Learned Regularization for Sparse-View CT: Accuracy, Lesion Detectability and Distribution Shift of a Plug-and-Play Prior against Tuned Total Variation
Organizations: Independent research
Abstract
Learned image priors are often proposed to replace hand-crafted regularizers in sparse-view computed tomography (CT), yet they are usually judged by PSNR on data resembling their training set. We evaluate one noise-conditioned DnCNN denoiser used as a Plug-and-Play prior against filtered back-projection (FBP), Tikhonov and total-variation (TV) regularization, with every method tuned on validation data and all differences paired. On synthetic 28x28 digit images, the prior beats box-constrained TV by about 3.4 dB at moderate noise (sigma = 0.05, 20 angles), but the advantage shrinks at low noise and reverses under distribution shift: an MNIST-trained prior is 3.1 dB (Fashion-MNIST) and 4.3-4.6 dB (Shepp-Logan phantom) below TV, whereas a Fashion-MNIST-trained prior beats TV by 2.2 dB on Fashion-MNIST, consistent with prior mismatch rather than a failure of PnP itself. In a lesion-insertion test the prior detects lesions worse than TV in distribution despite its higher PSNR, and under mismatch it produces more lesion-like false blobs than box-constrained TV for seven of eight lesion types. On real LoDoPaB-CT slices (patient-disjoint split; 30, 60 and 120 angles; three dose levels; 40 test patients), the prior does not consistently outperform a well-tuned TV in PSNR (-0.93 to +0.33 dB) and has lower SSIM in eight of nine conditions. Its lesion-detection AUC is higher than this TV's in all nine conditions (significant in six) but never significantly higher than that of Hann-filtered FBP, which the PSNR-tuned TV falls below in detection in all nine conditions (seven significant). On non-anatomical phantoms its PSNR falls below TV in all six conditions. PSNR and lesion detectability thus rank the methods differently. Our conclusions concern one small denoiser, and the CT analysis was developed after seeing results, without pre-specified primary comparisons.
Figures & tables
| Setting | Images | Operator and noise | Data used | Experiments |
|---|---|---|---|---|
| Digits, Gaussian noise (Stage 1) | MNIST; Fashion-MNIST, EMNIST letters, Shepp–Logan for shift | dense Radon matrix, 40 bins, ; additive Gaussian noise on the normalised sinogram | prior: MNIST training split (59 500 images); tuning: 100 validation images; test: 200 images | E1–E10 |
| Digits, Poisson noise (Stage 2) | as above | matrix-free operator (same geometry); Poisson photon noise, | as above | E11–E13 |
