PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 16, 2026Diagnostics2 citationsOpen Access

Adaptive Bandelet Transform and Transfer Learning for Geometry-Aware Thyroid Cancer Ultrasound Classification

View Full Paper
YHYassine HabchiHKHamza KheddarMGMohamed Chahine Ghanem

Key Points

  • To improve the classification of thyroid nodules in ultrasound images using geometry-adaptive strategies and transfer learning techniques.
  • Applied Bandelet Transform for enhanced feature representation in ultrasound images
  • Used SMOTE to address class imbalance
  • Incorporated targeted data augmentation to diversify datasets
  • Classified using multiple pre-trained architectures, focusing on VGG19 for optimal performance
  • Achieved 98.91% accuracy with the BT+TL (VGG19) model
  • Demonstrated improvement over classical wavelet representations
  • Obtained 98.11% sensitivity and 97.31% specificity
  • Produced a 98.89% F1-score, surpassing existing methods

Abstract

Background and Objectives: Classification of thyroid nodules (TN) in ultrasound remains challenging due to limited labelled data and the limited capacity of conventional feature representations to capture complex, multi-directional textures. This work aims to improve data-efficient TN classification by integrating a geometry-adaptive Bandelet Transform (BT) with transfer learning (TL) to enhance feature representation and generalisation. Methods: The proposed pipeline first applies BT to strengthen directional and structural encoding in ultrasound images via quadtree-driven geometric adaptation. It then mitigates class imbalance using SMOTE and increases data diversity through targeted data augmentation. The resulting representations are classified using multiple ImageNet-pretrained architectures, where VGG19 yields the most consistent performance. Results: Experiments on the publicly available DDTI dataset show that BT-based preprocessing consistently improves performance over classical wavelet representations across multiple quadtree thresholds, with the best results obtained at T=30. Under this setting, the proposed BT+TL (VGG19) model achieves 98.91% accuracy, 98.11% sensitivity, 97.31% specificity, and a 98.89% F1-score, outperforming comparable approaches reported in the literature. Conclusions: Coupling geometry-adaptive transforms with modern TL backbones provides a robust and data-efficient strategy for ultrasound TN classification, particularly under limited annotation and challenging texture variability. The complete project is publicly available.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Habchi et al. (2026) studied this question.

synapsesocial.com/papers/6992b45f9b75e639e9b094b3https://doi.org/10.3390/diagnostics16040554
Ask AI
Helpful
Bookmark
Share
View Full Paper