The Journal of Light Scattering, Volume. 37, Issue 2, 249(2025)

Optical Attenuation-Based Multi-modal OCT Imaging for Thyroid Cancer Recognition

HE Xiang1, ZHAO Chao1, DING Xiaodong2, LI Jiefu1, YU Chenglong1, and MA Jun1、*
Author Affiliations
  • 1Optics and Optoelectronics Laboratory, Ocean University of China, Qingdao, Shandong 266100, China
  • 2Thyroid Disease Diagnosis and Treatment Department, Qingdao Municipal Hospital, Qingdao, Shandong 266011, China
  • show less

    Analyzing the light attenuation and morphological features in optical coherence tomography (OCT) images can improve the diagnostic accuracy of papillary thyroid carcinoma (PTC) tissue. We propose the PDC-DRE attenuation coefficient depth-resolved model, which effectively resolves the systemic overestimation caused by discretization in conventional approaches. The optical attenuation coefficient (OAC=(3.08±0.64) mm-1) of PTC tissues is significantly lower than that of normal thyroid tissues (OAC=(4.47±0.85) mm-1). A multi-modal image dataset was established by integrating optical attenuation features with OCT morphological imaging, and a custom ThyOCTNet model was employed to differentiate between normal and PTC tissues. Results demonstrate that the ThyOCTNet model achieves enhanced sensitivity (98.5%), specificity (98.8%), and accuracy (98.6%) with multi-modal images compared to single-modal images. This study confirms that OCT multi-modal fusion imaging combining optical attenuation and morphological features significantly improves PTC identification accuracy, providing a new technical pathway for the clinical translation of non-invasive pathological diagnostic technologies.

    Tools

    Get Citation

    Copy Citation Text

    HE Xiang, ZHAO Chao, DING Xiaodong, LI Jiefu, YU Chenglong, MA Jun. Optical Attenuation-Based Multi-modal OCT Imaging for Thyroid Cancer Recognition[J]. The Journal of Light Scattering, 2025, 37(2): 249

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Mar. 14, 2025

    Accepted: Jul. 31, 2025

    Published Online: Jul. 31, 2025

    The Author Email: MA Jun (majun@ouc.edu.cn)

    DOI:10.13883/j.issn1004-5929.202502012

    Topics