A device and a method are disclosed for automated planning of an access path for a percutaneous, minimally invasive intervention on an area of the body, in particular on the lungs. In the method, 3D image data of the body area are prepared, from which bones and elements endangered by the intervention are automatically segmented. In a display of the 3D image data and/or of image data derived therefrom, a target position is marked in the 3D image data by a user. On the basis of one or more predefined path geometries, the target position and the segmented data, a planning module automatically determines one or more access paths to the target position which do not run through bones and which do not intersect any elements endangered by the intervention, or intersect only a minimal number of elements endangered by the intervention. The one or more access paths are presented to the user on a monitor for information and/or interactive selection and/or correction in a display of the 3D image data or of image data derived therefrom.