Predicting the Severity of Thyroid Nodules with YOLOv8 and CA+LSR Architecture

Kalpana Devi; Vidhya S; Therasa M; Praveena A; Ramesh Kumar M; Kalaivani E

doi:10.35882/jeeemi.v8i3.1137

Kalpana Devi Department of Computer Science and Engineering, Sri Ramakrishna Institute of Technology, Coimbatore, Tamil Nadu, India https://orcid.org/0000-0002-5906-8689
Vidhya S Department of Computer Science and Engineering, V.S.B Engineering College, Karur, Tamil Nadu, India. https://orcid.org/0000-0003-4575-7106
Therasa M Department of Computer Science and Engineering, Panimalar Engineering College, Chennai, Tamil Nadu, India. https://orcid.org/0000-0001-7301-978X
Praveena A Department of Computer Science and Engineering, Hindusthan Institute of Technology, Coimbatore, Tamil Nadu, India. https://orcid.org/0009-0003-6945-2187
Ramesh Kumar M Department of Information Technology, V.S.B College of Engineering Technical Campus, Coimbatore, Tamil Nadu, India. https://orcid.org/0009-0009-8313-5309
Kalaivani E Department of Computer Science and Engineering, Bannari Amman Institute of Technology, Sathyamangalam, Tamil Nadu, India https://orcid.org/0009-0003-3202-4621

DOI: https://doi.org/10.35882/jeeemi.v8i3.1137

Keywords: Thyroid cancer, YOLOv8, nodules, Coordinate Attention, Positional attributes

Abstract

The rise in thyroid cancer has significantly increased the burden on radiologists to diagnose thyroid nodules using sonography accurately. To address this challenge, a highly precise and efficient automatic computer-aided diagnosis system is needed. A retrospective analysis was conducted on a dataset consisting of 200 ultrasound images from 161 patients (84 benign and 77 malignant) at Wenzhou Central Hospital. This study presents an enhanced version of the You Only Look Once version 8 (YOLOv8) neural network, specifically designed to improve the accuracy of thyroid nodule diagnosis. YOLO has been objective in handling the required elements from the given input images or frames, and the article discusses the extensive benefits of the same. The proposed network incorporates a Coordinate Attention (CA) module and a Label Smoothing Regularization (LSR) module, which facilitate the extraction of positional information and enhance overall performance. The improved neural network demonstrates high accuracy in identifying lesion areas and classifying nodule types, achieving a mean average precision (mAP) of 90% with an average inference time of 8 milliseconds on the test dataset. The ablation experiment revealed that incorporating the CA and LSR modules adds 1.2 milliseconds of computational time per image while providing a significant 4.1% improvement in mean average precision (mAP). Compared with state-of-the-art networks, the enhanced YOLOv5 network performed exceptionally well in diagnosing benign and malignant thyroid nodules, even with a limited dataset. Furthermore, its high accuracy and efficiency suggest potential applicability to other sonographic diagnostic tasks, aiding radiologists in improving diagnostic accuracy and patient outcomes.

Downloads

Download data is not yet available.

References

I. Steinberg, D. M. Huland, O. Vermesh, H. E. Frostig, W. S. Tummers, and S. S. Gambhir, “Photoacoustic clinical imaging,” Photoacoustics, vol. 14, pp. 77–98, Jun. 2019.

C. P. Wild, E. Weiderpass, and B. W. Stewart, “World cancer report: Cancer research for cancer prevention,” in International Agency for Research on Cancer. Lyon, France : Avenue Tony Garnier, 2020.]

How Does the Thyroid Gland Work? Accessed: 2023. [Online]. Available: https://www.ncbi.nlm.nih.gov/

X. Zhou, Y. Li, and W. Liang, “CNN-RNN based intelligent recommendation for online medical pre-diagnosis support,” IEEE/ACM Trans. Comput. Biol. Bioinf., vol. 18, no. 3, pp. 912–921, May/Jun. 2021.

