Industry leading news about artificial intelligence in healthcare and diagnostics.

Using Smartphones to Perform Screening Tests is Among New AI Health Initiatives from Google

Google recently announced a series of new initiatives that apply artificial intelligence (AI) to healthcare. Notably, two of these endeavors will start using smartphones to perform screening tests traditionally performed by healthcare providers.

In a blog post on March 24, Greg Corrado, PhD, Google’s Head of Health AI, outlined three new AI health efforts from the company.

“Google Health has expanded its research and applications to focus on improving the care clinicians provide and allow care to happen outside hospitals and doctor’s offices,” Corrado wrote.

The first key initiative is the use of AI to detect diabetic changes to the retinas using smartphone cameras. Named the Automated Retinal Disease Assessment, or ARDA, this technology is already in use and has performed almost 100,000 retinal screenings.

While ARDA is currently using smartphones to perform screening tests of retinas for people who may have diabetes, the technology is being expanded for patients with other diseases.

“While this is in the early stages of research and development, our engineers and scientists envision a future where people, with the help of their doctors, can better understand and make decisions about health conditions from their own homes,” Corrado wrote.

Using Smartphones for Screening Tests of Heartbeats

Another Google Health push is the use of smartphone microphones to listen to and interpret heart sounds. This would provide patients with a way to have their heart function tested using solely their phone.

“Our latest research investigates whether a smartphone can detect heartbeats and murmurs,” Corrado explained. “We're currently in the early stages of clinical study testing, but we hope that our work can empower people to use the smartphone as an additional tool for accessible health evaluation.”

The third area where Google has recently announced a new focus on AI-development is in interpreting ultrasounds. Corrado notes that over half of pregnant mothers in developing countries do not receive ultrasounds during their pregnancies, “in part due to a shortage of expertise in reading ultrasounds.”

“We believe that Googles expertise in machine learning can help solve this and allow for healthier pregnancies and better outcomes for parents and babies,” Corrado continued, adding that Google has partnered with Northwestern Medicine in Chicago to develop and validate this new technology.

Google Explores Analyzing Unstructured Patient Data

Google also is involved in other AI-related undertakings that may improve how electronic health records are used.

A March 24 Wall Street Journal article explored a recently-announced collaboration between Mayo Clinic and Google to use AI to analyze language-based data from patient records.

A meaningful portion of patient data is unstructured, making it hard to analyze, including doctor’s notes, progress notes, or information created by a provider by entering data. Natural language processing tools, however, are able to analyze unstructured data in large quantities.

“We just have a lot of data that is not accessible and locked down because theyre in an unstructured format,” said Vish Anantraman, MD, Chief Technology Officer at Mayo Clinic in Rochester, Minn., told the Journal.

“I think the results are very promising,” Anantraman continued, adding that patient data examined in the collaboration with Google is now much more accessible, but that additional work is still needed. “I would refrain from saying its spectacular,” he concluded.

Google is not the only tech giant interested in unstructured patient data. In mid-July 2021, Amazon Web Services (AWS) introduced cloud-based software designed to not only store healthcare data, but also help interpret the information.

As Google makes new moves in the healthcare AI space, healthcare leaders will want to be aware of how these developments can impact future trends. New technologies that take advantage of using smartphones for screening tests may shift provider-driven care to something that is available in a patient’s pocket.

Other trends, however, like using natural language processing to analyze unstructured data may improve the insights that hospitals can ultimately obtain, increasing the value they provide to patients. Forward looking leaders will want to carefully weigh the implications of emerging AI technologies and the impact they may have.

—Caleb Williams

Related Resources:

The Check Up: Our latest health AI developments

Automated Retinal Disease Assessment

Teaching AI to read fetal ultrasound in low- and middle-income countries to identify high-risk patients

Mayo Clinic Assessing Google AI Tool as It Seeks More Insights From Patient Records

Amazon Bets Big on Precision Medicine with New Service to Help Hospitals Manage Precision Medicine Data