In one of a kind research, it has been established that Artificial Intelligence (AI) can be used to detect benign thyroid nodules.
These nodules are first noticed on thyroid ultrasound and are known to be non cancerous. The benefit of this technology is that it can help get rid of a number of unnecessary biopsies.
Benign thyroid nodules are widespread but they are not easy to detect. In order to detect them, fine needle aspiration biopsy is employed. And in this way, doctors diagnose thyroid cancer.
It’s also been seen that most biopsies produce benign (noncancerous) results and are also known to be avoidable. The above research was conducted by Nikita Pozdeyev, MD, PhD of the University of Colorado Anschutz Medical Campus in Aurora, CO.
In their clinical research, scientists managed to establish that AI assessment of ultrasound images can easily rule out thyroid cancer and also avoid biopsy in most of the cases.
But in the new research, they employed machine learning technique which is a kind of AI. It’s basically done to analyze ultrasound images of thyroid nodules.
AI Platform Identifies Benign Thyroid Nodules
In order to establish the experiment, researchers used over 30,000 images from as many as 600 Benign Thyroid Nodules that were initially employed to classify thyroid nodules in the category of “cancer” or “no cancer.
The above research was conducted in more than 145 different Benign Thyroid Nodules that were collected at a different healthcare center.
“This study demonstrates that the ultrasound-based AI classifier of thyroid nodules achieves sensitivity comparable to that of thyroid biopsy with fine-needle aspiration,” Pozdeyev said.
“The above technology could assist radiologists and endocrinologists in choosing which thyroid nodules should undergo biopsy, especially those in the community who may not review many thyroid ultrasound images.”