In the early phase of a nuclear reactor accident, in-vivo monitoring of impacted population would be highly useful to detect potential contamination during the passage of the cloud and to estimate the dose from inhalation of measured radionuclides. However, it would be important to take into account other exposure components: (1) inhalation of unmeasured radionuclides and (2) external irradiation from the plume and from the radionuclides deposited on the soil. This article presents a methodology to calculate coefficients used to convert in-vivo measurement results directly into doses, not only from the measured radionuclides but from all sources of exposure according to model-based projected doses. This early interpretation of in-vivo measurements will provide an initial indication of individual exposure levels. As an illustration, the methodology is applied to two scenarios of accidents affecting a nuclear power plant: a loss-of-coolant accident leading to core meltdown and a steam generator tube rupture accident.