Fix, Don't Retake: Physics-Simulated Training for Correcting Patient Positioning Errors in Dental Panoramic Radiographs
Authors: Sema Helali, Lina Abu Nada, Faleh Tamimi, Rafat Damseh
Organizations: Department of Computer Science and Software Engineering, College of Computing and Artificial Intelligence, United Arab Emirates University, UAE · Faculty of Dental Medicine and Oral Health Sciences, McGill University, Canada · Department of Oral and Craniofacial Health Sciences, College of Dental Medicine, University of Sharjah, UAE
Patient-positioning errors can distort panoramic radiographs and lead to repeat examinations. This study investigated artificial intelligence for correcting these errors in existing images. A physics-based simulator generated paired correctly positioned and mispositioned panoramics from cone-beam computed tomography volumes, with controlled individual and combined positioning errors. Five approaches were compared: U-Net, pix2pix, conditional diffusion, an error-conditioned model, and a mixture of error-specific specialists. Evaluation used image similarity and geometric metrics across three test sets. All approaches improved mean image similarity and reduced geometric errors, with U-Net and the two error-aware approaches outperforming pix2pix and conditional diffusion.
Figures & tables
Figure 1: Overview of paired panoramic data generation from CBCT volumes, including positioning errors and physics-based panoramic simulation.
Error
Mild
Moderate
Severe
Forward/backward (mm)
2–4
4–7
7–10
Chin down/up ( ∘ )
3–6
6–10
10–15
Head turn ( ∘ )
4.5–6
6–9
9–12
Head tilt ( ∘ )
2–4
4–6
6–9
Tongue air gap (mm)
1.5–3
3–6
6–10
Table 1: Severity ranges of simulated positioning errors.
Natural
Balanced
Unseen geometry M4
Approach
PSNR ↑
SSIM ↑
IWE ↓
LRE ↓
PSNR ↑
SSIM ↑
IWE ↓
LRE ↓
PSNR ↑
SSIM ↑
IWE ↓
LRE ↓
Uncorrected
18.81
.577
5.57
5.99
16.64
.458
5.91
6.62
17.90
.550
5.32
6.73
U-Net
23.25
.703
2.64
2.04
21.59
.659
1.78
1.93
21.25
.649
3.10
2.93
pix2pix
21.51
.634
3.79
3.17
19.72
.566
2.43
3.00
19.85
.586
4.01
4.33
Cond. diffusion
21.18
.624
3.62
3.30
19.32
.565
2.68
2.76
19.31
.562
3.96
3.62
Error-conditioned
23.62
.705
2.76
1.95
21.97
.667
2.33
1.80
21.54
.656
3.11
2.65
Table 2: Mean performance across the three test sets. PSNR is in dB; incisor-width (IWE) and left/right (LRE) errors are percentages. Higher ( ↑ ) or lower ( ↓ ) values are better. Best results are bold.
Figure 2: Qualitative comparison for three positioning errors. Overlays show the reference in green and the compared image in magenta; gray indicates agreement.
Level
Approach
PSNR ↑
SSIM ↑
IWE ↓
LRE ↓
Mild
Uncorrected
18.55
.516
2.45
3.35
U-Net
22.51
.692
2.08
1.61
pix2pix
20.77
.616
2.01
2.28
Cond. diffusion
20.35
.607
2.54
1.68
Error-conditioned
23.04
.698
2.53
1.35
Mixture of specialists
23.15
.712
2.21
1.44
Table 3: Mean results by severity on the balanced set (450 images per level). Best results within each level are bold.
