Enhancing Thyroid Cancer Detection in Ultrasound Images Using Augmentation-Driven Deep Learning Framework
DOI:
https://doi.org/10.6000/1929-6029.2026.15.13Keywords:
Thyroid cancer, Deep learning, Image augmentation, Classification, Segmentation, Ultrasound images, TransformersAbstract
Thyroid cancer has become the most rapidly increasing endocrine malignancy, and its incidence is on the rise globally. Deep learning methods demonstrated reliable prospects in improving thyroid cancer diagnosis based on ultrasound imaging. But data scarcity and class imbalance make it challenging to build robust models. In this paper, we propose an image augmentation approach for enhancing thyroid cancer classification and segmentation using deep learning. The Thyroid Ultrasound-Image Database, which contains 2,450 images with five ACR-TIRADS levels, was applied.
Pre-processing: Noise reduction, inverse intensity, and normalization were performed. Geometric transformations, photometric augmentations, GAN-based synthetic image generation and domain-specific augmentations were utilized. For classification and segmentation tasks, CNN-based architectures (e.g., VGG, ResNet) as well as transformer-based models were employed, along with U-Net variants.
Conclusions: Data augmentation was found to lead to a substantial improvement in the models' ability to generalize, with 10–15% gains in accuracy. Various augmentation techniques were performed differently, and the combination of multiple techniques was the most accurate. Qualitative results indicated the robustness of feature extraction, and quantitative comparisons showed that our method was competitive with several state-of-the-art methods. The present approach has potential clinical applications to help radiologists in the early detection of thyroid cancer. Potential future work involves adapting advanced augmentation methods and multimodal fusion to further increase the classification performance.
References
Franchini F, Colao A, Macchia PE, Nettore IC, Palatucci G, Ungaro P. Obesity and thyroid cancer risk: An update. Int J Environ Res Public Health 2022; 19(3): Art. no. 1116. DOI: https://doi.org/10.3390/ijerph19031116
Suteau V, Briet C, Rodien P, Munier M. Sex bias in differentiated thyroid cancer. Int J Mol Sci 2021; 22(23): Art. no. 12992. DOI: https://doi.org/10.3390/ijms222312992
Hu J, Simental A, Lee SC, Yuan X, Yuan IJ, Mirshahidi S. Thyroid carcinoma: Phenotypic features, underlying biology and potential relevance for targeting therapy. Int J Mol Sci 2021; 22(4): Art. no. 1950. DOI: https://doi.org/10.3390/ijms22041950
Delon C, Vernon S, Payne NWS, Kotrotsios Y, Brown KF, Shelton J. Differences in cancer incidence by broad ethnic group in England, 2013–2017. Br J Cancer 2022; 126(12): 1765-1773. DOI: https://doi.org/10.1038/s41416-022-01718-5
Nabhan F, Dedhia PH, Ringel MD. Thyroid cancer: Recent advances in diagnosis and therapy. Int J Cancer 2021; 149(5): 984-992. DOI: https://doi.org/10.1002/ijc.33690
Jiang Z-P, Shao Z-E, Liu Y-Y, Huang K-W. An improved VGG16 model for pneumonia image classification. Appl Sci 2021; 11(23): Art. no. 11185. DOI: https://doi.org/10.3390/app112311185
Khader F, et al. Denoising diffusion probabilistic models for 3D medical image generation. Sci Rep 2023; 13(1): Art. no. 7303. DOI: https://doi.org/10.1038/s41598-023-34341-2
Müller D, Kramer F. MIScnn: A framework for medical image segmentation with convolutional neural networks and deep learning. BMC Med Imaging 2021; 21(1): DOI: https://doi.org/10.1186/s12880-020-00543-7
Anari S, Tataei Sarshar N, Rezaie A, Dorosti S, Mahjoori N. Review of deep learning approaches for thyroid cancer diagnosis. Math Problems Eng 2022; 2022: 1-8. DOI: https://doi.org/10.1155/2022/5052435
Alsaif H, et al. A novel data augmentation-based brain tumor detection using convolutional neural network. Appl Sci 12(8): Art. no. 3773. DOI: https://doi.org/10.3390/app12083773