J. Zhang, F. Zhang, C. Zhao, Q. Xu, C. Liang, Y. Yang, H. Wang, Y. Shang, Y. Wang, X. Mu, D. Zhu, C. Zhang, J. Yang, M. Yao, and L. Zhang, “Dysbiosis of the gut microbiome is associated with thyroid cancer and thyroid nodules and correlated with clinical index of thyroid function,” Endocrine, vol. 64, no. 3, pp. 564–574, Jun. 2019, doi: 10.1007/s12020-018-1831-x.

L. Aversano, M. L. Bernardi, M. Cimitile, A. Maiellaro, and R. Pecori, “A systematic review on artificial intelligence techniques for detecting thyroid diseases,” PeerJ Comput. Sci., vol. 9, p. e1394, Jun. 2023.

M. Grussendorf, I. Ruschenburg, and G. Brabant, “Malignancy rates in thyroid nodules: A long-term cohort study of 17,592 patients,” Eur. Thyroid J., vol. 11, no. 4, Aug. 2022, Art. no. e220027.

Z.-Q. Zhao, P. Zheng, S.-T. Xu, and X. Wu, “Object detection with deep learning: A review,” IEEE Trans. Neural Netw. Learn. Syst., vol. 30, no. 11, pp. 3212–3232, Nov. 2019.

Y. Hang, “Thyroid nodule classification in ultrasound images by fusion of conventional features and res-GAN deep features,” J. Healthcare Eng., vol. 2021, pp. 1–7, Jul. 2021.

X. Zhou, W. Liang, W. Li, K. Yan, S. Shimizu, and K. Wang, “Hierarchical adversarial attacks against graph neural network based IoT network intrusion detection system,” IEEE Internet of Things J., vol. 9, no. 12, pp. 9310–9319, Jun. 2022.

F. Li, W. Sun, L. Liu, Z. Meng, and J. Su, “The application value of CDFI and SMI combined with serological markers in distinguishing benign and malignant thyroid nodules,” Clin. Transl. Oncol., vol. 22, no. 11, pp. 2200–2209, 2022, doi: 10.1007/s12094-022-02880-1.

S. Keestra, V. H. Tabor, and A. Alvergne, “Reinterpreting patterns of variation in human thyroid function: An evolutionary ecology perspective,” Evol., Med., Public Health, vol. 9, no. 1, pp. 93–112, 2021.

W. Song, “Multitask cascade convolution neural networks for automatic thyroid nodule detection and recognition,” IEEE J. Biomed. Health Informat., vol. 23, no. 3, pp. 1215–1224, May 2019.

X. Zhang, C. Xuan, J. Xue, B. Chen, and Y. Ma, “LSR-YOLO: A high-precision, lightweight model for sheep face recognition on the mobile end,” Animals, vol. 13, no. 11, p. 1824, May 2023.

A. Saini, K. Guleria, and S. Sharma, “Machine learning approaches for early identification of thyroid disease,” in Proc. World Conf. Commun. Comput. (WCONF), Jul. 2023, pp. 1–6.

O. Moussa, H. Khachnaoui, R. Guetari, and N. Khlifa, “Thyroid nodules classification and diagnosis in ultrasound images using fine-tuning deep convolutional neural network,” Int. J. Imag. Syst. Technol., vol. 30, pp. 185–195, 2020.

S. Prathibha, D. Dahiya, C. R. Rene Robin, C. Venkata Nishkala, and S. Swedha, “A novel technique for detecting various thyroid diseases using deep learning,” Intell. Autom. Soft Comput., vol. 35, no. 1, pp. 199–214, 2023.

G. Mariani, M. Tonacchera, M. Grosso, F. Orsolini, P. Vitti, and H. W. Strauss, “The role of nuclear medicine in the clinical management of benign thyroid disorders—Part 1: Hyperthyroidism,” J. Nucl. Med., vol. 62, no. 3, pp. 304–312, Mar. 2021.

