The use of smartphones as diagnostic tools is still a work in progress as the overall potential of the technology is still uncertain, The Washington Post reported Jan. 15.
Smartphones' sensors are capable of monitoring a patient's vital signs, assessing people for concussions, screening for atrial fibrillation and conducting mental health wellness checks, and companies and researchers are tapping into phones to build out various apps and products that can connect people to care.
But the lack of clarity from the FDA on how to regulate artificial intelligence and machine learning-based medical devices is hindering the smartphone's ability to play a larger role in healthcare, as smartphone apps rely on algorithms built by machine learning and artificial intelligence to collect data, rather than the physical tools typically used in hospitals, making it difficult to compare their performance to existing medical devices.
Failure to build in assurances of clinical accuracy and validation can undermine the technology's goals of easing costs and access, because a physician still must verify results being received from the smartphone, according to Eugene Yang, MD, a clinical professor of medicine at the University of Washington.
Algorithms based on AI and machine learning also need to be trained, which takes time and a lot of data.
"We're not there yet," Dr. Yang told the Post.