Text-Conditioned Multi-Expert Regression Framework for Fully Automated Multi-Abutment Design
Authors: Mianjie Zheng, Xinquan Yang, Xuefen Liu, Xuguang Li, Kun Tang, He Meng, Linlin Shen
Organizations: College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, China · AI Research Center for Medical Image Analysis and Diagnosis, Shenzhen University, Shenzhen, China · National Engineering Laboratory for Big Data System Computing Technology, Shenzhen University, China · Department of Stomatology, Shenzhen University General Hospital, Shenzhen, China · School of Artificial Intelligence, Shenzhen University, Shenzhen, China
Dental implant abutments serve as the geometric and biomechanical interface between the implant fixture and the prosthetic crown, yet their design relies heavily on manual effort and is time-consuming. Although deep neural networks have been proposed to assist dentists in designing abutments, most existing approaches remain largely manual or semi-automated, requiring substantial clinician intervention and lacking scalability in multi-abutment scenarios. To address these limitations, we propose TEMAD, a fully automated, text-conditioned multi-expert architecture for multi-abutment design. This framework integrates implant site localization and implant system, compatible abutment parameter regression into a unified pipeline. Specifically, we introduce an Implant Site Identification Network (ISIN) to automatically localize implant sites and provide this information to the subsequent multi-abutment regression network. We further design a Tooth-Conditioned Feature-wise Linear Modulation (TC-FiLM) module, which adaptively calibrates mesh representations using tooth embeddings to enable position-specific feature modulation. Additionally, a System-Prompted Mixture-of-Experts (SPMoE) mechanism leverages implant system prompts to guide expert selection, ensuring system-aware regression. Extensive experiments on a large-scale abutment design dataset show that TEMAD achieves state-of-the-art performance compared to existing methods, particularly in multi-abutment settings, validating its effectiveness for fully automated dental implant planning.
Figures & tables
Figure 1: Illustration of the traditional CAD–CAM workflow for multi-abutment design, where each implant site must be positioned and measured individually before designing the corresponding abutment based on the measured parameters.
Figure 2: Illustration of the self-supervised masked autoencoder pretraining framework.
Figure 3: The overall framework of the proposed method.
Figure 4: Visualization of the remeshing process. The original intraoral scan mesh is first converted into a base mesh through manifold repair and mesh simplification. The base mesh is then refined through three successive subdivision steps, where each face is split into four ordered faces at each step. Finally, the refined mesh is partitioned into Fn patches, where each patch contains 64 faces generated after three subdivisions, serving as the input to the network.
Figure 5: Comparison of parameter-specific performance for multi-abutment prediction. Each subplot corresponds to one key parameter. The horizontal axis denotes IoU performance, while the vertical axis lists the values of the corresponding parameter.
Figure 6: Visualization of pre-trained reconstruction performance.
Network
Params(M)
FLOPs(G)
AISL (%)
Acc@1(%)
Acc@2(%)
Acc@3(%)
AISC (%)
PointFormer
12.91
43.01
19.86
32.98
33.64
33.64
26.57
PointNet
1.61
8.5
33.20
51.27
52.60
52.67
48.89
PointNet++
1.74
8.5
82.46
92.76
92.90
92.93
86.43
Table 1: The performance of IAPM. AISL : Implant Site Localization Accuracy, AISC : Implant Site Count Accuracy, Acc@k denotes the proportion of ground-truth implant sites whose predicted locations fall within k adjacent tooth positions (intra-arch distance), Params: Parameters, FLOPs: Floating Point Operations Per Second.
ISIN
SPMoE
Transgingival
Diameter
Height
✗
✔
29.42
72.00
40.01
✔
✗
26.24
55.63
19.91
✔
✔
29.88
70.78
40.45
Table 2: Ablation experiments on the ISIN and SPMoE module.
Dataset
Modality
Method
Transgingival
Diameter
Height
SA
PointNet
24.62
62.17
22.83
PointNet++
26.30
57.38
15.39
Point Cloud
PointMAE
25.10
55.83
43.16
PointMamba
30.14
62.17
46.49
PointFEMAE
26.60
57.70
19.16
Mesh
MeshMAE
28.61
62.53
18.44
Table 3: Comparison of TEMAD with other mainstream. SA denotes the dataset containing only single-abutment cases, while SMA includes both single- and multi-abutment samples.
Figure 7: Ablation study on encoder pretraining. The horizontal axis denotes IoU performance, while the vertical axis corresponds to the three key abutment parameters, including transgingival, diameter, and height. w/o PT and w/ PT indicate model variants trained without and with encoder pretraining, respectively.
