59 Quick summary: A promising non-invasive imaging technique called Second Harmonic Generation microscopy has successfully analysed structural changes in collagen fibres to distinguish cancerous thyroid nodules from benign ones. This optical approach utilizes interpretable statistical modelling instead of artificial intelligence black box methods, offering a faster and far more objective diagnostic tool to complement conventional pathology. For public policy and healthcare practice, implementing this label free assessment method could significantly reduce the number of inconclusive biopsies and unnecessary surgical procedures, directly protecting patient well-being and lowering diagnostic anxiety. A promising non-invasive imaging technique could improve how doctors diagnose papillary thyroid cancer, the most common type of thyroid cancer, according to researchers at Houston Methodist in the US. The technique, known as Second Harmonic Generation (SHG) microscopy, analyses subtle changes in collagen, the body’s primary structural protein. These changes in tissue architecture may help distinguish cancerous thyroid nodules from benign ones, potentially reducing reliance on more invasive diagnostic procedures. The study, published in the Journal of Biomedical Optics, was co-led by Stephen Wong, the John S. Dunn Presidential Distinguished Chair in Biomedical Engineering at Houston, and Raksha Raghunathan, assistant professor at the Houston. “This work is significant for the field because it could make thyroid cancer diagnosis faster, more accurate, and more objective,” Wong said. “It offers a new way to improve diagnosis by complementing conventional cytology.” According to the National Institutes of Health, thyroid cancer is the most common cancer of the endocrine system. It is also one of the most common cancers diagnosed in adolescents and young adults, with incidence rates particularly high among people aged 15 –39 years. “Unlike artificial intelligence black-box approaches, this study used interpretable statistical modelling to identify biologically meaningful collagen signatures associated with thyroid cancer,” Wong said. By analysing collagen remodelling using light-based imaging, the technique provides clearer and more consistent diagnostic information. The researchers believe it could complement conventional pathology and cytology, helping clinicians make more accurate and objective diagnostic assessments. Thyroid nodules are common, but determining which require treatment remains a clinical challenge. Current diagnostic pathways often rely on fine-needle aspiration cytology, which can produce indeterminate or inconclusive results, sometimes leading to repeat biopsies or diagnostic surgery. The researchers hope SHG microscopy could become a reliable adjunctive diagnostic tool, improving diagnostic confidence and potentially reducing the number of unnecessary surgical procedures and their associated risks, including injury to the vocal cords or parathyroid glands. The incidence of thyroid cancer has increased substantially over recent decades, largely because of improved detection of small tumours through the widespread use of ultrasound and other imaging technologies. Many of these cancers, particularly in younger patients, are slow-growing and highly treatable, contributing to excellent survival rates when diagnosed at an early stage. In the UK, thyroid cancer accounts for around 1% of all new cancer cases, with approximately 4,300 people diagnosed each year. Incidence has risen markedly since the 1990s, although mortality has remained relatively low. Women are diagnosed more frequently than men, and the disease often occurs in younger adults. Experts emphasise the importance of balancing early detection with avoiding unnecessary treatment of indolent tumours. The Houston team’s focus on papillary thyroid cancer is particularly significant because this subtype accounts for the vast majority of thyroid cancer cases. Rather than relying on molecular markers or chemical stains, SHG microscopy analyses structural changes in collagen fibres associated with malignant transformation, providing an objective, label-free optical assessment of tissue architecture. Future work will focus on validating the method in larger and more diverse patient cohorts, as well as expanding the technology to distinguish between additional thyroid cancer subtypes. The researchers believe optical techniques such as SHG microscopy could be integrated into existing pathology workflows. If confirmed in larger clinical studies, the approach could support more personalised treatment decisions and help reduce diagnostic uncertainty for patients. Additional collaborators on the study included Lin Wang, Hong Zhao, Helmi Khadra, and Elizabeth Jacobi from Houston; Wesley Poon, Orhun Davarci, and Reid Master from Texas A&M University; and Jun Liu from Shanghai Jiao Tong University. The research was funded by the National Cancer Institute, the T. T. & W. F. Chao Foundation, and the John S. Dunn Research Foundation.