Stroke has a high incidence rate and often leads to permanent disability, particularly if it is not treated promptly. However, no blood biomarkers for early diagnosis are available to date. Therefore, we sought to detect stroke-specific blood biomarkers by identifying proteins associated with the underlying coagulation mechanism, which accounts for more than 80% of all stroke cases. Protein profiling was performed using blood samples from 16 healthy controls and 18 patients who suffered a stroke as the discovery set. We identified upregulated proteins (> 1.5-fold change and p value < 0.05) in patients who suffered a stroke relative to the corresponding levels in healthy controls by nano-liquid chromatography-tandem mass spectrometry using data-independent acquisition based on sequential window acquisition of all theoretical mass spectra, which was developed to improve the consistency and accuracy of candidate proteins. Pathway analysis confirmed that the upregulated proteins were mainly involved in blood coagulation. Among these, we selected prothrombin, plasminogen, fibrinogen alpha-chain, and histidine-rich glycoprotein as candidate biomarkers. Multiple reaction monitoring analysis was performed on a validation set of 61 serum samples (31 healthy controls and 30 stroke patients) to assess the diagnostic value of the candidate biomarkers. All four proteins showed higher expression levels in patients with stroke than in healthy controls. The areas under the receiver operating characteristic curve were greater than 0.9, confirming their clinical value. These four blood coagulation proteins may help in diagnosing stroke more accurately and quickly.