Influenza is a serious health concern, with an urgent need for improvement in vaccine efficacy. Classically, neutralizing antibodies are the driving force of antiviral immunity to influenza. While it is known that CD4+ T cells are also important, their precise role remains poorly characterized. To address this, we used high-dimensional immune cell profiling by mass cytometry (CyTOF®) on peripheral blood mononuclear cells from 46 healthy donors, at Day 0 and Day 7 after influenza vaccination. By applying a novel clustering algorithm called Denoised Ragged Pruning, we identified nine CD4+ T cell clusters, three of which (Clusters 2, 4, and 7) were responsive to influenza peptide stimulation. Since the differences between clusters are subtle, we applied a new visualization tool called Penalized Supervised Starplots to quickly and intuitively visualize the salient features driving cluster separation. We found that the three influenza-responsive clusters expressed markers of central memory (Cluster 2), effector memory (Cluster 4) and TEMRA (Cluster 7) phenotypes. Clusters 2 and 7 were responsive to vaccination, being significantly above background at Day 7 post-vaccination but not at Day 0. Cluster 4 was significantly above background at both Day 0 and Day 7, indicating that it is a pre-existing influenza-specific cluster. All three clusters showed distinct combinations of cytokines, activation markers and cytotoxic markers. Interestingly, the study cohort could be divided into three subgroups, based on its patterns in cluster abundance. Our findings provide valuable insights into CD4+ T cell subsets, and can aid vaccine improvement and baseline prediction of vaccine response.