Susceptibility-weighted imaging (SWI) is a magnetic resonance imaging (MRI) sequence used for evaluating traumatic brain injury (TBI). Although SWI is being increasingly used in veterinary medicine, there are no systematic studies regarding its use. We aimed to evaluate TBI lesions by using conventional MRI and SWI in 11 dogs and determine the correlation between clinical status and conventional MRI or SWI findings. The modified Glasgow coma scale (MGCS) at presentation and a previously used MRI grading system (MRGr; grades 1–6) were used to evaluate the brain lesions, and correlations between MGCS score and each MRGr were assessed. Conventional MRI revealed 23 lesions in 11 dogs with variable MGCS scores (range: 11–17). SWI showed comparable findings for all of the lesions except for subdural hemorrhage, and it revealed additional lesions in four dogs. The median MRGr was 2 on both conventional MRI and SWI. The MRGr of the conventional MRI assessments and the MGCS scores showed a significant negative correlation (
r
= −0.685). In conclusion, SWI had better TBI lesion-detection ability, but conventional MRI had a better correlation with early clinical status and subdural hemorrhage. Thus, a combination of conventional MRI and SWI examinations can improve TBI diagnosis in dogs.