| LoDoPaB-CT (Stage 3) | slices | matrix-free parallel-beam, 513 bins, of 1000 angles; Poisson noise, photons per detector bin and fractions of it | patient-disjoint: 8 960 training slices, 20 validation slices, 40 test slices (one per patient) | E14–E19 |
| Condition | FBP | Tikhonov | TV-ADMM | TV-box | PnP-DnCNN |
|---|---|---|---|---|---|
| , 20 angles | 24.2 | 28.9 | 36.0 | 42.5 † | 29.6 |
| , 20 angles | 23.1 | 24.1 | 29.0 | 29.8 | 29.1 |
| , 20 angles | 15.5 | 17.1 | 20.1 | 20.7 | 24.0 |
| , 20 angles | 10.4 | 14.7 | 16.9 | 17.2 | 19.9 |
| , 20 angles | 6.6 | 12.9 | 14.3 | 14.6 | 15.3 |
| 5 angles, | 9.1 | 14.2 | 15.5 | 16.5 | 17.7 |
| Inverse crime | Finer grid (bicubic) | |||
| Condition | vs TV-ADMM | vs TV-box | vs TV-ADMM | vs TV-box |
| , 20 angles | [ ] | [ ] † | [ ] | [ ] |
| , 20 angles | [ ] | [ ] | [ ] | [ ] |
| , 20 angles | [ ] | [ ] | [ ] | [ ] |
| , 20 angles | [ ] | [ ] | [ ] | [ ] |
| , 20 angles | [ ] | [ ] | [ ] | [ ] |
| Constant offset | Per-angle jitter ( = s.d.) | |||||
| TV | PnP | PnP TV | TV | PnP | PnP TV | |
| 20.15 | 24.00 | 20.15 | 24.00 | |||
| 20.13 | 23.91 | 20.14 | 23.99 | |||
| 20.04 | 23.62 | 20.10 | 23.93 | |||
| 19.71 | 22.61 | 19.99 | 23.65 | |||
| 18.56 | 20.05 | 19.19 | 22.18 | |||
| TV | PnP | ||||
|---|---|---|---|---|---|
| vs true | vs rotated | vs true | vs rotated | Ceiling | |
| 20.15 | 20.15 | 24.00 | 24.00 | ||
| 20.04 | 20.36 | 23.62 | 24.14 | 35.9 | |
| 19.71 | 20.66 | 22.61 | 24.23 | 29.0 | |
| 18.56 | 20.99 | 20.05 | 24.14 | 22.1 | |
| PSNR (dB) | SSIM | |||||
|---|---|---|---|---|---|---|
| Test set | TV | PnP | gap (wins) | TV | PnP | |
| MNIST (in-distribution) | 20.27 | 23.97 | (100%) | 0.742 | 0.944 | |
| EMNIST letters (near shift) | 20.09 | 22.03 | (92%) | 0.842 | 0.938 | |
| Fashion-MNIST (far shift) | 19.61 | 16.57 | (11%) | 0.680 | 0.675 | |
| Shepp–Logan (far shift) | 19.19 | 15.03 | (0%) | 0.722 | 0.517 | |
| Test set | TV-ADMM | TV-box | PnP | PnP TV-ADMM | PnP TV-box |
|---|---|---|---|---|---|
| MNIST | 20.21 | 20.69 | 24.05 | ||
| EMNIST letters | 20.28 | 20.35 | 22.07 | ||
| Fashion-MNIST | 19.42 | 19.45 | 16.32 | ||
| Shepp–Logan | 19.28 | 19.50 | 14.96 |
| Test set | TV-ADMM | PnP | PnP | best | FBP |
|---|---|---|---|---|---|
| MNIST (in-distribution) | 20.27 | 23.97 | 22.09 | 23.97 (0.1, 5) | 15.53 |
| EMNIST letters | 20.09 | 22.03 | 20.11 | 22.03 (0.1, 5) | 15.47 |
| Fashion-MNIST | 19.61 | 16.57 | 13.80 | 17.86 (0.2, 1) | 14.94 |
| Shepp–Logan | 19.19 | 15.03 | 12.40 | 16.36 (0.1, 1) | 14.77 |
| Data model | Model error | |||||
