AI helps stroke patients adhere to medical treatment
April 11, 2017
Results from a randomised controlled trial published this month in the American Heart Association's journal Stroke demonstrate that stroke survivors were twice as likely to take anti-blood clot treatment (anticoagulants) when using an artificial intelligence (AI) platform than patients receiving treatment as usual.
The AI platform from AiCure in New York uses software algorithms on smartphones to confirm automatically patient identity, the medication and medication ingestion. In addition, patients receive automated reminders and precise dosing instructions. Real-time data were available for review by health care workers. Early detection of non-adherence allowed for immediate follow up and ensured patients stayed on track in the intervention group.
The artificial intelligence visually confirms medication ingestion on any smartphone letting the clinically validated platform continuously monitor patients.
"In the absence of routine laboratory monitoring, artificial intelligence has the potential to automate a critical component of care – adherence monitoring – and provide continuity of care between visits to ensure patients persist with their therapy and get full therapeutic benefit," said Daniel Labovitz, lead author and director of the Stern Stroke Center at Montefiore Medical Center.
The 12-week study included ischemic stroke (clot caused) survivors randomly assigned to an AI platform for daily monitoring or treatment as usual groups. To measure treatment adherence, researchers tested the concentration of medication in blood samples. Blood tests showed that all patients in the intervention arm took medication regularly, compared to only half in the control arm. Participants were average age 57 years, 54 per cent female, 46 per cent black, 43 per cent Hispanic, and 11 per cent white.
In the USA, stroke is the fifth-leading cause of death and a major cause of long-term disability among survivors. Since the danger of having another stroke increases after a first stroke, preventive treatment is critical. In the past, healthcare providers needed to monitor anti-blood clot treatment for bleeding complications, but recently approved medications do not require monitoring. However, the reduced need for monitoring has placed pressure on patients to self-manage and leaves room for suboptimal adherence to go undetected; routine laboratory tests are not reliable indicators of adherence.
"Many patients are unable to self-manage and are at increased risk of stroke and bleeding," said Laura Shafner, study co-author and chief strategy officer at AiCure. "The use of technology and artificial intelligence has the potential to significantly improve health outcomes and reduce costs in clinical care."
The study demonstrated that patients with little prior experience using a smartphone were able to learn easily how to use the technology and use it consistently. The AI platform is being deployed in drug development and population health settings. The AI platform is said to be the only automated technology to be validated in ambulatory settings against drug concentration levels.
Co-authors are Morayma Reyes Gil, Deepti Virmani and Adam Hanina. The research was supported by National Center for Advancing Translational Sciences of the National Institutes of Health under grant number R44 TR00087302.