---
title: Teeth Segmentation Accuracy
url: https://www.emergentmind.com/topics/teeth-segmentation-accuracy-tsa
type: topic
---

# Teeth Segmentation Accuracy

Teeth Segmentation Accuracy (TSA) is a central metric family for evaluating the fidelity of automated dental image segmentation models across 2D radiographs, 3D intraoral scans, and volumetric imaging such as cone-beam CT (CBCT). TSA quantifies how accurately a system delineates individual teeth (or tooth regions) relative to expert-annotated ground truth, typically capturing both pixelwise or pointwise correctness, spatial overlap, and boundary integrity. Robust TSA metrics underpin the validation and comparison of deep learning models in digital dentistry, enabling precise diagnosis, treatment planning, and a variety of clinical workflows.

## 1. Core Definitions and Variants of TSA

The most widely adopted TSA metrics are derived from classical segmentation performance indices but are tailored to the dental domain’s instance- and boundary-specific challenges.

- **Intersection-over-Union (IoU):** 
  For binary or multi-class mask $P$ (prediction) and $G$ (ground truth), 
  $$
  \mathrm{IoU} = \frac{|P \cap G|}{|P \cup G|}
  $$
  This is frequently averaged over all tooth instances, yielding a mean IoU (mIoU) as a global measure [2310.17176, 2511.18533, 2503.23702].

- **Dice Similarity Coefficient (DSC):**
  $$
  \mathrm{DSC} = \frac{2|P \cap G|}{|P| + |G|}
  $$
  Closely related to IoU via
  $$
  \mathrm{DSC} = \frac{2\,\mathrm{IoU}}{1+\mathrm{IoU}}
  $$

- **Accuracy (ACC/OA):**
  Pointwise or pixelwise accuracy,
  $$
  \mathrm{ACC} = \frac{TP + TN}{TP + TN + FP + FN}
  $$
  where $TP$, $TN$, $FP$, $FN$ are true/false positives/negatives [2204.02939, 2511.18533, 2204.08797].

- **Boundary Metrics:**
  - **Rotated IoU (RIoU):** Quantifies the overlap between oriented bounding boxes (OBBs) of predicted and ground truth masks [2310.17176].
  - **Boundary IoU:** Restricts overlap computation to crown–gingiva or tooth-tooth junctions [2503.23702]. 
  - **Hausdorff Distance (HD):** Measures maximal deviation between predicted and GT boundaries [2409.03982].

- **Composite/Challenge Scores:**
  Leaders in public challenges employ composite TSA scores aggregating multiple terms, such as:
  $$
  \mathrm{TSA\,Score} = w_1\,\mathrm{Dice} + w_2\,\mathrm{IoU} + w_3\,(1 - H(d))
  $$
  with empirically derived weights [2409.03982].

- **Instance-Level F1/TSA (3D challenges):**
  $$
  \mathrm{TSA} = \frac{1}{|\mathcal{T}|} \sum_{t\in\mathcal{T}} F1_t
  $$
  where $F1_t$ is the harmonic mean of precision and recall for tooth $t$ [2305.18277].

## 2. Quantification and Evaluation Protocols

TSA computation is protocol- and modality-dependent:

- **2D Panoramic X-rays:**
  - Per-pixel comparison of predicted and GT binary or multiclass masks.
  - Instance metrics (IoU, Dice) are averaged over the set of teeth in each image and then across the dataset [2310.17176, 2311.03749].
  - Some models include orientation-aware metrics (RIoU) using OBBs derived via PCA over the mask contour [2310.17176].

- **3D Scans/Meshes:**
  - Per-point or per-face label comparison; TSA given as overall accuracy or mIoU [2210.16627, 2503.23702].
  - For challenge settings, TSA is averaged per tooth instance (instance-level F1), as in 3DTeethSeg’22 [2305.18277].
  - Additional evaluation of boundary quality is realized through boundary IoU or HD [2503.23702, 2409.03982].