| Same operator (inverse crime) | 0 | |||||
| Fine grid, nearest | 0.03 | |||||
| Fine grid, bicubic (default) | 0.09 | |||||
| Fine grid, bilinear | 0.10 | |||||
| Inverse crime, vs TV-box | 0 | |||||
| Bicubic, vs TV-box | 0.09 |
| Data model | 5 | 10 | 20 | 40 | 90 |
|---|---|---|---|---|---|
| Same operator (inverse crime) | |||||
| Fine grid, nearest | |||||
| Fine grid, bicubic (default) | |||||
| Fine grid, bilinear |
| Condition | FBP | Tikhonov | TV-ADMM | PnP-DnCNN |
|---|---|---|---|---|
| , 20 angles | 24.3 | 27.4 | 33.2 | 29.9 |
| , 20 angles | 23.2 | 23.9 | 28.4 | 29.4 |
| , 20 angles | 15.5 | 17.1 | 20.1 | 24.1 |
| , 20 angles | 10.4 | 14.7 | 16.8 | 19.9 |
| , 20 angles | 6.6 | 12.9 | 14.3 | 15.3 |
| 5 angles, | 9.1 | 14.1 | 15.5 | 17.8 |
| error ratio PnP/TV | |||||||
|---|---|---|---|---|---|---|---|
| Test set | in | in | total | TV | PnP | TV | PnP |
| MNIST (in-dist.) | 0.48 | 0.40 | 0.42 | 0.58 | 0.76 | 0.45 | 0.76 |
| EMNIST letters | 0.70 | 0.64 | 0.65 | 1.22 | 1.55 | 0.43 | 0.67 |
| Fashion-MNIST | 1.27 | 2.32 | 2.05 | 0.29 | 0.80 | 0.36 | 0.37 |
| Shepp–Logan | 2.59 | 2.64 | 2.63 | 0.44 | 3.27 | 0.09 | 0.23 |
| exact null | ||||||
|---|---|---|---|---|---|---|
| Test set | ratio | ratio | ratio | |||
| MNIST (in-dist.) | 0.49 | 0.72 | 0.48 | 0.76 | 0.47 | 0.76 |
| EMNIST letters | 0.68 | 1.56 | 0.70 | 1.55 | 0.70 | 1.56 |
| Fashion-MNIST | 1.21 | 0.70 | 1.27 | 0.80 | 1.27 | 0.83 |
| Shepp–Logan | 2.64 | 3.68 | 2.59 | 3.27 | 2.21 | 2.74 |
| Inverse crime | Bilinear fine grid | |||||
|---|---|---|---|---|---|---|
| Test set | best | best | ||||
| MNIST (in-dist.) | 0.43 | 0.43 (5) | 0.71 | 0.43 | 0.42 (5) | 0.65 |
| EMNIST letters | 0.64 | 0.64 (5) | 1.01 | 0.66 | 0.62 (5) | 0.93 |
| Fashion-MNIST | 2.16 | 1.56 (2) | 4.13 | 2.65 | 1.47 (2) | 3.71 |
| Shepp–Logan | 2.60 | 1.92 (1) | 4.77 | 3.44 | 1.89 (1) | 4.58 |
| TV | MNIST-trained (m03) | Fashion-trained (f03) | ||||||
|---|---|---|---|---|---|---|---|---|
| Test set | ADMM | box | PnP | gap | gap (box) | PnP | gap | gap (box) |
| MNIST | 20.21 | 20.69 | 24.05 | 21.60 | ||||
| EMNIST letters | 20.28 | 20.35 | 22.07 | 20.73 | ||||
| Fashion-MNIST | 19.42 | 19.45 | 16.32 | 21.63 | ||||
| Shepp–Logan | 19.28 | 19.50 | 14.96 | 19.10 | ||||
| pAUC | Ret. | FL | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Prior test set | TV | box | PnP | TV | box | PnP | TV | box | PnP | |
| MNIST MNIST (in-dist.) | 0.75/0.5 | 0.91 | 0.90 | 0.69 | 0.59 | 0.65 | 0.28 | 0.005 | 0.010 | 0.000 |
| 1.25/0.25 | 0.90 | 0.90 | 0.69 | 0.62 | 0.67 | 0.30 | 0.005 | 0.010 | 0.000 | |
