Skip to main content

Table 1 Table summarizing performance over different substitute CT approaches

From: A review of substitute CT generation for MRI-only radiation therapy

Approach category MRI sequence/contrast Number Site Performance metric Note References
main sub 1 sub 2 sub 3     ΔDose [%] MAE [HU] DSC bone Other   
Voxel Semi-automatic T1w   20 brain     1 ΔROI [61]
Threshold dUTE   1 phantom 2   0.81 87 Cbone [6264]
dUTE   5-19 brain 0.5 230 0.49-0.65 90 Ctissue [22, 35, 65]
Dixon   2 brain     94/77 SSbone [66]
dUTE Dixon 6-98 brain    0.75 88/88,81 SSbone, Ctissue [6769]
Probabilistic Clustering Fuzzy c-means T2w Dixon 2-5 brain     97-98/75-94 SSbone [70, 71]
T2w T1w 10 brain     98/74 SSair [52]
   dUTE Dixon 9 brain   130     [72]
Bayesian Markow RF dUTE   5 brain 1 204-247 0.53-0.59    [22]
Regression Discriminant dUTE T2w 3 phantom 2.3 88     [73]
dUTE T2w 3 brain 1.2 153     [74]
Gaussian dUTE T2w 5-9 brain 0.9-1.5 137-140 0.85 68-94,98 γ11, γ33 [23, 7577]
dUTE   5 brain 1 136-148 0.67-0.72    [22, 48]
Random F dUTE   5 brain 1 128 0.74    [22]
Semi-automatic
Dual
T1w T2w 9 prostate 0.7 75 0.91 99.9 a, γ22 [78]
T1w T2*w 10-15 prostate 0.3-2 135   93 γ11 [21, 79, 80]
PCA T1w   10 torso     5 Δbone [81]
Sinogram T1w   10 brain    0.85   b [82]
Neural network T1w   3 brain    0.78    [83]
dUTE   4 brain    0.83    [84]
Pattern recognition dUTE Dixon 10 brain     76 Cair [85]
Hybrid Neural Network Template dUTE   4 brain    0.77    [86]
Gaussian Spatial info dUTE T2w 9 brain   130     [75]
Random F Spatial info T1w   9-10 brain    0.92-0.98   c [87, 88]
Atlas Pattern recognition Patch
Patch Probabilistic
T1w   5
40
brain 0.5 85 0.84
0.75
   [48]
[89]
Deformable T1w   28 whole body     0.88 DSCbrain [90]
T1w T2w 5-17 brain 1-2.5 97-114 0.63-0.83 1.7 ΔROI [24, 48, 9194]
T2w   37 prostate 1.5   0.79    [25]
  T2w   10 cervix 0.3      [95]
Hybrid Deformable Patch T1w   17 brain   101     [96]
     T2w   39 Prostate 0.3 40.5 0.91 100 γ22 [97]
Hybrid Deformable Probabilistic   T1w   9-27 brain   126 0.86 86/90 SSbone [26, 98]
   T2w   10 prostate 0.2 36.5   99.9 γ21 [27]
Regression Random F T2w   20 prostate     0.83 DSCprostate [99]
   Threshold   dUTE   154 brain    0.81    [100]
  1. Column 1: Overall approach: Voxel, atlas or hybrid. Column 2: Sub categories within each main approach. Column 3: MRI sequences/contrasts applied in the studies. Column 4: Number of subjects included in the studies (patients, volunteers, phantoms). Column 5: Anatomical site investigated. Column 6: Performance metrics. Column 7: Specification of “Other” metric and comments. Column 8: References to included studies. SSx = % specificity @ % sensitivity for tissue x, Cx = % correctly classified voxel for tissue x, Δx = mm distance bt. CT and sCT for tissue x, γxy = % of points with γx %ymm < 1 where x and y are the dosimetric and geometric deviations, respectively, aMAE of whole FOV and DSC based on 2D DRR, b Overlap ratio similar to DSCbone. c Bone > 600 HU. Method abbreviations: Bayesian Bayesian statistics, Markow RF Markow Random Fields, Discriminant Discriminant analysis, Random F Random Forest, PCA Principal Component Analysis, Patch = cluster/collection of MRI voxels, Deformable = deformable registration