Abstract: Traditional bulk cell studies obscure the heterogeneity between individual cells, whereas single-cell omics provides a novel research perspective to reveal cellular diversity and its associated molecular mechanisms, offering a crucial support for the precise analysis of complex biological processes. To address the unavoidable sample loss in conventional proteomics sample preparation methods, an integrated sample processing platform was established for single-cells using active-matrix digital microfluidics (AM-DMF), that manipulated nanoliter level droplets as the reaction vessels. Combining AM-DMF and timsTOF Pro 2 mass spectrometer enabled low-loss and high-sensitivity single-cell proteomic analysis. In this study, the surfactant used in the AM-DMF platform was firstly optimized to ensure proper droplet movement and sorting on the chip surface, and chromatographic conditions were also optimized to determine the ideal column length for single-cell protein identification. Then, the AM-DMF platform was compared to cellenONE, which was another commonly used single-cell sorting and sample processing platform. The results showed that using the AM-DMF platform combined with data-independent acquisition (DIA), nearly 3 000 proteins can be identified on average from a single HeLa cell, with an average single-cell sorting time of 2 s. Compared to other sample preparation methods, the identification depth increases by 58%, and the single-cell sorting time is reduced by 74%. The quantitative results exhibit high reproducibility, demonstrating the stability and consistency of this method. These findings validate the advantages of the AM-DMF platform in terms of identification depth, sensitivity, and sample processing throughput. Finally, the single-cell proteomes of HepG2, A549, and HeLa cells were systematically compared on the AM-DMF platform. The results revealed that an average of 2 831, 2 541, and 2 403 proteins are identified in single HepG2, A549, and HeLa cells were systematically compared on the AM-DMF platform, respectively. Principal component analysis (PCA) showed that single-cell proteomic data from the same cell line cluster together, while those from different cell lines are clearly separated, indicating the high stability of AM-DMF in single-cell sample preparation and qualitative and quantitative consistency, preserving the intrinsic proteomic characteristics and differences of the cells. Furthermore, an analysis of variance (ANOVA) with adjusted p-values less than 0.05 identifies 149 proteins with significantly different abundances across the three cell types. Gene ontology (GO) enrichment analysis of differentially expressed proteins showed that three cell types are enriched in distinct biological processes. Through a comparative analysis of HepG2, A549 and HeLa cell lines at the single-cell proteome level, this study highlights distinct proteomic expression patterns across different cell types, demonstrating the potential of the AM-DMF platform for single-cell proteomics research. This approach is expected to provide technical support for further investigation in complex biological questions, such as cell differentiation, tumor heterogeneity, and the immune microenvironment, and also holds promise for applications in the analysis of even smaller samples, such as single-cell extracellular vesicles and secreted proteins, offering refined molecular-level insights for precision medicine, drug screening, and personalized therapy.