How is AI revolutionizing the stroke care?

We are thrilled to announce our latest scientific publication in 2022, which is the result of perfect teamwork and has led to a promising discovery.

The world’s 3rd biggest killer is stroke. Stroke strikes every five minutes. All strokes are serious medical emergencies, but not all strokes are the same. The most common type of stroke is the so-called ischemic stroke, which occurs when a blood vessel that supplies blood to the brain becomes blocked. If one of the main arteries in the brain becomes blocked, it is considered a major vascular occlusion or LVO stroke. Depending on how severe the stroke is, the patient may need very different medical care.

The classification scale used today (NIHSS) is quite complex, requiring professional assistance even in ambulances. Our previous telemedicine solution focused on this limitation by “teleporting” the stroke specialist to the site via a secure communication platform, MOBIX. However, current research has gone even further.

Why don’t we simplify the LVO classification? What if stroke severity could be diagnosed with a simple test in ambulance and with an intelligent ML model? In the study our aim was to comprehensively assess the predictive ability of several clinical variables for LVO prediction and to develop an optimal combination of them using machine learning tools.

The study highlighted that machine learning tools are extremely useful in reducing the size of large data sets and in in assessing and optimizing predictive ability. The results of the study showed the potential of molecular biomarkers to optimize stroke diagnosis models.

Authors believe that these results underline the need for further detailed research, as it has great potential to revolutionize stroke diagnosis. For example, one extremely promising area is screening for a large number of potential biomarkers, i.e. the “omics” approach and the combined analysis of multi-omics data. Data analysis platforms, such as our DLX Smart DataLake solution can facilitate to organize and analyze large amounts of data with modern machine learning methods.

At E-Group we are continuously working on solutions to improve healthcare and open new avenues in AI driven digital health solutions.

Link to study: https://www.mdpi.com/2075-1729/12/2/230/htm