Biodosimetry-based discrimination between homogeneous total-body photon exposure and complex irradiation scenarios (partial-body shielding and/or neutron + photon mixtures) can improve treatment decisions after mass-casualty radiation-related incidents. Our study objective was to use high-throughput biomarkers to: a) detect partial-body and/or neutron exposure on an individual basis, and b) estimate separately the photon and neutron doses in a mixed exposure.We developed a novel approach, where metrics related to the shapes of micronuclei distributions per binucleated cell in ex-vivo irradiated human lymphocytes (variance/mean, kurtosis, skewness, etc.) served as predictors in machine learning or parametric analyses of the following scenarios: (A) Homogeneous gamma-irradiation, mimicking total-body exposures, vs. mixtures of irradiated blood with unirradiated blood, mimicking partial-body exposures. (B) X rays vs. various neutron + photon mixtures. Classification of samples as homogeneously vs. heterogeneously irradiated (scenario A) achieved a receiver operating characteristic curve area (AUROC) of 0.931 (uncertainty range of 0.903-0.951), and R 2 for actual vs. reconstructed mean dose was 0.87. Detection of samples with ≥ 10% neutron contribution (scenario B) achieved AUROC of 0.916 (0.893-0.943), and R 2 for reconstructing photon-equivalent dose was 0.77. These encouraging findings demonstrate a proof-of-principle for the proposed approach of analyzing micronuclei/cell distributions to detect clinically-relevant complex radiation exposure scenarios.biodosimetry approach in this scenario was to distinguish neutron + photon mixtures from pure photon exposures, and to quantify the neutron contribution.