[en] Full text: As the sizes of mouse organ are comparable with the range of the high-energy beta particles emitted by the radionuclides commonly used in radionuclide therapy a significant amount of beta radiation emitted could be imparted to the adjacent tissues. The often assumption that beta particles are fully-absorbed at the emission site is not satisfied and cross-irradiation should be included into the dose estimation formulas. Keeping in mind that the radiation effects are correlated with the absorbed dose in the target the inclusion of cross-irradiation in the dose estimation must be evaluated. The MIRD's formulation was used to perform absorbed dose calculation in mice using absorbed fractions previously reported for ¹³¹I, ⁹⁰Y and ¹⁷⁷Lu. Two approaches were considered: a) cross irradiation when a fraction of beta particles emitted can escape from the organ source and, b) full self- irradiation when the beta particles are considered fully absorbed at the emission site. The formulation of linear-quadratic model was readapted to be used in the radionuclide therapy. Treatment with a single administration in mice was simulated and radiation effects on tumor, bone marrow and kidneys under the assumption of cross-irradiation were predicted. A biphasic repair kinetics was considered in the calculation of irradiation effects on kidneys. Typical published biokinetic data for radiopharmaceutical assayed in mice and radiobiological parameters were used in the calculations. The influence of cross irradiation condition was diverse for the tissues analyzed here. The absorbed dose values in kidneys calculated for both methods were no significantly different for low energies, but variations around to 40-50% (over or under-estimation) in absorbed dose were obtained for high energies. Approximately a 30% of the beta radiation emitted from bone will cross irradiates the bone marrow. For injected activities values higher than 10MBq (300μCi), as a single injection, the absorbed dose in BM exceeds the tolerable limits (2Gy). The formulation presented here could be used in the design of refined experiments for radionuclide therapy with mice model where the radio myelotoxicity and/or toxicity in kidneys needs to be controlled. (author)