Improving diagnostic precision in thyroid nodule segmentation from ultrasound images with a self-attention mechanism-based Swin U-Net model
Bibliographic record
Abstract
Background: Accurate segmentation of thyroid nodules in ultrasound imaging remains a significant challenge in medical diagnostics, primarily due to edge blurring and substantial variability in nodule size. These challenges directly affect the precision of thyroid disorder diagnoses, which are crucial for metabolic and hormonal regulation. Methods: This study proposes a novel segmentation approach utilizing a Swin U-Net architecture enhanced with a self-attention mechanism. The model integrates residual and multiscale convolutional structures in the encoder path, with long skip connections feeding into an attention module to improve edge preservation and feature extraction. The decoder path employs these refined features to achieve precise segmentation. Comparative evaluations were conducted against traditional models, including U-Net and DeepLabv3+. Results: The Swin U-Net model demonstrated superior performance, achieving an average Dice Similarity Coefficient (DSC) of 0.78, surpassing baseline models such as U-Net and DeepLabv3+. The incorporation of residual and multiscale convolutional structures, along with the use of long skip connections, effectively addressed issues of edge blurring and nodule size variability. These advancements resulted in significant improvements in segmentation accuracy, highlighting the model's potential for addressing the inherent challenges of thyroid ultrasound imaging. Conclusion: The enhanced Swin U-Net architecture exhibits notable improvements in the robustness and accuracy of thyroid nodule segmentation, offering considerable potential for clinical applications in thyroid disorder diagnosis. While the study acknowledges dataset size limitations, the findings demonstrate the effectiveness of the proposed approach. This method represents a significant step toward more reliable and precise diagnostics in thyroid disease management, with potential implications for enhanced patient outcomes in clinical practice.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".