Segmentation of medical images is very important for clinical research and diagnosis， leading to a requirement for robust automated methods. An essential part of a successful radiotherapy planning system for breast cancer treatment is the accurate segmentation of target organs at risk， such as lnngs. Distinguishing of the lnng cavities is not trivial due to largely changing lung shapes， low contrast and poorly defined bonndaries even in the absence of prominent neighboring structures. Inthis study， we address the lnng segmentation problem in pulmonary magnetic resonance imaging and propose an automated method based on a robust region-aided geometric snake with a modified diffused region force into the standard geometric model definition. (This extra region force which is created by Fuzzy C-Means algorithm gives the snake a global complementary view of the lnng bonndary information within the image. Along with the local gradient flow， it helps detect fuzzy bonndaries and overcome noisy regions in our MR images.) The proposed method has been successful in segmenting the lnngs in every slice of 30 magnetic resonance images with 80 consecutive slices in each image. We present results by comparing our automatic method to manually segmented lnng cavities provided by an expert radiologist and with those of previous works， showing encouraging results and high robustness of our approach.