R. Liu, S. Zhou, Y. Guo, Y. Wang, and C. Chang, “Nodule localization in thyroid ultrasound images with a joint-training convolutional neural network,” J. Digit. Imag., vol. 33, no. 5, pp. 1266–1279, Oct. 2020.

X. Wu, D. Sahoo, and S. C. H. Hoi, “Recent advances in deep learning for object detection,” Neurocomputing, vol. 396, pp. 39–64, Jul. 2020.

S. W. Kwon, I. J. Choi, J. Y. Kang, W. I. Jang, G.-H. Lee, and M.-C. Lee, “Ultrasonographic thyroid nodule classification using a deep convolutional neural network with surgical pathology,” J. Digit. Imag., vol. 33, no. 5, pp. 1202–1208, Oct. 2020.

M. Wang, C. Yuan, D. Wu, Y. Zeng, and W. Qiu, “Automatic segmentation and classification of thyroid nodules in ultrasound images with convolutional neural networks,” in Proc. Int. Conf. Med. Image Comput. Comput.-Assisted Interv., 2021, pp. 109–115.

J. Zhan, L.-H. Zhang, Q. Yu, C.-L. Li, Y. Chen, W.-P. Wang, and H. Ding, “Prediction of cervical lymph node metastasis with contrast-enhanced ultrasound and association between presence of BRAFV600E and extrathyroidal extension in papillary thyroid carcinoma,” Therapeutic Adv. Med. Oncol., vol. 12, Jan. 2020, Art. no. 1758835920942367, doi: 10.1177/1758835920942367.

I. K. Kang, C. K. Jung, K. Kim, J. Park, J. S. Kim, and J. Bae, “Papillary thyroid carcinoma in a separate pyramidal lobe mimicking thyroglossal duct cyst carcinoma: A case report,” J. Endocrine Surg., vol. 22, no. 4, p. 138, 2022.

F. Abdolali, J. Kapur, J. L. Jaremko, M. Noga, A. R. Hareendranathan, and K. Punithakumar, “Automated thyroid nodule detection from ultrasound imaging using deep convolutional neural networks,” Comput. Biol. Med., vol. 122, Jul. 2020, Art. no. 103871.

A. Kunapinun, M. N. Dailey, D. Songsaeng, M. Parnichkun, C. Keatmanee, and M. Ekpanyapong, “Improving GAN learning dynamics for thyroid nodule segmentation,” Ultrasound Med. Biol., vol. 49, no. 2, pp. 416–430, Feb. 2023.

C.-L. Cao, Q.-L. Li, J. Tong, L.-N. Shi, W.-X. Li, Y. Xu, J. Cheng, T.-T. Du, J. Li, and X.-W. Cui, “Artificial intelligence in thyroid ultrasound,” Frontiers Oncol., vol. 13, May 2023, Art. no. 1060702.

X. Zhou, W. Liang, K. Wang, and L. T. Yang, “Deep correlation mining based on hierarchical hybrid networks for heterogeneous Big Data recommendations,” IEEE Trans. Computat. Social Syst., vol. 8, no. 1, pp. 171–178, Feb. 2021.

W.-H. Li, W.-Y. Yu, J.-R. Du, D.-K. Teng, Y.-Q. Lin, G.-Q. Sui, and H. Wang, “Nomogram prediction for cervical lymph node metastasis in multifocal papillary thyroid microcarcinoma,” Frontiers Endocrinology, vol. 14, May 2023, Art. no. 1140360, doi: 10.3389/fendo.2023.1140360.

A. Bikas and K. D. Burman, “Epidemiology of thyroid cancer,” in The Thyroid and Its Diseases: A Comprehensive Guide for the Clinician. Cham, Switzerland : Springer, 2019, pp. 541–547.

A. Shahroudnejad, “Tun-Det: A novel network for thyroid ultrasound nodule detection,” in Proc. Int. Conf. Med. Image Comput. Comput.-Assist. Intervent., Strasbourg, France. Cham, Switzerland : Springer, Sep. 2021, pp. 656–667.

