Best model Adj R2 Std Err of Est Function
Scanner A to B
V1V2
V6V7
V11V12
Aggregate		 Quadratic
Compound
Quadratic
Model
Averaging		 0.989
0.986
0.995		 0.577
0.035
0.329		 0.025*x*x-0.821*x+15.263
3.193*(1.050**x)
0.037*x*x-1.784*x+33.158
0.020*x*x-0.868*x+16.140+1.064*
(1.050**x)
Scanner A to C
V1V3
V6V8
V11V13
V16V18
Aggregate		 Quadratic
Quadratic
Quadratic
Quadratic
Quadratic		 0.988
0.987
0.996
0.992		 0.522
0.541
0.246
0.374		 0.025*x*x-0.955*x+28.032
0.019*x*x-0.571*x+22.036
0.028*x*x-1.342*x+36.825
0.025*x*x-1.091*x+31.752
0.024*x*x-0.990*x+29.661
Scanner A to D1200
V1V4
V6V9
V11V14
V16V19
V21V22
Aggregate		 Quadratic
Quadratic
Quadratic
Quadratic
Quadratic
Quadratic		 0.987
0.984
0.994
0.993
0.990		 0.389
0.456
0.213
0.271
0.266		 0.027*x*x-1.333*x+48.314
0.025*x*x-1.179*x+45.197
0.030*x*x-1.675*x+55.794
0.026*x*x-1.306*x+48.897
0.013*x*x-0.464*x+36.589
0.024*x*x-1.191*x+46.958
Scanner A to D600
V1V5
V6V10
V11V15
V16V20
V21V23
Aggregate		 Quadratic
Quadratic
Quadratic
Quadratic
Quadratic
Quadratic		 0.984
0.983
0.994
0.992
0.991		 0.371
0.397
0.181
0.235
0.203		 0.019*x*x-0.890*x+53.962
0.020*x*x-0.886*x+53.164
0.023*x*x-1.269*x+61.897
0.021*x*x-1.029*x+56.914
0.009*x*x-0.287*x+46.876
0.018*x*x-0.872*x+54.563
Scanner B to C
V2V3
V7V8
V12V13
Aggregate		 Quadratic
Quadratic
Quadratic
Quadratic		 0.998
0.999
0.997		 0.166
0.100
0.196		 0.043*x*x-0.349*x+21.570
0.043*x*x-0.354*x+21.814
0.030*x*x-0.191*x+22.162
0.039*x*x-0.298*x+21.849
Scanner B to D1200
V2V4
V7V9
V12V14
Aggregate		 Quadratic
Quadratic
Quadratic
Quadratic		 0.997
0.999
0.995		 0.148
0.112
0.207		 0.027*x*x-0.305*x+33.390
0.027*x*x-0.242*x+32.702
0.015*x*x-0.133*x+33.141
0.023*x*x-0.226*x+33.078
Scanner B to D600
V2V5
V7V10
V12V15
Aggregate		 Quadratic
Quadratic
Quadratic
Quadratic		 0.998
0.999
0.995		 0.118
0.062
0.155		 0.027*x*x-0.267*x+45.311
0.028*x*x-0.259*x+44.587
0.019*x*x-0.178*x+45.368
0.025*x*x-0.235*x+45.089
Scanner C to D1200
V3V4
V8V9
V13V14
V18V19
Aggregate		 Quadratic
Quadratic
Quadratic
Quadratic
Quadratic		 0.993
0.992
0.996
0.995		 0.271
0.304
0.190
0.201		 0.018*x*x-0.394*x+33.212
0.020*x*x-0.468*x+34.022
0.022*x*x-0.669*x+37.601
0.018*x*x-0.428*x+35.062
0.019*x*x-0.490*x+34.974
Scanner C to D600
V3V5
V8V10
V13V15
V18V20
Aggregate		 Quadratic
Quadratic
Quadratic
Quadratic
Quadratic		 0.991
0.989
0.994
0.996		 0.267
0.299
0.189
0.156		 0.009*x*x+40.911
0.014*x*x-0.267*x+43.841
0.015*x*x-0.405*x+46.810
0.013*x*x-0.297*x+45.628
0.013*x*x-0.242*x+44.298
Scanner D1200 to D600
V4V5
V9V10
V14V15
V19V20
V22V23
Aggregate		 Linear
Linear
Linear
Linear
Linear
Linear		 0.993
0.993
0.991
0.994
0.992		 0.278
0.274
0.284
0.229
0.292		 1.237*x+9.425
0.725*x+21.302
0.668*x+23.668
0.683*x+22.771
1.074*x+13.201
0.877*x+18.074
Table 4: Examples of fitted correction functions from Stage II and Stage III. Bold functions are aggregates of individual functions and these are the final voxel-based correction functions. Functions are only applicable for the valid ranges of voxel based MTRs given in Table 1.