Artificial intelligence in medicine Embedded AI preemptively identifies a stroke

Related Vendors

AITAD GmbH

FAULHABER Drive Systems

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

Wuxi InfiMotion Technology Co., Ltd.

Stroke is becoming a common disease. Modern technologies like embedded AI can save lives. Valuable time is no longer lost when a stroke is quickly detected thanks to embedded AI. For this, near-infrared images of veins are analyzed.

Modern technologies such as Embedded AI (Embedded Artificial Intelligence) assist in the fast detection and reaction: Stroke is becoming a common health issue. More and more people suffer from an acute injury of brain areas due to a vascular occlusion or a cerebral hemorrhage. In 2023 alone in Germany, 270,000 people suffered from a stroke, of which 200,000 for the first time and 70,000 for the repeated time.

For the future, experts predict even higher numbers. This is connected, among other things, with risk factors that are constantly increasing. Society is getting older, and the probability of suffering a stroke increases exponentially from the age of 55. In addition, almost two-thirds of men are overweight [1]. Therefore, stroke care will become more and more important in the future.

While one in five affected die within the first weeks after the stroke, 64 percent of patients require care—almost 15 percent of them need to be placed in care facilities. This results in major financial and social challenges that can only be mitigated by prevention. In addition to individual efforts, such as a healthier diet and quit smoking, modern technology can help detect and react early.

Quickly recognize stroke

Hardly anything is as crucial in the treatment of strokes and curbing the consequences as timely detection and response. Dr. Christian Seiß, a specialist in neurology, explains: "The more time that elapses—the longer brain tissue remains without oxygen—the more likely are severe and lasting damage after a stroke." If a stroke is treated in the first 90 minutes after onset, every fourth patient is symptom-free. If the treatment takes place within a period of 90 minutes to three hours, only every ninth patient is symptom-free. If the span of three hours between occurrence and treatment is exceeded, only one in 14 patients is symptom-free.

The costs for treating these impairments have significantly increased in recent years. They rose from 7.1 billion euros in 2004 to 17 billion euros in 2023—due to current developments, a further increase of 30 percent is expected by 2040.

Embedded AI detects a stroke

Already today, embedded AI can help to reduce consequential costs or even save lives by speeding up the detection of strokes and thus enabling quick medical treatment. The detection and detection of an impending stroke are made possible with near-infrared sensors that analyse different influences and investigate indications of conspicuous features. The characteristics can include an increased stress level or a changed vein structure.

"Imagine this technology in a mobile phone. Heart rate can already be measured with special apps today. Why not other diseases as well. We look at our display so much. The review could run in the background without any problems—and give a warning in time. The same applies to laptops, when employees—whether in the office, at the airport or in a restaurant before a customer meeting—are working," describes Viacheslav Gromov, CEO of AI specialist AITAD, an innovative business model.

Near-infrared can help

Infrared sensors, for example, can recognize the vein structure in a driver's face. Embedded AI analyzes this and can detect an impending stroke early. If the person is in the car, the vehicle could in such a case pull over to the side of the road and automatically call for help. The car is just one example: Analysis can take place anywhere that facial recognition is easily possible. This includes, but is not limited to, smartphones, screens, or displays. Mirrors or shower cubicles would also be suitable for such solutions.

To detect a stroke, near-infrared sensors are used which are mounted on embedded AI boards. This combines the possibilities of early detection with the general advantages of embedded AI. Due to the independent operation of embedded AI and the resulting network independence, the analysis can be carried out quickly and without delay. In addition, the sensitive personal data and biological conditions are better protected by the few interfaces than is the case with AI solutions with cloud connections.

"The demographic development forces us to find answers to new medical questions—such as the rising risk of stroke. Already today, we can detect these quickly and effectively and thus minimize financial but above all human damage. The applications of the solutions are more diverse than many companies think. So let's save lives together," concludes Gromov. (heh)

Sources

[1] Healthy nutrition. "You can change your habits!". German 'Tagesschau'. Retrieved on July 1, 2024.

[2] Stroke: Causes & Risk Factors Prevention. Retrieved on July 1, 2024.