H. Du, F. Chen, H. Li, K. Wang, J. Zhang, J. Meng, H. Li, X. Xu, J. Qu, R. Wu, J. Li, M. Zhang, F. Zhang, and X. Zhu, “Deep-learning radiomics based on ultrasound can objectively evaluate thyroid nodules and assist in improving the diagnostic level of ultrasound physicians,” Quant. Imag. Med. Surgery, vol. 14, no. 8, pp. 5932–5945, Aug. 2024.

C. F. Li, R. Q. Du, Q. Y. Luo, R. Wang, and X. H. Ding, “A novel model of thyroid nodule segmentation for ultrasound images,” Ultrasound Med. Biol., vol. 49, no. 2, pp. 489–496, 2023.

T. Khan, “Application of two-class neural network-based classification model to predict the onset of thyroid disease,” in Proc. 11th Int. Conf. Cloud Comput., Data Sci. Eng., Jan. 2021, pp. 114–118.

X. Zhou, X. Yang, J. Ma, and I. K. Wang, “Energy efficient smart routing based on link correlation mining for wireless edge computing in IoT,” IEEE Internet of Things J., vol. 9, no. 16, pp. 14988–14997, Aug. 2022.

L. Chang, Y. Zhang, J. Zhu, L. Hu, X. Wang, H. Zhang, Q. Gu, X. Chen, S. Zhang, M. Gao, and X. Wei, “An integrated nomogram combining deep learning, clinical characteristics and ultrasound features for predicting central lymph node metastasis in papillary thyroid cancer: A multicenter study,” Frontiers Endocrinol., vol. 14, Feb. 2023, Art. no. 964074, doi: 10.3389/fendo.2023.964074.

M. D. McCradden, J. A. Anderson, E. A. Stephenson, E. Drysdale, L. Erdman, A. Goldenberg, and R. Zlotnik Shaul, “A research ethics framework for the clinical translation of healthcare machine learning,” Amer. J. Bioethics, vol. 22, no. 5, pp. 8–22, May 2022.

H. He, J. Zhu, Z. Ye, H. Bao, J. Shou, Y. Liu, and F. Chen, “Using multimodal ultrasound including full-time-series contrast-enhanced ultrasound cines for identifying the nature of thyroid nodules,” Frontiers Oncol., vol. 14, Aug. 2024, Art. no. 1340847.

J. Sun, B. Wu, T. Zhao, L. Gao, K. Xie, T. Lin, J. Sui, X. Li, X. Wu, and X. Ni, “Classification for thyroid nodule using ViT with contrastive learning in ultrasound images,” Comput. Biol. Med., vol. 152, Jan. 2023, Art. no. 106444.

R. Song, “Dual-branch network via pseudo-label training for thyroid nodule detection in ultrasound image,” Int. J. Speech Technol., vol. 52, no. 10, pp. 11738–11754, Aug. 2022.

A. Prochazka, S. Gulati, S. Holinka, and D. Smutek, “Patch-based classification of thyroid nodules in ultrasound images using direction independent features extracted by two-threshold binary decomposition,” Computerized Med. Imag. Graph., vol. 71, pp. 9–18, Jan. 2019.

R. Srivastava and P. Kumar, “Optimizing CNN based model for thyroid nodule classification using data augmentation, segmentation and boundary detection techniques,” Multimedia Tools Appl., vol. 82, no. 26, pp. 41037–41072, Nov. 2023.

X. Zhou, X. Xu, W. Liang, Z. Zeng, and Z. Yan, “Deep-learning-enhanced multitarget detection for end-edge-cloud surveillance in smart IoT,” IEEE Internet of Things J., vol. 8, no. 16, pp. 12588–12596, Aug. 2021.

N. G. Inan, O. Kocadağlı, D. Yıldırım, İ. Meşe, and Ö. Kovan, “Multi-class classification of thyroid nodules from automatic segmented ultrasound images: Hybrid ResNet based UNet convolutional neural network approach,” Comput. Methods Programs Biomed., vol. 243, Jan. 2024, Art. no. 107921.