It is estimated that one in eight women will be diagnosed with breast cancer during their lifetime. Generally, the first noticeable symptom of breast cancer is a lump that feels different from the rest of the breast tissue. In fact, more than 80% of breast cancer cases are discovered when the woman (or man…men get breast cancer too!) feels a lump. Finding breast cancer early can mean improved chances that the disease can be successfully treated. Fortunately, mammograms are often able to detect breast cancers early, and involve a simple low energy x-ray of the breast.
The use of mammograms as a screening tool for the detection of early breast cancer in otherwise healthy women is not without controversy. A mammogram that generates a false positive can cause significant stress and lead to unnecessary surgical interventions, and mammography can also generate false negatives. It is estimated that the numbers of cancers missed by mammography is ~20%, either due to observer error or because the cancer is hidden by other dense tissue in the breast. While mammograms do save lives, improved mammography could lead to earlier and more accurate detection of breast cancer.
Which is why we were excited to see FDA clear CureMetrix to sell its software “triage” service to screen mammograms for signs of cancer. CureMetrix is based here in San Diego, and in March 2019, the company was given the green light to market cmTriage™, which according to the company’s website is “software intended to provide a notification triage code to the radiologist’s mammography work list based on the presence of a suspicious region of interest found by the underlying algorithm. This workflow optimization tool enables a radiologist to customize their mammography work list based on cases that may need immediate attention.”
According to CureMetrix CEO Kevin Harris, “the cloud-based service sends back the results in three to four minutes. What we’re seeing in preliminary studies is the triage software can help doctors read through their work list up to 40 percent faster.” In addition to having an impact on clinical efficiency, CureMetrix says radiologists can elect to get immediate notification of suspicious results, so patients can be made aware of a concern before they have even left the clinic.
We are definitely seeing more companies like CureMetrix, focused on leveraging artificial intelligence and deep learning to develop the next generation of medical image analysis technology, and FDA is working hard to keep pace. FDA recently published a Discussion Paper and Request for Feedback on the topic of Artificial Intelligence and Machine Learning in Software as a Medical Device. FDA is considering a total product lifecycle-based regulatory framework for these technologies that would allow for modifications to be made from real-world learning and adaptation, while still ensuring that the safety and effectiveness of the software as a medical device is maintained.
Artificial intelligence and machine learning technologies have the potential to transform health care by deriving new and important insights from the vast amount of data generated during the delivery of health care every day. Information is power…we just have to figure out how to harness it to better assist health care providers and improve patient care!