Smartphone App Helps Researchers to Detect Early Signs of Stroke

Researchers gathered at the Society of NeuroInterventional Surgery's (SNIS) 20th Annual Meeting to discuss a smartphone app designed to detect early signs of stroke accurately, leveraging the power of machine learning.

According to a press release, the study, titled "Smartphone-Enabled Machine Learning Algorithms for Autonomous Stroke Detection," was a collaborative effort between researchers from the UCLA David Geffen School of Medicine and multiple medical institutions in Bulgaria.

They collected data from 240 stroke patients across four metropolitan stroke centers.

Smartphone App to Assess Symptoms of Stroke

During the first 72 hours of symptom onset, the researchers utilized smartphones to capture patient videos, assessing facial asymmetry, arm strength, and speech changes, which are the indicators of stroke.

The team employed machine learning to assess facial asymmetry, meticulously analyzing 68 facial landmark points. They used data from a smartphone's internal 3D accelerometer, gyroscope, and magnetometer to examine arm weakness.

To identify speech changes, they utilized mel-frequency cepstral coefficients, a widely used sound recognition method converting sound waves into images, enabling a comparison between normal and slurred speech patterns.

The results showed that the app displayed high sensitivity and specificity, accurately detecting stroke in nearly all cases, as verified through neurologists' reports and brain scan data.

Dr. Radoslav Raychev, a vascular and interventional neurologist from UCLA's David Geffen School of Medicine, is enthusiastic about the app's potential impact.

He believes that using emerging technology and machine learning will aid more patients in identifying stroke symptoms early on. Rapid and accurate assessment of symptoms is crucial for ensuring that people with stroke have a better chance of survival and regaining their independence.

Machine Learning in the Field of Stroke Care

Dr. Radoslav Raychev and his team hope that deploying this app will change lives and revolutionize the future of stroke care.

"It's exciting to think how this app and the emerging technology of machine learning will help more patients identify stroke symptoms upon onset," said Raychev.

"Quickly and accurately assessing symptoms is imperative to ensure that people with stroke survive and regain independence. We hope the deployment of this app changes lives and the field of stroke care," he added.

While the smartphone app is still not public, it promises to transform stroke care by enabling individuals to recognize symptoms early on, seek prompt medical attention, and increase their chances of successful treatment and recovery.

The use of machine learning may just add a layer of precision, empowering patients to act swiftly and potentially save lives. The study's abstract can be found here.