In order to increase the accuracy of image guided neurosurgery (IGNS), we developed a brain deformation corrective system (BDCS) based on linear elastic model. To investigate the fidelity of such a system, two cases of cerebral tumor resection were conducted in our paper. According to our experiments, the average shift error after craniotomy was reduced to (1.00 +/- 0.63)mm from (4. 97 +/- 3. 53)mm and the average prediction rate was (72.27 +/- 20.37)%. The results showed that the system was convenient and fast to operate and presented a promising pathway to compensating for brain deformation during neurosurgery.