| MNIST Fashion (shifted) | 0.75/0.5 | 0.91 | 0.92 | 0.82 | 0.63 | 0.74 | 0.99 | 0.030 | 0.090 | 0.430 |
| 1.25/0.25 | 0.93 | 0.91 | 0.76 | 0.59 | 0.69 | 0.87 | 0.030 | 0.095 | 0.428 | |
| Fashion Fashion (in-dist.) | 0.75/0.5 | 0.91 | 0.90 | 0.75 | 0.63 | 0.55 | 0.22 | 0.030 | 0.025 | 0.005 |
| Prior test set | Operating point | PSNR (dB) | pAUC | FL |
|---|---|---|---|---|
| MNIST MNIST (in-dist.) | TV (PSNR-tuned) | 20.11 | 0.910 | 0.005 |
| TV, lesion | 20.13 | 0.904 | 0.005 | |
| PnP, PSNR-tuned | 24.07 | 0.693 | 0.000 | |
| PnP, constrained | 20.84 | 0.849 | 0.060 | |
| PnP, lesion | 19.57 | 0.916 | 0.315 | |
| MNIST Fashion (shifted) | TV (PSNR-tuned) | 19.37 | 0.914 | 0.030 |
| Method | Rel. diff. | PSNR (dense = MF) | Time dense (s) | Time MF (s) |
|---|---|---|---|---|
| Tikhonov | 16.36 | 0.08 | 0.26 | |
| TV-ADMM | 19.01 | 0.16 | 7.50 | |
| PnP ( ) | 23.10 | 0.76 | 0.94 |
| (photons/ray) | 1000 | 300 | 100 | 30 | 10 |
|---|---|---|---|---|---|
| effective | 0.018 | 0.032 | 0.056 | 0.106 | 0.198 |
| MNIST (train dist.) | (0.70) | (0.68) | (0.71) | (0.88) | (0.99) |
| EMNIST | (0.75) | (0.83) | (0.86) | (1.02) | (1.00) |
| Fashion-MNIST | (1.04) | (1.18) | (1.18) | (1.19) | (1.07) |
| Shepp–Logan | (0.90) | (1.79) | (1.73) | (1.77) | (1.75) |
| (photons/ray) | 1000 | 300 | 100 | 30 | 10 |
|---|---|---|---|---|---|
| MNIST (train dist.) | (1.00) | (1.00) | (1.00) | (1.00) | (0.99) |
| EMNIST | (0.96) | (0.99) | (1.00) | (0.98) | (0.98) |
| Fashion-MNIST | (0.47) | (0.50) | (0.55) | (0.52) | (0.67) |
| Shepp–Logan |
| pAUC | Ret. | FL | pAUC | Ret. | FL | ||||||||
| Prior test set | TV | PnP | TV | PnP | TV | PnP | TV | PnP | TV | PnP | TV | PnP | |
| MNIST MNIST | 0.75/0.5 | .89 | .60 | .60 | .29 | .010 | .007 | .77 | .63 | .48 | .34 | .005 | .067 |
| 1.25/0.25 | .88 | .60 | .60 | .27 | .020 | .007 | .75 | .65 | .46 | .46 | .005 | .068 | |
| MNIST Fashion | 0.75/0.5 | .81 | .72 | .64 | .84 | .240 | .443 | .73 | .63 | .42 | .57 | .230 | .518 |
| 1.25/0.25 | .84 | .72 | .65 | .84 | .255 | .438 | .71 | .61 | .42 | .54 | .225 | .518 | |
| 30 | 60 | 120 | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| dose | 1 | 0.25 | 0.1 | 1 | 0.25 | 0.1 | 1 | 0.25 | 0.1 | |
| real | 29.34 | 27.95 | 26.57 | 30.15 | 28.40 | 26.73 | 30.61 | 28.54 | 27.89 | |
| simulated | 29.41 | 27.90 | 26.52 | 30.27 | 28.45 | 26.72 | 30.76 | 28.55 | 27.85 | |
| dose | Hann | TV | PnP-const | PnP-TVinit | |
|---|---|---|---|---|---|
| 30 | 1 | 0.2 | 0.003 | (0.02, 12) | (0.02, 4) |
| 30 | 0.25 | 0.16 | 0.01 | (0.05, 8) | (0.08, 4) |
