Please use this identifier to cite or link to this item:
|Scopus||Web of Science®||Altmetric|
|Title:||Multiple mucociliary transit marker tracking in synchrotron X-ray images using the global nearest neighbor method|
|Citation:||Proceedings of the 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC' 17), 2017 / vol.2017, pp.1824-1827|
|Series/Report no.:||Proceedings of Annual International Conference of the IEEE Engineering In Medicine And Biology Society|
|Conference Name:||39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC' 17) (11 Jul 2017 - 15 Jul 2017 : Jeju Island, Korea)|
|Hye-Won Jung, Ivan Lee, Sang-Heon Lee, David Parsons, Martin Donnelley|
|Abstract:||Recent research has enabled in-vivo examination of mucociliary transit in live airways by analysing the movement patterns of micron-sized markers in high resolution synchrotron X-ray images. However, high levels of false positives and false negatives severely impact the performance of many automated tracking algorithms. This paper proposes an improved approach to track valid mucociliary transit markers using a modified gating region and cost matrix. The proposed method offers a more effective way to associate markers with the correct trackers. Improved visualization methods are also introduced to assist the interpretation of the tracking results. The tracking method has achieved a tracking accuracy of 81.7% track purity and 71.3% track effectiveness.|
|Keywords:||Respiratory System; Cluster Analysis; Algorithms; Synchrotrons; X-Rays; Pattern Recognition, Automated|
|Appears in Collections:||Paediatrics publications|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.