Systems and methods for managing motions of an anatomical region of interest of a patient during image-guided radiotherapy are disclosed. An exemplary system may include an image acquisition device, a radiotherapy device, and a processor device. The processor device may be configured to control the image acquisition device to acquire at least one 2D image. Each 2D image may include a cross-sectional image of the anatomical region of interest. The processor device may also be configured to perform automatic contouring in each 2D image to extract a set of contour elements segmenting the cross-sectional image of the anatomical region of interest in that 2D image. The processor device may be further configured to match the set of contour elements to a 3D surface image of the anatomical region of interest to determine a motion of the anatomical region of interest and to control radiation delivery based on the determined motion.