Objective：To compare the diagnostic value of volume CT digital subtraction angiography（VCTDSA） and matched mask bone elimination（MMBE） for intracranial arteriovenous malformation（AVM）. Methods：A retrospective analysis was performed on the clinical and imaging data of 37 patients who underwent head or head-and-neck computed tomographic angiography（CTA） and was diagnosed with intracranial AVM by digital subtraction angiography（DSA） from February 2015 to February 2018 in our hospital. The original images were reconstructed by the same operator using VCTDSA and MMBE. The image quality of malformed vessels，feeding arteries，and draining veins，and their automatic subtraction image and manual subtraction image were evaluated，and the post-processing time was compared. Results：There was a significant difference in the automatic，manual，and total subtraction time between VCTDSA and MMBE（P=0.000）. VCTDSA had a shorter time of automatic subtraction than MMBE，but a longer time of manual and total subtraction than MMBE. In 37 patients with intracranial AVM，the image quality of automatic subtraction was mainly classi－fied as grade Ⅱ by MMBE（59.5%，22/37） and grade Ⅲ by VCTDSA（62.2%，23/37）；there was a significant difference between the two methods. The image quality of manual subtraction was mainly classified as grade Ⅰ by VCTDSA（56.8%，21/37） and grade Ⅱ by MMBE（62.2%，23/37）；there was a significant difference between the two methods（P=0.011）. In 19 patients with AVM located in the superficial brain parenchyma，VCTDSA had significantly better image quality of the draining veins than MMBE（P=0.035）. There was no significant difference in the image quality of mal－formed vessels and feeding arteries of all the 37 patients between VCTDSA and MMBE（P=0.317 and 0.739）. Conclusion：VCTDSA has better image quality than MMBE for the AVM located in superficial brain parenchyma，especially the draining veins. However，MMBE subtracts the bone thoroughly，has better image quality of automatic subtraction，and can save more time in post-processing.