TY - JOUR T1 - Anatomically derived regional measurement of lung function JF - European Respiratory Journal JO - Eur Respir J VL - 38 IS - Suppl 55 SP - p4613 AU - Michael Bennett AU - Tom Havelock AU - Joy Conway AU - John Fleming Y1 - 2011/09/01 UR - http://erj.ersjournals.com/content/38/Suppl_55/p4613.abstract N2 - Existing methods for quantifying lung ventilation and perfusion SPECT scans provide measurements which are either global, based on division of the lung into non-anatomical regions or based on the application of a generic lung model of lung anatomy. The objective of this work is to develop a technique to use HRCT data to automatically define sub-lobar regions corresponding to anatomical lung segments and to align these with SPECT/CT data acquired separately from the same subject. This can be achieved by performing a non-rigid registration between the HRCT data, acquired at full inspiration, and the low-res CT data captured as part of the SPECT procedure, acquired during tidal breathing. Once the transform required to bring these two data sets into alignment has been determined, it can be applied to the sub-lobar regions to map these to the low-res CT and hence to the SPECT data. This process of non-rigid registration is achieved using a free-form polynomial warp algorithm, which automatically determines the optimum elastic transform required to bring the HRCT and low-res CT data sets into alignment.HRCT scans are analysed using commercially available software to define the airway tree geometry, which is then used to seed a Voronoi space division algorithm to divide the lung into segments based on the branches of the airway tree. Early results have shown that it is possible to robustly divide the lung into the regions based on the segmental branches of the airway tree and good alignment between single 2D slices and the SPECT images has been obtained.The method will be extended to work in 3D and the results so far indicate that this will be a viable method to produce anatomically relevant measurements of lung function. ER -