Ab (Mathworks, Natick, MA, USA), which analyses video sequences acquired with an alveolar endoscope (Sch ly, Denzlingen, Germany) [2]. The user has to provide a pointer for the alveoli that shall be traced and whose modifications in size and shape are to be determined. Filters, smoothing splines and expectation-driven fine tuning is performed to attain robust and predictable results in the intrati