Purpose: Early detection and diagnosis are the keys to improving the survival rate and reducing the fatality rate linked to gastric cancer. The precancerous lesion of gastric cancer is referred to as Gastric high-grade intraepithelial neoplasia (HGIN). Both sensitivity and specificity of current biomarkers that aid in the diagnosis of gastric HGIN are still relatively low. The proteomic data of gastric HGIN are still scarce. The study aimed to explore candidate protein biomarkers for gastric HGIN screening with proteomics and bioinformatics technology. Methods: A total of 10 serum samples were collected and categorized into two groups, that is, the gastric HGIN and the healthy control groups. Label-free quantification (LFQ) in conjunction with liquid chromatography with tandem mass spectrometry (LC-MS/MS) was employed to identify the probable biomarkers for gastric HGIN. Further, differentially expressed proteins (DEPs) were quantified by the proteomic analysis. Results: In total, 1192 distinct serum proteins were discovered between the gastric HGIN group and the healthy control group. DEPs were recruited in the conduction of further analyses, utilizing a threshold of 1.5-fold difference in expression level (P ＜ 0.05) in comparison with the control group. There were 18 upregulated and 12 downregulated proteins in the gastric HGIN group in contrast with the control group. Bioinformatics analyses were performed according to Gene Ontology (GO), cluster, and KEGG pathway enrichment analyses. Conclusions: Our data provide more biological information for the formation of gastric HGIN and provide clues for further research on the pathogenesis of early gastric cancer.