According to Gizmodo, Google successfully trained AI to detect breast cancer “with greater accuracy than doctors.”
A recent study, funded by Google, compared analyses of around 29,000 mammograms from the UK and U.S.; the study found that Google’s technology reduced false negatives by 9.7 percent, and false positives by 5.7 percent in the U.S., and reduced 2.7 percent of false negatives and 1.2 false positives in the UK.
The AI’s strength lies in its ability to detect “hard-to-spot cancer hidden behind dense tissue,” Gizmodo says. It may also help make human jobs easier, too: “it could ‘reduce the workload’ for radiologists in the UK, where, unlike in the U.S., patients customarily get second opinions on each mammogram.”
Tech experts predict that AI will have hugely positive impacts in healthcare in the next ten to 20 years. Google will continue to contribute to the healthcare space; it has already been working on tech solutions to detect diabetic eye disease, heart disease, and to analyze data on the progression of multiple sclerosis.
Hold the Excitement
While Google is making strides with AI in healthcare, not everyone is excited. Gizmodo warns the tech industry that Google might have ulterior motives, and “will gladly scoop your data as a shady means to benevolent ends.” The publication specifically points to the UK accusing the tech giant of illegally obtaining health records for over a million people for a DeepMind study on kidney injuries; and Google allegedly gathered healthcare data without users’ consent in Project Nightingale. “Google doesn’t seem to make moral determinations with its powers,” Gizmodo says.
As a result, decision makers should be wary when moving into the new future of AI in healthcare. While AI solutions may improve disease detection and treatment options by leaps and bounds, decision makers should consider how users’/patients’ data is being handled. A wholly successful tech development should include the care and consideration of a user’s health, privacy, and data.