A study published in Nature suggests its model was able to spot cancer in de-identified screening mammograms with fewer false positives and false negatives than experts.
WHY IT MATTERS The findings, published in Nature, indicate that Google’s AI model spotted breast cancer in de-identified screening mammograms with greater accuracy, with fewer false positives and false negatives than experts.
Google's London-based AI subsidiary DeepMind worked with Cancer Research UK Imperial Centre, Northwestern University and Royal Surrey County Hospital to train and deploy the AI model, which scanned data from more than 76,000 women in the U.K. and more than 15,000 women in the U.S.
The AI model was also able to more effectively screen for breast cancer using less information than human doctors, relying solely on X-ray images, while doctors had access to patient histories and prior mammograms.
THE LARGER TREND The use of machine learning technologies in breast cancer screening could have huge implications, as spotting and diagnosing breast cancer early remains a challenge, with radiologists overloaded and the disease affecting millions of women across the globe.
The latest research builds on Google’s work with deep learning algorithms, which it developed to help doctors spot breast cancer more quickly and accurately in pathology slides.
This isn't the first time artificial intelligence has shown big promise for better mammography detection. In 2016, researchers at Houston Methodist developed an AI software they said could improve readings to 99 percent accuracy by analyzing values from X-ray images and the text of clinical reports.
The opportunities for clinical improvements using AI are broad and diverse, and the technology stands to continue making big advances across healthcare in 2020, experts say.
For instance, AI technology based on a deep learning model has also shown promise of helping cardiologists predict irregular heart rhythm, atrial fibrillation, before it develops. That was the conclusion drawn from two studies presented at the American Heart Association Scientific Sessions 2019 and conducted by Geisinger researchers.
ON THE RECORD "Looking forward to future applications, there are some promising signs that the model could potentially increase the accuracy and efficiency of screening programs, as well as reduce wait times and stress for patients," said Shravya Shetty, technical lead at Google Health, in a blog post describing the mammography findings.
"But getting there will require continued research, prospective clinical studies and regulatory approval to understand and prove how software systems inspired by this research could improve patient care," he said.
Source at Health Care IT News