Objective To explore the proportion of immune cell infiltration in lung squamous cell carcinoma and to establish a prognostic risk assessment model for the patients. Methods Based on the whole transcriptome data of lung squamous cell carcinoma from The Cancer Genome Atlas (TCGA) database, R 4.0.3 software and CIBERSORT deconvolution algorithm were used to calculate the proportion of different immune cell infiltrations. The prognostic risk assessment model was established by univariate and multivariate Cox regression analysis, and the effectiveness of the model was evaluated by receiver operating characteristic (ROC) curve. The nomogram for predicting 3-, 5-, and 10-year survival rates of patients was plotted with clinical variables. Results A total of 22 types of immune cells were involved. The relative infiltration rates of 8 types of immune cell subsets (naive B cells, plasma cells, CD4 memory activated T cells, follicular helper T cells, regulatory T cells, M0 macrophage, M1 macrophage, and resting dendritic cells) were significantly lower than those in lung squamous carcinoma tissue (P<0.05 or P<0.01), while the infiltration rates of other 8 types (CD4 memory resting T cells, resting natural killer (NK) cells, monocytes, M2 macrophages, activated dendritic cells, resting mast cells, eosinophils, and neutrophils) were significantly higher (P<0.01 or P<0.05); and there were no significant differences in the rest of immune cell subsets (memory B cells, CD8 T cells, CD4 naive T cells, γ/δ T cells, activated NK cells, or activated mast cells) (all P>0.05). CD4 naive T cells and CD4 memory resting T cells were risk factors in patients with lung squamous cell carcinoma (HR>1), while CD4 memory activated T cells, follicular helper T cells, and resting dendritic cells were protective factors (HR<1). The prognosis of patients with high infiltration rates of CD4 memory activated T cells and resting dendritic cells was better than those of patients with low rates (P<0.05). The prognostic risk assessment model of patients with lung squamous cell carcinoma constructed by CD4 memory activated T cells, follicular helper T cells and resting dendritic cells was effective, with the area under curve value of (ROC) curve being 0.678. Conclusion Resting dendritic cells and CD4 memory activated T cells are associated with the occurrence and prognosis of lung squamous cell carcinoma, and can be used as independent prognostic factors. The prognostic risk assessment model constructed by prognosis related immune cell subsets is effective.