Ischemic stroke and vascular dementia, as common cerebrovascular diseases, with the former causing irreversible neurological damage and the latter causing cognitive and memory impairment, are closely related and have long received widespread attention. Currently, the potential causative genes of these two diseases have yet to be investigated, and effective early diagnostic tools for the diseases have not yet emerged. In this study, we screened new potential biomarkers and analyzed new therapeutic targets for both diseases from the perspective of immune infiltration. Two gene expression profiles on ischemic stroke and vascular dementia were obtained from the NCBI GEO database, and key genes were identified by LASSO regression and SVM-RFE algorithms, and key genes were analyzed by GO and KEGG enrichment. The CIBERSORT algorithm was applied to the gene expression profile species of the two diseases to quantify the 24 subpopulations of immune cells. Moreover, logistic regression modeling analysis was applied to illustrate the stability of the key genes in the diagnosis. Finally, the key genes were validated using RT-PCR assay. A total of 105 intersecting DEGs genes were obtained in the 2 sets of GEO datasets, and bioinformatics functional analysis of the intersecting DEGs genes showed that GO was mainly involved in the purine ribonucleoside triphosphate metabolic process，respiratory chain complex，DNA-binding transcription factor binding and active transmembrane transporter activity. KEGG is mainly involved in the Oxidative phosphorylation, cAMP signaling pathway. The LASSO regression algorithm and SVM-RFE algorithm finally obtained three genes, GAS2L1, ARHGEF40 and PFKFB3, and the logistic regression prediction model determined that the three genes, GAS2L1 (AUC: 0.882), ARHGEF40 (AUC: 0.867) and PFKFB3 (AUC: 0.869), had good diagnostic performance. Meanwhile, the two disease core genes and immune infiltration were closely related, GAS2L1 and PFKFB3 had the highest positive correlation with macrophage M1 (p < 0.001) and the highest negative correlation with mast cell activation (p = 0.0017); ARHGEF40 had the highest positive correlation with macrophage M1 and B cells naive (p < 0.001), the highest negative correlation with B cell memory highest correlation (p = 0.0047). RT-PCR results showed that the relative mRNA expression levels of GAS2L1, ARHGEF40, and PFKFB3 were significantly elevated in the populations of both disease groups (p < 0.05). Immune infiltration-based models can be used to predict the diagnosis of patients with ischemic stroke and vascular dementia and provide a new perspective on the early diagnosis and treatment of both diseases.

Polycrystalline γ-LiAlO₂ pellets were sequentially irradiated with 120 keV He⁺ and 80 keV D₂⁺ ions to the fluences of 1 × 10¹⁷ and 2 × 10¹⁷ (He⁺+ D⁺)/cm² at 573 K. Additional irradiation was performed to a fluence of 2 × 10¹⁷ (He⁺+ D⁺)/cm² at 773 K. The irradiated pellets were characterized using scanning transmission electron microscopy, time-of-flight secondary ion mass spectrometry, and grazing incidence x-ray diffraction. Lattice damage, amorphization, and fractures are observed with no evidence for the formation of secondary-phase precipitates in the pellets irradiated up to an ion fluence of 2 × 10¹⁷ (He⁺+ D⁺)/cm² at 573 K. In contrast, faceted precipitates with sizes larger than 100 nm formed in a pellet irradiated to 2 × 10¹⁷ (He⁺+ D⁺)/cm² at 773 K. Analyses of the diffraction and composition data suggest that the precipitates have a spinel-type structure, likely a non-stoichiometric LiAl₅O₈ with Li depletion. This could be an intermediate phase with Li atoms at the octahedral and possibly tetrahedral sites as well. It is speculated that as the dose increases, Li loss will continue and the precipitates will approach a composition of alumina primarily in phases of α-Al₂O₃ and amorphized Al₂O₃.