Deep learning-based classification of thyroid nodules in ultrasound images: A case-level evaluation of CNN and vision transformer architectures

Thyroid nodule classification by ultrasound presents challenges in that nodule categories with varying risk degrees can have overlapping ultrasound characteristics. In this paper, we assess deep learning-based TI-RADS classification of thyroid nodules based on a case-based evaluation strategy that minimizes information leakage. The Digital Database of Thyroid Ultrasound Images (DDTI) was re-evaluated and resulted in 349 ultrasound images belonging to 298 cases. TI-RADS 2–3 nodule categories are categorized as low suspicion cases, whereas TI-RADS 4a–5 cases are categorized as high suspicion. Three deep learning architectures, DenseNet121, EfficientNetB0, and ViT-Tiny, were evaluated using five-fold case-level cross-validation. We calculated class weights only based on training data. The evaluation metrics used were accuracy, sensitivity, specificity, precision, F1 score, balanced accuracy, ROC-AUC, and PR-AUC with bootstrapped confidence intervals and statistical tests. DenseNet121 had the highest accuracy (0.718) and F1 score (0.813), but EfficientNetB0 had the highest ROC-AUC (0.724) and specificity (0.673). ViT-Tiny had an accuracy of 0.634 and an ROC-AUC of 0.648. DenseNet121 achieved significantly higher accuracy and F1 score than EfficientNetB0, whereas differences in ROC-AUC, PR-AUC, and balanced accuracy were not statistically significant. The TN5000 evaluation demonstrated poor transfer performance of 0.498 balanced accuracy and 0.533 ROC-AUC. The deep learning model showed moderate performance for TI-RADS-based risk stratification. Peritumoral information may help to add additional value, but cross-dataset transfer is still difficult.

Authors

Institutions

Publication Details

Journal
Research on Biomedical Engineering
Published
2026-11-16
DOI
https://doi.org/10.1007/s42600-026-00506-x
Primary Topic
AI in cancer detection
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Deep learning-based classification of thyroid nodules in ultrasound images: A case-level evaluation of CNN and vision transformer architectures

Nabila Zrira, Rajaa Sebihi, Sanae Azzouz, Rajaa Bouzidi Idrissi
2 links
Research on Biomedical Engineering
AI in cancer detection
article

Deep learning-based classification of thyroid nodules in ultrasound images: A case-level evaluation of CNN and vision transformer architectures

Nabila Zrira, Rajaa Sebihi, Sanae Azzouz, Rajaa Bouzidi Idrissi
article en

Abstract

Thyroid nodule classification by ultrasound presents challenges in that nodule categories with varying risk degrees can have overlapping ultrasound characteristics. In this paper, we assess deep learning-based TI-RADS classification of thyroid nodules based on a case-based evaluation strategy that minimizes information leakage. The Digital Database of Thyroid Ultrasound Images (DDTI) was re-evaluated and resulted in 349 ultrasound images belonging to 298 cases. TI-RADS 2–3 nodule categories are categorized as low suspicion cases, whereas TI-RADS 4a–5 cases are categorized as high suspicion. Three deep learning architectures, DenseNet121, EfficientNetB0, and ViT-Tiny, were evaluated using five-fold case-level cross-validation. We calculated class weights only based on training data. The evaluation metrics used were accuracy, sensitivity, specificity, precision, F1 score, balanced accuracy, ROC-AUC, and PR-AUC with bootstrapped confidence intervals and statistical tests. DenseNet121 had the highest accuracy (0.718) and F1 score (0.813), but EfficientNetB0 had the highest ROC-AUC (0.724) and specificity (0.673). ViT-Tiny had an accuracy of 0.634 and an ROC-AUC of 0.648. DenseNet121 achieved significantly higher accuracy and F1 score than EfficientNetB0, whereas differences in ROC-AUC, PR-AUC, and balanced accuracy were not statistically significant. The TN5000 evaluation demonstrated poor transfer performance of 0.498 balanced accuracy and 0.533 ROC-AUC. The deep learning model showed moderate performance for TI-RADS-based risk stratification. Peritumoral information may help to add additional value, but cross-dataset transfer is still difficult.

Research on Biomedical EngineeringVol. 42(4)
Mohammed V University (MA), National School of Mineral Industry (MA)
Good health and well-being
AI in cancer detection
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon