New AI Technology Can Detect Cancer On Mammograms Sooner

AI technology in mammograms spark a new hope for breast cancer detection, but still come with their own set of downsides.

A new service introducing artificial intelligence (AI) products to conduct mammograms can lead to earlier breast cancer detection. However, doctors warn of its pricing variables and potential lack of efficacy on patients of color.

AI programs catch what the traditional mammogram may miss. According to The New York Times, mammograms currently detect 87% of breast cancer cases. However, an AI model of the X-ray may find hidden forms of the illness.

Traditional mammograms can also lead to false positives and unnecessary precautionary treatments. AI models would significantly remedy these issues, but they also have downsides.

As the technology becomes more available, pricing for its inclusion varies. With no official billing code set for medical providers, the patient is responsible for covering the additional cost. While the price ranges from $40 to $100, with reimbursements by insurance companies viable, doctors do not consider AI-model mammograms a necessary standard for all.

Moreover, there are specific racial disparities regarding this new technology. Whether the AI tools can detect cancer in darker skin tones remains a concern. Doctors call for diversity in the development progress to ensure it works for patients of color.

“There’s a need for diverse training and testing of these AI tools and algorithms in order to develop them across different races and ethnicities,” Dr. Dodelzon said. AI is just a tool that learns based on what it sees.”

However, death rates for Black women with breast cancer remain the highest for the demographic. According to the American Cancer Society, they have a 40% higher chance of dying from the disease. Given the statistics, AI tech could significantly reduce this death rate through the benefit of earlier detection.

Patients at the highest risk are encouraged to pursue the AI model, even as these looming problems restrict their full potential.