The clinical value of artificial intelligence (AI) technology capable of detecting blood clots (thrombi) in just three seconds in emergency stroke situations?where every minute and second counts?has been demonstrated.
JLK, a medical AI specialist company led by CEO Kim Dongmin, announced that its cerebral infarction analysis solution, 'JLK-CTL', has proven through large-scale clinical validation that it can rapidly and accurately detect large vessel occlusions. The results of this study were published in the latest issue of the official journal of the Korean Neurological Association, the 'Journal of Clinical Neurology', thereby earning international recognition for the reliability of the technology.
Stroke, especially severe 'large vessel occlusion', is an ultra-emergency condition in which a major blood vessel in the brain is blocked by a thrombus. In such cases, a non-contrast CT scan, which is routinely performed in emergency rooms, can reveal a hyperdense artery sign (HAS), where the artery blocked by a clot appears whiter than the surrounding tissue. This is a crucial early clue for identifying the blocked vessel. However, this signal is extremely subtle, making interpretation challenging, and diagnostic results can vary depending on the experience of the specialist.
JLK-CTL is an AI solution trained with deep learning technology to detect this very 'hyperdense artery sign'. It has learned from a vast number of CT images, enabling it to accurately identify even the most subtle clues that may be missed by the human eye, and to analyze the location, length, and volume of the thrombus in three dimensions.
The study was conducted using extensive patient data from 1,812 individuals at seven major stroke centers in Korea. It also included data from CT equipment made by different manufacturers, thereby verifying compatibility and generalizability in real-world hospital environments.
Most importantly, the study demonstrated that the larger the thrombus volume analyzed by the AI, the greater the actual area of brain damage (cerebral infarction). This suggests that JLK-CTL goes beyond simply detecting thrombi and can provide important evidence for predicting patient prognosis and determining treatment strategies in the future.
Based on this clinical data, JLK expects to accelerate global regulatory approvals, including from the US Food and Drug Administration (FDA) and Japan's PMDA, by securing the reliability and effectiveness of JLK-CTL. In particular, the algorithm was developed using data from various CT equipment vendors, giving it strong compatibility and applicability in global hospital settings.
Ryu Wiseon, Chief Medical Officer at JLK, stated, "JLK-CTL can support rapid and accurate diagnosis even in environments with a shortage of imaging specialists, so we expect high demand in the global market. The results of this study will also serve as important evidence for overseas expansion and market development."
This research was supported by the Korea Health Industry Development Institute and the Multi-Ministry Medical Device Development Project Group. Based on these results, JLK plans to expand cooperation with major hospitals in Korea and abroad, as well as with countries considering the adoption of AI medical devices.
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
