Should You Get an AI Assisted Mammogram?
Here’s What Experts Say

AI assisted mammogram are now gradually being offered in the developed coutnries, where a computer program analyzes mammograms alongside a radiologist. While the technology holds promise, there are also questions about its effectiveness and cost.

The mammogram is a crucial tool in the fight against breast cancer. Early detection can significantly improve a woman’s chances of survival. However, mammograms aren’t perfect. They can miss cancers, particularly in younger women with dense breasts, and sometimes lead to unnecessary biopsies due to false positives.

Can AI Improve Breast Cancer Detection?

Mammogram images contain a wealth of information about breast tissue. Certain patterns, like bright white spots with jagged edges, might indicate cancer. However, differentiating these abnormalities from normal tissue can be challenging for humans. This is where AI steps in.

“AI models can, in some cases, ‘see what we cannot see,'” explains Dr. Katerina Dodelzon, a radiologist specializing in breast imaging at NewYork-Presbyterian/Weill Cornell Medical Center. The software highlights suspicious areas for the radiologist’s review. Some models even score images, helping prioritize which scans require the most attention.

Studies show promise. A large Swedish study found an AI model improved breast cancer detection by 20%, identifying six cancers per 1,000 women compared to radiologists finding five [1]. Another Danish study showed AI could reduce false positives, meaning fewer women needed additional testing [2].

Learn The Danger of Fast Food, Fried, and Crispy Snacks

Unanswered Questions and Room for Improvement

However, the impact of AI on breast cancer mortality remains unclear. Will finding more cancers earlier translate to more lives saved? Additionally, researchers are unsure how well these models, often trained on European data, will perform on diverse American populations.

“There’s a need for more diverse training and testing of these AI tools and algorithms,” Dr. Dodelzon emphasizes. “AI is just a tool that learns based on what it sees” [2].

Some experts caution against rushing this technology into widespread use. A similar situation occurred in the 1980s with computer-aided detection (CAD) technology, initially hailed as a breakthrough. Later studies revealed CAD did not improve mammogram accuracy [3].

“With AI analysis of mammograms, we may not find out for a couple of years if our performance went down,” warns Dr. Lewin, chief of breast imaging at Smilow Cancer Hospital and Yale Cancer Center [3].

Learn How Does Cell Phone Radiation Obliterate Brain Fertility?

AI’s Limitations: Why a Radiologist’s Expertise Matters

AI can’t replace a radiologist’s expertise. For instance, AI struggles to differentiate between surgical scars and tumors. “You just need a human for that,” says Dr. Carolyn Malone, a breast radiologist at John Theurer Cancer Center [4]. Radiologists can leverage patient medical history and their own experience to identify these anomalies.

Should You Pay Extra for an AI Mammogram?

The FDA has authorized several AI mammography products. Some clinics are offering them to patients, with out-of-pocket costs ranging from $40 to $100. Other hospitals are absorbing the cost or keeping the technology for research purposes until its value is clearer.

Wider adoption will likely lead insurance companies to reimburse the cost, but that could take time. For now, most patients likely don’t need AI for their mammograms, according to Dr. Dodelzon. However, it might offer some peace of mind for those with high anxiety about their results.

Learn Gen Z Health Crisis: Screens and the Next Patient Wave

The Takeaway: A Promising Technology Needs More Research

AI-assisted mammography holds promise for improving breast cancer detection. However, more research is needed to determine its long-term impact on patient outcomes and how well it performs on diverse populations. Currently, the cost-effectiveness of AI mammograms remains unclear.

Women should discuss the pros and cons of AI with their doctor to decide if it’s right for them. The good news is, advancements in technology are continuously evolving, offering hope for a future where AI can become a valuable tool in the fight against breast cancer.

Sources:

• [1] (2023) Performance of a Deep Learning Convolutional Neural Network for Risk Stratification in Breast Cancer Screening. Retrieved from https://pubmed.ncbi.nlm.nih.gov/37217249/
• [2] (2022) Artificial intelligence-aided interpretation