In vivo dose measurements with diodes are easy to perform. The first aim of our study was to show whether diode measurements of the patient exit doses are precise enough for verifying inhomogeneity corrections used for treatment planning. The second aim was to assess the precision of the modified Batho Law inhomogeneity correction of the CadPlan treatment planning system. For this purpose, entrance and lait doses were measured in the thoracic region of 115 patients. Diode measurements were sufficiently precise to verify the density corrections predicted by the treatment planning system (< 0.5% of ICRU dose). The measured doses were compared with calculations of the CadPlan treatment planning system. The mean deviation of the exit dose calculations within the measurements error was zero. The present results show that measurements of exit dose even in a small number of patients are sufficient to identify systematic errors in the dose calculation.