Interest in generating realistic synthetic medical images has grown in recent years. Such datasets can mitigate the scarcity of public data for artificial intelligence research and support education. Work on synthesizing dental panoramic radiographs (PRs) has so far relied almost exclusively on generative adversarial networks (GANs), and three limitations recur: the synthesized field of view is often restricted to a cropped dentoalveolar region rather than a complete PR; observer studies of GAN-generated PRs report that clinicians still identify synthetic images with accuracies of roughly 78-82%, implying persistent visible giveaways; and adversarial training remains sensitive to instability and mode collapse. Generating a full-size PR directly at diagnostic resolution with a diffusion model is an alternative, but sampling at that resolution is costly. We therefore present PanoDiff-SR, an end-to-end two-stage pipeline developed and evaluated specifically for panoramic radiography: a denoising diffusion implicit model (PanoDiff) generates a low-resolution (LR) seed PR (256×128), which a fine-tuned hybrid attention transformer (HAT) super-resolution (SR) stage upscales to a high-resolution (HR) PR of 1024×512. Decoupling the expensive generative step from the resolution step yields complete, full-arch PRs at diagnostic resolution on a single 47 GB GPU. The Fréchet inception distance between 7243 real and synthetic HR images is 40.5. Inception scores were 2.68, 2.38, 2.90 and 2.98 for real HR, synthetic HR, real LR and synthetic LR images, respectively. Six clinical experts, each evaluating 100 synthetic and 100 real PRs under a time limit, distinguished real from synthetic images with an average accuracy of 68.5% (50% corresponds to random guessing), substantially closer to chance than reported for GAN-based studies. Project page: https://s4nyam.github.io/panodiff/
Sanyam Jain, Bruna Neves de Freitas, Andreas Basse-OConnor +2
Department of Dentistry and Oral Health, Aarhus University, 8000 Aarhus, Denmark · Aarhus Institute of Advanced Studies, Aarhus University, 8000 Aarhus, Denmark · Department of Mathematics, Aarhus University, 8000 Aarhus, Denmark +1
Panoramic dental radiographs require fine-grained spatial reasoning, bilateral symmetry understanding, and multi-step diagnostic verification, yet existing vision-language models operate under a static single-pass paradigm that limits their clinical reliability. In this paper, we introduce OralGPT-Plus, an agentic vision-language model designed to perform iterative and symmetry-aware diagnostic reasoning for panoramic dental radiograph analysis. To support this paradigm, we construct DentalProbe, a five-thousand-image dataset with expert-curated diagnostic trajectories that provide structured supervision for localized inspection and contralateral comparison. We further develop a Reinspection-driven reinforcement learning framework that encourages clinically meaningful re-examination and stabilizes long-horizon reasoning with rubric-based reward and conditioned diagnostic-driven reward. In parallel, we present MMOral-X, the first benchmark for holistic panoramic diagnosis, containing 300 open-ended questions and region-level annotations across multiple difficulty levels. OralGPT-Plus demonstrates consistent and reliable improvements over strong baselines on MMOral-X and established panoramic benchmarks, indicating the effectiveness of interactive and symmetry-informed reasoning. Our work highlights the value of agentic modeling for dental imaging and provides a foundation for future research in clinically aligned panoramic radiograph analysis.
Yuxuan Fan, Jing Hao, Hong Chen +5
The Hong Kong University of Science and Technology (GZ) · School of Computer Science, Peking University · Faculty of Dentistry, The University of Hong Kong +3
An adequate diagnostic quality of radiographs is essential for reliable diagnoses and treatment planning. The patient's pose during radiography is one of the most important factors determining the diagnostic quality. Since patient positioning is difficult and not standardized, an automated AI-based approach using depth images to automatically assess the patient's pose before the radiograph has been taken would be helpful. Due to regulatory hurdles, however, it is difficult in practice to acquire the required depth images and corresponding radiographs. In this paper, we present a framework that can generate such training data synthetically from Computed Tomography scans. We further show that by pretraining on our generated synthetic dataset consisting of 3077 image pairs of upper ankle joints, the pose assessment of real upper ankle joints can be improved by up to 11 percentage points.
Manuel Laufer, Dominik Mairhöfer, Malte Sieren +7
Institute for Neuro- and Bioinformatics, University of Lübeck, Germany · University Medical Center Schleswig-Holstein, Lübeck, Germany · IMAGE Information Systems Europe GmbH, Rostock, Germany +1