- **Volumetric Imaging (CBCT):**
  - Voxelwise segmentation accuracy using Dice, IoU, and surface-based distances (HD, ASSD) [2102.06060, 2002.02143].

- **Domain-Aware/Hybrid Metrics:**
  Frameworks such as ViSTooth compute additional metrics (shape via Hu moments, position, angle) and aggregate them for composite TSA assessment [2405.08573].

### Common Experimental Setups

- **Cross-validation and held-out test sets** are standard for reporting TSA [2310.17176, 2204.02939].
- **Challenge protocols** require locked test sets and no access to ground truth, demanding strict generalizability [2305.18277, 2409.03982].
- **Ablation studies** systematically measure TSA gains from architectural innovations, loss terms, and boundary-focused strategies.

## 3. Model Architectures and Loss Functions Optimizing TSA

Architectures achieving state-of-the-art TSA integrate domain-specific advances:

- **Encoder–Decoder and Attention Mechanisms:** 
  EfficientNet-B7 encoders, grid-based attention gates, and parallel squeeze-excitation modules in FUSegNet [2310.17176]; recurrent convolutional modules and residual bridges in S-R2F2U-Net [2204.02939]; M-Net U-shape with Swin transformers and tooth-dedicated attention blocks (TAB) [2311.03749].

- **Boundary-aware Modules:** 
  Reverse-attention-based boundary extraction (BFEM), feature cross-fusion (FCFM) [2409.03982], instance-boundary loss [2512.24201], and contrastive learning at the tooth–gingiva interface [2503.23702].

- **Proposal-free Instance Segmentation:** 
  Transformer-style mask embedding heads enable robust handling of missing or malposed teeth [2512.24201].

- **Geometry-guided Losses:** 
  Curvature-aware focal losses upweight high-curvature (boundary) points, leading to improved TSA and smoother boundaries [2311.13234, 2210.16627, 2512.24201].

- **Hybrid and Regularization Losses:** 
  Hybrid Dice + Focal/cross-entropy [2310.17176, 2204.02939], squared Dice for class imbalance [2311.03749], L2 regularization [2406.03747], and composite losses balancing global and boundary-focused terms [2409.03982].

## 4. Quantitative Benchmarks

Multiple models and challenges provide direct comparative values for TSA across varying datasets and modalities:

| Model / Setting                          | TSA Metric        | Value(s)             | Dataset / Task         |
|:----------------------------------------- |:------------------|:---------------------|:-----------------------|
| DE-KAN [2511.18533]                      | Accuracy / Dice   | 98.91% / 97.1%       | CDPR 2D radiographs    |
| iMeshSegNet [2109.11941]                 | Dice              | 0.964 ± 0.054        | 3D intraoral mesh      |
| FUSegNet+AG+P-scSE [2310.17176]          | IoU / Dice / RIoU | 82.43% / 90.37% / 82.82% | Panoramic X-ray |
| S-R2F2U-Net [2204.02939]                 | Accuracy / Dice   | 97.31% / 93.26%      | Dental X-ray           |
| CGIP@3DTeethSeg'22 [2305.18277]          | Instance-level F1 | 0.9859               | 3D challenge           |
| BATISNet [2512.24201]                    | mIoU / mAP        | 84.42% / 81.93%      | Point cloud instance   |
| BFFNet [2409.03982] (STS Challenge)      | "TSA score"       | 0.91                  | 2D challenge           |
| TSegFormer [2311.13234]                  | TSA (Acc) / mIoU  | 97.97% / 94.34%      | 3D IOS, 16,000 scans   |
| CrossTooth [2503.23702]                  | mIoU / Bound. IoU | 95.86% / 82.06%      | 3D mesh, boundary      |
| TSGCN [2204.08797]                       | OA / mIoU         | 96.96% / 91.69%      | Mesh, 3D scanner       |

Best-in-class 2D deep learning models (DE-KAN, FUSegNet variants, OralBBNet) approach ≥98% accuracy and ≥97% Dice; 3D transformer-based methods (TSegFormer, TFormer) achieve ≥97.8% overall pointwise accuracy with mIoU in the mid-90% range. On more challenging boundary and instance-level metrics, recent instance-segmentation aware designs (BATISNet, TSegFormer, CrossTooth) show gains of 2–7 percentage points over prior semantic baselines with improved separation in adverse scenarios (e.g., malposed or missing teeth).

