The present invention provides a medical device 100 that includes a processor 134 and a memory 183 that includes machine-executable instructions 140. Execution of the machine-executable instructions causes the processor to receive a first magnetic resonance image data set 146 describing a first region of interest 122 of the subject 118 (200) and describe a second region of interest 124 of the subject. At least one second magnetic resonance image data set 152, 152 'is received (202). The first region of interest at least partially includes the second region of interest. Execution of the machine-executable instructions further causes the processor to receive an analysis region 126 in both the first region of interest and the second region of interest (204). Executing the machine-executable instructions further causes the processor to create a cost function that includes an intra-scan uniformity measure separately for each of the first magnetic resonance imaging dataset and the at least one second magnetic resonance imaging dataset. 206). The cost function further includes an inter-scan similarity measure calculated using both the first magnetic resonance image data set and each of the at least one second magnetic resonance image data set. The execution of the machine-executable instructions further causes the processor to use an intensity correction algorithm to generate a first intensity correction map 153 for the first magnetic resonance image data set in the analysis region and at least one second map in the analysis region. Optimize the cost function by calculating at least one second intensity correction map 156 for each of the magnetic resonance image data sets (208). Execution of the machine-executable instructions further causes the processor to calculate a first corrected magnetic resonance image 158 describing the analysis region using the first magnetic resonance image data set and the first intensity correction map ( 210). The execution of the machine-exec
