image position dicom tag between two PET DICOM images. Is the spacing of 3.5 mm or 1 mm?</span></font></div>
</div><div><font face="Lucida Grande, Verdana, Geneva, Arial, sans-serif"><span style="font-size:12px;line-height:22px"><br></span></font></div><div><font face="Lucida Grande, Verdana, Geneva, Arial, sans-serif"><span style="font-size:12px;line-height:22px">Cheers,</span></font></div>
<div class="gmail_extra">
</div></div>