A radiologist at SimonMed Imaging in Tempe, Arizona, identified cancerous tumors in Deirdre Hall’s left breast after artificial intelligence software flagged a suspicious area during a routine mammogram last summer, according to both the patient and the physician.
Hall, 55, a respiratory therapist, was diagnosed with Stage 1 breast cancer after the AI-driven alert prompted an ultrasound and subsequent biopsy. Doctors determined the cancer had been camouflaged by the density and complex layering of her breast tissue, a pattern that could have been missed without the AI’s assistance, said Dr. Sean Raj, chief medical officer and chief innovation officer at SimonMed Imaging.
The case illustrates a growing trend in breast cancer screening: radiologists are increasingly using AI to augment their interpretation of mammograms. The technology is designed to identify subtle anomalies that might be overlooked by the human eye, potentially leading to earlier and more accurate diagnoses.
AI systems are trained by inputting data from hundreds of thousands, or even millions, of mammograms. The software learns to differentiate between normal breast tissue and patterns indicative of cancer, creating a mathematical representation of both. As the AI is exposed to more images, its accuracy improves, according to experts in the field.
Research suggests AI can not only detect cancerous tissue more quickly and accurately but also predict an individual’s risk of developing breast cancer between mammograms. However, concerns remain regarding potential biases within the AI algorithms and the need for further validation through large-scale clinical trials, particularly within the United States.
The use of AI in breast cancer diagnosis is part of a broader shift toward leveraging artificial intelligence in healthcare. AI is already used in various medical applications, from guiding music playlists to responding to health questions via chatbots. In breast cancer imaging, AI is being applied across multiple modalities, including mammography, digital breast tomosynthesis, ultrasound, magnetic resonance imaging, and nuclear medicine techniques, according to a recent study published in J Breast Cancer.
Hall stated she is grateful the cancer was detected early, with no spread to the lymph nodes. She credits the AI software with saving her life.
Despite the promising results, some experts caution that more research is needed to demonstrate that AI consistently improves patient outcomes and does not lead to unnecessary follow-up tests for benign growths. The long-term impact of AI on breast cancer mortality rates remains to be seen.
