Meeting Banner
Abstract #3980

Automatic Segmentation of Lung Parenchyma Using Fuzzy Clustering

Andr Fischer1, Christian Oliver Ritter1, Dietbert Hahn1, Herbert Kstler1

1Institute of Radiology, University of Wuerzburg, Wuerzburg, Germany

This work describes the ability of Fuzzy C-Means (FCM) clustering to accurately distinguish between pulmonary parenchyma, pulmonary vessels, the heart, and the surrounding tissue in dynamic contrast enhanced (DCE)-MRI. FCM clustering achieves this by clustering voxels with similar temporal signal courses together. A 3D DCE-MRI dataset was accordingly segmented and is presented in this work. This technique enables user independent automatic segmentation of the lung parenchyma necessary to quantify lung perfusion. Thereby, a subjective bias in data analysis as often present in manual parenchyma segmentation is lowered.

Abstract PDF E-Poster Video

Keywords

accurately achieve affected agent alternatively always amplitude anatomic anatomy another aorta applied approaches arrow arterial arteries artifacts asymmetric automatic automatically belong blood blue boundaries capability cardiac caused channel characteristics circumvent classifies cluster clustered clustering clusters coil combination combined comes containing context cost course curves dataset depiction desirable determine differ distinguish distributed drawing drawn early encoding enhancement erroneous exhibit falsely feasible final finally fuzzy hard heart hence hypothesis in vivo included includes increasing indicated indicates induced influences institute interaction introduced invest larger leading leads light like lung lungs manual manually many middle misaligning moderately motion network notice obviously oliver onset originally outside overestimation overlay partial particular partition passage perfusion permeated projection properly propose pulmonary quantified quantitative radiology reduced regional reported required resolution respective restrict samples segment segmentation selected shape significantly structures studies subjectivity subsequent successfully surrounding temporal thereby therein thoracic thorax tissue tolerated transit traversed trio truncation truncations user various vessels volume washed yellow