The blend sign on CT imaging is a key predictor of hematoma growth in intracerebral hemorrhage (ICH), yet the underlying pathological mechanisms remain poorly understood. To address this, hematoma specimens from nine ICH patients with blend signs were collected via stereotactic minimally invasive puncture, targeting both high-density and low-density bleeding regions. Proteins were extracted and analyzed using isotope-labeled iTRAQ-based quantitative proteomics combined with LC-MS/MS. Bioinformatic analysis identified differentially expressed proteins, associated pathways, and regulatory microRNAs, followed by validation with Western blotting and ELISA. A total of 72 differentially expressed proteins were identified, with enrichment in processes including inflammation, apoptosis, oxidative stress, and metabolism. Key molecules such as cytochrome C, growth-associated protein 43, and tau were significantly upregulated in high-density regions. Integration with microRNA prediction suggested regulatory interactions, and drug screening highlighted potential therapeutic candidates such as dexketoprofen and 2-mercaptoethanesulfonic acid. This dataset provides raw mass spectrometry files and processed quantification results, offering a valuable resource for exploring the molecular basis of hematoma expansion and for identifying potential biomarkers and therapeutic targets in ICH.