In this project we present a system that uses Time-of-Flight (ToF) technology for automatic classification and verification of breathing patterns. Due to the correlation of respiratory motion and tumor movement a stable breathing pattern during tumor treatment is mandatory. Assuming the patient breaths regular, a mapping between the respiratory motion and the tumor movement can be established prior to the treatment. Therefore it is important to detect changes in the breathing pattern in order to adjust these mapping continuously. The proposed algorithm calculates multiple volume signals of different anatomical regions of the upper part of the patient’s body. Therefore parallel regions of interest are defined for the patient’s abdomen and thorax. These feature signals are used for classifying the respiration progression over time by using different signal energy criteria. Changing breathing patterns can be visualized in a 2-D histogram, which is also used to classify and detect abnormal breathing phases.