## 5. Critical Factors Impacting High TSA

The primary determinants of elevated TSA include:

- **Boundary modeling:** Explicit attention to boundaries, via reverse-attention, boundary-aware losses, or curvature-guided sampling, significantly improves TSA in regions prone to tooth adhesion or annotation ambiguity [2409.03982, 2311.13234, 2503.23702, 2512.24201].

- **Instance-awareness:** Proposal-free instance heads and transformer-based querying overcome the limitations of semantic-only approaches, especially for cases of missing, fused, or supernumerary teeth [2512.24201].

- **Multi-scale and multi-modal fusion:** Integration of local and global features through cross-fusion modules, multi-path encoders, or dual-stream graph networks enhances anatomical coherence [2311.03749, 2511.18533, 2204.08797].

- **Annotation quality and data augmentation:** Comprehensive annotations and augmentation strategies expand generalization over variable anatomy, ages, and imaging artifacts (e.g., metal in CBCT) [2311.13234, 2503.23702, 2002.02143].

- **Loss function design:** Incorporation of hybrid/weighted loss functions provides a more balanced optimization for both foreground (teeth) and challenging margins [2310.17176, 2409.03982].

## 6. Limitations and Future Directions

Despite significant advances, several challenges persist:

- **Boundary degradation under label noise** or for highly worn, supernumerary, or missing teeth remains a major source of error [2409.03982, 2503.23702].
- **Generalization to atypical anatomies** (edentulous jaws, complex prosthetics) requires further expansion of training sets and semantically-aware post-processing [2503.23702].
- **Computational cost:** High-accuracy models (DE-KAN, TSegFormer, BATISNet) may require greater memory and inference time; efficiency improvements and knowledge distillation are noted areas for future research [2511.18533].
- **Weak or unsupervised learning:** Reducing annotation burden and addressing out-of-distribution generalization can further broaden clinical applicability [2409.03982, 2311.07398].
- **Clinical metrics:** Continued refinement of composite TSA scores combining shape, position, orientation, boundary, and overlap metrics is advocated for translational robustness and explainability [2405.08573].

## 7. Research Directions and Best Practices

- **Hybrid multi-objective loss and deep supervision** across decoder stages can consistently raise TSA [2409.03982, 2311.03749].
- **Boundary-aware training and augmentation**, such as focal losses on high-curvature points, yield sharper boundaries and improve segmentation where it is most clinically relevant [2512.24201, 2503.23702, 2311.13234].
- **Model selection and validation** should leverage multiple TSA metrics (Dice, IoU, boundary scores, instance-level F1, and domain-specific attributes) to avoid overfitting to a single index [2305.18277, 2405.08573].
- **Human-in-the-loop retraining** and visual analytics frameworks (e.g., ViSTooth’s glyph + scatterplot) accelerate the discovery and remediation of rare segmentation errors, supporting continual improvement [2405.08573].

Teeth Segmentation Accuracy thus encapsulates a suite of metrics and practices. Continued progress depends on not just absolute scores, but a nuanced, region- and instance-aware assessment of segmentation reliability—especially at clinically critical boundaries. This evolving landscape mandates coordinated improvement in modeling, loss design, data curation, and metric selection, as exemplified by the leading approaches in 2D/3D imaging and international benchmark challenges.

Source: https://www.emergentmind.com/topics/teeth-segmentation-accuracy-tsa