In the design of surgical guides for implant placement, determining the precise implant position is a critical step. However, the implant region itself is often characterized by a lack of distinctive texture in medical images. Consequently, artificial intelligence (AI) models must infer the correct implant position and angulation (slope) primarily by analyzing the texture of the surrounding teeth, which poses a significant challenge. To address this, we propose ImplantMamba, a network architecture designed for long-range sequential modeling to integrate texture information from adjacent teeth. Our approach explicitly couples the regression of the implant position with its slope. The core of ImplantMamba is a hybrid encoder that combines Convolutional Neural Networks (CNNs) with Mamba layers. This design enables the network to hierarchically extract local anatomical features through CNNs while simultaneously modeling global contextual dependencies across the entire scan volume via Mamba's selective scan operations, leading to a more comprehensive understanding of the implant site. Furthermore, we introduce a Slope-Coupled Prediction Branch (SCP). This branch is designed to connect the prediction of implant position with the slope, ensuring internal consistency and anatomical plausibility by thereby enforcing a coherent relationship between the predicted implant location and its angulation. Extensive experiments on a large-scale dental implant dataset demonstrate that the proposed ImplantMamba achieves superior performance compared to existing methods.
Xinquan Yang, Congmin Wang, Xuguang Li +3
School of Artifical Intelligence, Shenzhen University, Shenzhen, China · National Engineering Laboratory for Big Data System Computing Technology, Shenzhen University, China · Huangpu People’s Hospital, Zhongshan City, China +1
Dental crown restoration is one of the most common treatment modalities for tooth defect, where personalized dental crown design is critical. While computer-aided design (CAD) systems have notably enhanced the efficiency of dental crown design, extensive manual adjustments are still required in the clinic workflow. Recent studies have explored the application of learning-based methods for the automated generation of restorative dental crowns. Nevertheless, these approaches were challenged by inadequate spatial resolution, noisy outputs, and overextension of surface reconstruction. To address these limitations, we propose \totalframework, a margin-aware mesh generation framework comprising CrownDeformR and CrownSegger. Inspired by the clinic manual workflow of dental crown design, we designed CrownDeformR to deform an initial template to the target crown based on anatomical context, which is extracted by a multi-scale intraoral scan encoder. Additionally, we introduced \marginseg, a novel margin segmentation network, to extract the cervical margin of the target tooth. The performance of CrownDeformR improved with the cervical margin as an extra constraint. And it was also utilized as the boundary condition for the tailored postprocessing method, which removed the overextended area of the reconstructed surface. We constructed a large-scale intraoral scan dataset and performed extensive experiments. The proposed method significantly outperformed existing approaches in both geometric accuracy and clinical feasibility.
Metal artifacts generated by dental implants significantly degrade cone-beam computed tomography (CBCT) volumes, obscuring critical anatomical structures and compromising diagnostic precision. To address this, an unsupervised deep learning framework has been proposed for Metal Artifact Reduction (MAR) utilizing a Cycle-Consistent Adversarial Network (CycleGAN) optimized for high-fidelity restoration. Unlike supervised methods that rely on unattainable voxel-aligned paired datasets, the proposed approach leverages an unpaired dataset of approximately 4,000 images, curated from the public ToothFairy dataset. The architecture integrates U-Net-based generators and PatchGAN discriminators, specifically tuned to mitigate generative hallucinations and preserve morphological integrity. Quantitative benchmarking on a held-out test set demonstrates a 34.6% improvement in the Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE) score, a substantial reduction in Fréchet Inception Distance (FID) from 207.03 to 157.04, and a superior Structural Similarity Index Measure (SSIM) of 0.9105. The framework achieves real-time efficiency with a 3.03 ms inference time per slice, effectively suppressing artifacts while preserving anatomical detail. Expert validation confirms high fidelity; however, to ensure reliability in extreme cases, the architecture is recommended as a clinical decision-support tool under human-in-the-loop oversight. By enhancing diagnostic clarity via a scalable software pipeline, this study provides a robust solution for high-fidelity dental implant imaging.
G. L. T. Chamika, S. N. A. Dhanapala, P. H. S. V. Nimalaweera +2
Department of Electrical and Electronic Engineering, University of Peradeniya, Peradeniya, Sri Lanka · Department of Oral Medicine and Periodontology, Faculty of Dental Sciences, University of Peradeniya, Peradeniya, Sri Lanka