Halicek M, Fei B, Dormer JD, Little JV. Chen AY. Tumor detection of the thyroid and salivary glands using hyperspectral imaging and deep learning. Biomed Opt Express 2020; 11(3): 1383. DOI: https://doi.org/10.1364/BOE.381257
Wang L. Deep learning techniques to diagnose lung cancer. Cancers 2022; 14(22): Art. no. 5569. DOI: https://doi.org/10.3390/cancers14225569
Anaya-Isaza A, Mera-Jimenez L. Data augmentation and transfer learning for brain tumor detection in magnetic resonance imaging. IEEE Access 2022; 10: 23217-23233. DOI: https://doi.org/10.1109/ACCESS.2022.3154061
Khan MA, Sharif M, Hsu C, Kadry S, Akram T. A two-stream deep neural network-based intelligent system for complex skin cancer types classification. Int J Intell Syst 2021; 37(12): 10621-10649. DOI: https://doi.org/10.1002/int.22691
Celard P, Iglesias EL, Sorribes-Fdez JM, Romero R, Vieira AS, Borrajo L. A survey on deep learning applied to medical images: From simple artificial neural networks to generative models. Neural Comput Appl 2022; 35(3): 2291-2323. DOI: https://doi.org/10.1007/s00521-022-07953-4
Sun L, Chen J, Gong M, Xu Y, Batmanghelich K, Yu K. Hierarchical amortized GAN for 3D high resolution medical image synthesis. IEEE J Biomed Health Inform 2022; 26(8): 3966-3975. DOI: https://doi.org/10.1109/JBHI.2022.3172976
Pandian J, Hnatiuc M, Kumar V, Geman O, Arif M, Kanchanadevi K. Plant disease detection using deep convolutional neural network. Appl Sci 2022; 12(14): Art. no. 6982. DOI: https://doi.org/10.3390/app12146982
Pekova B, et al. NTRK fusion genes in thyroid carcinomas: Clinicopathological characteristics and their impacts on prognosis. Cancers 2021; 13(8): Art. no. 1932. DOI: https://doi.org/10.3390/cancers13081932
Li F, et al. Using ultrasound features and radiomics analysis to predict lymph node metastasis in patients with thyroid cancer. BMC Surg 2020; 20(1). DOI: https://doi.org/10.1186/s12893-020-00974-7
Dar SA, et al. Improving Alzheimer’s Disease Detection with Transfer Learning. Int J Stat Med Res 2025; 14: 403-415. DOI: https://doi.org/10.6000/1929-6029.2025.14.39
Ayadi W, Dar SA. AI-Powered CNN Model for Automated Lung Cancer Diagnosis in Medical Imaging. Int J Stat Med Res 2025; 14: 616-625. DOI: https://doi.org/10.6000/1929-6029.2025.14.58
Dar SA, ELnazer AA, Rathi S, Khalid MN, Tirva D, Rather AA. Automated Detection of Posterior Tibial Slope on X-Ray Images Using VGG19. Int J Stat Med Res 2025; 14: 676-687. DOI: https://doi.org/10.6000/1929-6029.2025.14.63
Dar SA, Palanivel S, Kalaiselvi Geetha M. Autonomous Taxi Driving Environment Using Reinforcement Learning. International Journal of Modern Education and Computer Science 2022; 14: 88-102. DOI: https://doi.org/10.5815/ijmecs.2022.03.06
Dar SA, Harshitha K, Rekha P. Indoor Scene Classification Using Deep Learning Techniques. ICTACT Journal on Image and Video Processing (IJIVP) 2025; 16: 3739-3745. DOI: https://doi.org/10.21917/ijivp.2025.0529
Mithra C, Dar SA, Preethi A. Artificial Intelligence and Machine Learning Algorithms to Analyse Integrated Nutrient Management Strategies for Optimal Yield and Quality of Oilseed Crops for Sustainable Development. Proceedings of the 2024 4th International Conference on Soft Computing for Security Applications (ICSCSA), IEEE, 2024; pp. 119-126. DOI: https://doi.org/10.1109/ICSCSA64454.2024.00026
Saravanan S, Dar SA. Rather AA, Qayoom D, Ali I. Deep Learning Models for Intrusion Detection Systems in MANETs: A Comparative analysis. Decision Making Advances 2025; 3(1): 96-110. DOI: https://doi.org/10.31181/dma31202556
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Policy for Journals/Articles with Open Access
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are permitted and encouraged to post links to their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work
Policy for Journals / Manuscript with Paid Access
Authors who publish with this journal agree to the following terms:
- Publisher retain copyright .
- Authors are permitted and encouraged to post links to their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work .
