Characterizing the fatigue properties of metallic additively manufactured (AM) lattice components is essential and it has been the subject of many studies. Accurate assessment of fatigue strength requires computational methods that account for the influence of manufacturing-induced internal pores and surface defects with adequate resolution. This work investigates the fatigue behavior of two laser powder bed fusion (LPBF) Ti6Al4V lattice components by employing: (a) an image-based finite cell method (FCM) to explicitly consider the effect of geometrical defects captured via computed tomography (CT) scans, and (b) the average strain energy density (SED) approach to characterize fatigue failure based on a critical SED value. The image-based FCM has proven effective in resolving fine geometric details of various as-built AM parts, with promising preliminary applications in fatigue life prediction. Moreover, SED-based methods have been successfully utilized for analyzing the fatigue life of diverse metallic components. In conjunction with the above approach, a two-scale modeling workflow can be used to obtain detailed stress states required for such local fatigue assessment, particularly for components within larger structures, e.g. nodes of space frames. In such a workflow, the fully resolved local stress field associated with the AM part, computed by analyzing the detailed 3D geometry of the AM component under boundary conditions derived from a global analysis, serves as the direct input for fatigue characterization with the average SED approach. In this work, we first compare the critical SED value derived from physical experiments to the computationally estimated critical SED value using the volume-based SED criterion. Then we integrate a two-scale workflow to assess the fatigue life of individual components within a large structure. We conclude that (a) the computed critical SED values demonstrate close agreement with the experimental fatigue results, validating the approach for assessing the fatigue life of metallic AM lattice components with complex geometric features, and (b) the introduction of the two-scale workflow to the fatigue life characterization approach can provide additional insights into the fatigue life of an AM component within a large-scale structure.