| 30 | 0.1 | 0.12 | 0.03 | (0.08, 12) | (0.05, 8) |
| 60 | 1 | 0.3 | 0.003 | (0.02, 8) | (0.02, 2) |
| 60 | 0.25 | 0.2 | 0.01 | (0.08, 4) | (0.02, 8) |
| 60 | 0.1 | 0.12 | 0.03 | (0.05, 12) | (0.02, 12) |
| dose 1 | dose 0.25 | |||||
|---|---|---|---|---|---|---|
| Method | PSNR | SSIM | TV | PSNR | SSIM | TV |
| FBP-Hann | 30.65 | 0.738 | 28.73 | 0.699 | ||
| TV | 31.91 | 0.786 | – | 29.82 | 0.750 | – |
| PnP-const [d8] | 31.74 | 0.773 | 29.72 | 0.734 | ||
| PnP-TVinit [d8] | 31.94 | 0.784 | 30.03 | 0.753 | ||
| PnP-anneal [d8] | 31.26 | 0.769 | 29.94 | 0.750 | ||
| Method | 1 | 0.25 | 0.1 | 1 | 0.25 | 0.1 | 1 | 0.25 | 0.1 |
|---|---|---|---|---|---|---|---|---|---|
| PSNR (dB) | |||||||||
| FBP-Hann | 29.02 | 27.72 | 26.63 | 30.59 | 28.66 | 27.11 | 31.70 | 29.18 | 27.94 |
| TV | 30.98 | 29.15 | 27.69 | 32.04 | 29.83 | 27.93 | 32.64 | 30.07 | 29.13 |
| TV-long | 31.31 | 29.21 | 27.72 | 32.12 | 29.88 | 27.98 | 32.50 | 30.11 | 29.21 |
| PnP-const | 30.38 | 29.05 | 27.86 | 31.85 | 29.68 | 28.31 | 32.64 | 30.23 | 28.86 |
| dose | PSNR: PnP-const TV-long | SSIM: PnP-const TV-long | PSNR: PnP-TVinit TV-long | PSNR: PnP-const FBP-Hann | |
|---|---|---|---|---|---|
| 30 | 1 | ||||
| 30 | 0.25 | ||||
| 30 | 0.1 | ||||
| 60 | 1 | ||||
| 60 | 0.25 | ||||
| 60 | 0.1 |
| dose | PnP-const TV-long | PnP-const TV | PnP-const FBP-Hann | TV-long FBP-Hann | PnP-TVinit TV-long | |
|---|---|---|---|---|---|---|
| 30 | 1 | |||||
| 30 | 0.25 | |||||
| 30 | 0.1 | |||||
| 60 | 1 | |||||
| 60 | 0.25 | |||||
| 60 | 0.1 |
| dose | : PnP-const TV-long | : PnP-const TV | : PnP-const TV-long | |
|---|---|---|---|---|
| 30 | 1 | |||
| 30 | 0.25 | |||
| 30 | 0.1 | |||
| 60 | 1 | |||
| 60 | 0.25 | |||
| 60 | 0.1 |
| false blobs, (vs TV) | vs TV-long | |||||
|---|---|---|---|---|---|---|
| dose | AUC | PSNR | SSIM | |||
| 30 | 1 | |||||
| 30 | 0.25 | |||||
| 60 | 1 | n/a | ||||
| 60 | 0.25 | n/a | ||||
| 120 | 1 | |||||
| set | , dose | FBP-Hann | TV | TV-long | PnP-const | PnP-TVinit | PnP-const FBP |
|---|---|---|---|---|---|---|---|
| CT | 30, 0.25 | 9.78 | 5.56 | 6.06 | 0.75 | 0.64 | |
| CT | 30, 1 | 5.78 | 1.04 | 1.69 | 0.50 | 0.32 | |
| CT | 120, 0.25 | 5.79 | 0.02 | 0.02 | 1.38 | 0.39 | |
| CT | 120, 1 | 26.79 | 0.12 | 0.00 | 0.24 | 0.18 | |
| phantom | 30, 0.25 | 15.48 | 27.95 | 30.20 | 16.50 | 11.65 | |
| phantom | 30, 1 | 31.60 | 75.72 | 83.90 | 24.48 | 57.30 |
Sparse-View CT Reconstruction
Sparse-view computed tomography (CT) reconstruction recovers 3D images from limited X-ray projections while suppressing artifacts and preserving anatomy.
See all 56 papers