Predictive model uses 6 ultrasound features to differentiate between benign and malignant thyroid nodules
Experts have developed a predictive nomogram model capable of accurately differentiating between benign and malignant thyroid nodules with peripheral calcifications.
These calcifications have been a source of debate for some time. Years ago, peripheral calcifications were considered a characteristic of benign nodules. However, more recently instances of peripheral ring or arc calcifications have been more frequently linked to cases of papillary, undifferentiated or follicular carcinomas.
“Although the correlation between thyroid malignancy and peripheral calcification has been well established, there is no unified statement about the associations of related features of peripheral calcification with malignancy,” Fajin Dong, MD, from the department of ultrasound at Shenzhen People’s Hospital in China, and colleagues noted in a new paper describing their model in Academic Radiology.
As such, there is a need for more accurate methods of distinguishing benign from malignant nodules containing these characteristics. To do this, experts sought to develop a nomogram model that combines both clinical and imaging features to predict malignancy.
The model was trained on images of 268 nodules with peripheral calcifications and confirmed pathological results, while another 112 nodules were included in a validation set. Candidate variables included age, gender and ultrasound imaging features.
Through this, researchers were able to highlight several imaging features indicative of cancer; halo sign, extrusion beyond calcification, type of peripheral calcification, margin, internal echogenicity and composition all showed strong associations with malignancy. These six features yielded excellent calibration and discrimination in both the training and validation cohorts, and achieved AUCs of 0.904 and 0.882, respectively.
“In our study, similar to other thyroid nodules, the internal echogenicity, internal composition, and margin contributed to the diagnosis of malignant thyroid nodules with peripheral calcifications,” the group noted. “Consistent with previous studies, solid structures were significantly associated with a higher risk of malignancy in peripheral calcified nodules.”
The authors added that calcification type had the highest predictive value, with type 2 and 3 warranting “heightened vigilance.”
Although their study had limitations, such as its retrospective nature, the team indicated that once the model is further validated, it could help guide providers in determining how to manage thyroid nodules.
Learn more about the findings here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.