newsletter
But check where it comes from.
Improve remote patient monitoring with AI
Razmi sees the potential for digital health AI to improve RPM and enable remote diagnosis. He cites the example of how patients can now test themselves for urinary tract infections using deep learning apps that can analyze scans of strips dipped in urine samples.
Meanwhile, given the shortage of clinicians, AI could help providers diagnose conditions like atrial fibrillation or abnormal heart rhythms, he says.
“[RPM apps] Take advantage of deep learning AI and unstructured data. Because it is FDA approved, it can still be used today,” says Razmi. “They were deemed safe and accurate enough to be used in routine medical practice.”
RPM allows patients to receive better treatment earlier. Some services also offer physical therapy for musculoskeletal disorders at home. AI can monitor a patient’s movements and provide feedback, he says.
Read more: Remote patient monitoring improves nurse workflow.
In another example, Razmi points out that an AI app can detect if depression is worsening by analyzing a person’s voice. You can then provide instructions and advise the patient to follow up with their provider.
AI tools allow clinicians to set threshold parameters for remote monitoring. AI can change a clinician’s workflow by providing critical alerts similar to those of critical laboratory value, Elliott explains. Additionally, this tool allows your baseline to change over time. For example, if a patient’s blood pressure has been elevated over several weeks or months, she adds.
“It’s important to set the right warning thresholds and have a sufficiently diverse data set to know which thresholds are important and which are not (signal vs. noise),” Elliott says.
The potential of AI and virtual care
If the FDA approves medical AI tools, they could help healthcare organizations triage patients and manage chronic diseases, Razmi says.
In the future, Elliott predicts, AI will handle all administrative tasks and correct incorrect data in real time, including names, addresses, insurance information, duplicate accounts and pharmacies.
Decision support applications incorporating AI will become part of connected care. Currently, AI helps with clinical decision support, but in the future, Elliott says, these tools will become part of standard care.
“AI will be used for triage and intake – think virtual AI medical assistants,” Elliott says. “This is done in a rudimentary way today, but the future may involve many paramedical activities that are currently inefficient and not the best use of time.”