Thyroid cancer is the most-common endocrine malignancy. Its incidence has tripled in the last thirty years. Diagnosis of thyroid cancer is difficult because 50% of people older than 65 have at least 1 thyroid nodule, but only 10% of the nodules are cancerous. Approximately 1.6 billion dollars are spent annually in the US to detect thyroid cancer. Conventional ultrasound is used to identify nodules that warrant a needle biopsy. However, 65% of needle biopsies are negative for cancer and 30% are ?indeterminate.? The indeterminate nodules are surgically removed for definitive diagnosis and 75% of them prove to be benign. Therefore, well more than 80% of initially presenting nodules undergo unnecessary biopsies and more than 20% of them also undergo subsequent unnecessary surgery procedures. Accordingly, the broad objective of the proposed study is to assess the feasibility of using quantitative-ultrasound (QUS) methods to distinguish cancerous from benign nodules reliably and thereby to reduce the enormous cost and risks associated with unnecessary biopsies and surgical excisions.
The first aim of the project is to develop and asses the ability of QUS to distinguish cancerous from benign nodules and to compare the ability of QUS to the ability of conventional methods to select nodules that warrant biopsies;
the second aim i s to expand QUS methods by combining existing QUS measures developed by Riverside Research with measures derived from so-called B-flow- imaging (BFI) and shear-wave-elasticity (SWE) techniques developed by GE;
the third aim i s to formulate an objective basis for planning future, prospective studies to translate the findings of the present study to a commercial instrument that can bring QUS-based nodule evaluation into the clinic. To achieve these three aims, QUS performance in classifying cancerous and benign nodules will be compared to the performance of conventional ultrasound and the results of fine needle cytology, molecular marker analyses, and, in the cases of that undergo surgical excision, histology, will used as gold standards. Classification will be performed using standard, well understood, linear, and non-linear methods, such as linear-discriminant analysis and support- vector machines respectively. If feasibility is successfully demonstrated in the proposed project, and if the demonstration of feasibility ultimately leads to future incorporation into an instrument capable of real-time QUS analysis for reliable nodule evaluation, then a highly significant technological advance will be realized that can provide valuable, risk-reducing, cost-effective health-care benefits for patients presenting with thyroid nodules.
Thyroid cancer rates have tripled in the last thirty years, but current methods of diagnosis are very inefficient. Far too many thyroid biopsies and surgeries are performed with >80% of biopsies and >25% of thyroid surgeries being performed on benign nodules. The advanced ultrasound methods to be evaluated in this proposal seek to drastically reduce the number of unnecessary biopsies and surgeries of benign thyroid nodules.
