Objective To investigate the application of susceptibility-weighted imaging （SWI） in the diagnosis of intracranial arterial thrombosis in patients with hyperacute cerebral infarction.Methods A retrospective analysis was performed for 46 patients with hyperacute cerebral infarction within 6 hours after onset. Plain CT scan of the head， conventional magnetic resonance imaging， diffusion-weighted imaging， and SWI were performed within 6 hours after onset， and then digital subtraction angiography （DSA） was performed. Cerebral hyperdense artery sign （CHAS） on CT and susceptibility vessel sign （SVS） on SWI were analyzed in terms of their sensitivity and specificity in the diagnosis of arterial thrombosis， and the Kappa test was used to investigate the consistency between CHAS and SVS.Results Of all 46 patients， 15 were diagnosed with cerebral arterial thrombosis by DSA. With DSA as the reference standard， CHAS had a sensitivity of 86.7% （13/15）， a specificity of 93.5% （29/31）， a diagnostic efficiency of 91.3% （42/46）， a positive predictive value of 86.7% （13/15）， and a negative predictive value of 93.5% （29/31） in the diagnosis of cerebral arterial thrombosis， while SVS had a sensitivity of 100% （15/15）， a specificity of 93.5% （29/31）， a diagnostic efficiency of 95.7% （44/46）， a positive predictive value of 91.3% （42/46）， and a negative predictive value of 100% （29/29）. CHAS and SVS had good consistency in the diagnosis of arterial thrombosis （k=0.758， P <0.05）.Conclusions SVS has an important value in displaying cerebral arterial thrombosis， and SWI can be used routinely in patients with hyperacute cerebral infarction.