JLK (CEO Dongmin Kim), a stroke AI specialized company, announced that it presented the abstract of a study on acute cerebral infarction detection using JLK AI solution at the International Stroke Conference (ISC 2025) held at the LA Convention Center from the 5th to the 7th (local time).
This abstract presentation was given by Professor Jun Sunwoo, who concurrently serves as Professor of Radiology at Seoul National University Bundang and Chief Medical Officer (Director) at JLK. On the evening of the 7th (local time), Professor Sunwoo presented an abstract on the topic of ‘Improving the Detection Sensitivity of High-Difficulty Acute Cerebral Infarction Lesions on Diffusion-Weighted Imaging (DWI) Using JLK Artificial Intelligence Solution JBS-01K (JLK-DWI)’ at the ISC 2025 Special Session.
The study focused on the fact that even experts often miss minute lesions in the brainstem or hyperacute strokes in diffusion-weighted imaging (DWI), which is essential for diagnosing acute ischemic stroke. These lesions are classified as difficult cases even for medical professionals who have dealt with stroke frequently. Professor Seonwoo suggested that these difficult cases can be supplemented with JLK's stroke solution JBS-01K (JLK-DWI).
The study, which involved a total of 3,986 patients, was conducted from February 2017 to November 2021 on patients who underwent initial and follow-up DWI. It consisted of 130 high-difficulty cases and 120 controls out of a total of 250 cases.
As a result, the sensitivity improved from 74.6% for the diagnosis by medical staff alone to 90.6% when using the JLK AI solution, and the lesion segmentation accuracy (DSC) improved from 0.523 to 0.742. In addition, the sensitivity skyrocketed to 98.0% when diagnosed by AI alone, and it also detected an additional 79.6% of cases that the readers had missed.
The five medical professionals who participated in this study evaluated that the diagnostic reliability of difficult cases with AI assistance was higher. Based on this, the researchers concluded that JLK's AI solution JBS-01K (JLK-DWI) significantly improved diagnostic performance and lesion segmentation accuracy for microscopic lesions or hyperacute strokes when utilizing the AI solution.
Professor Seonwoo Jun said, “With the publication of this research abstract, we were able to widely publicize the excellent verification capabilities of JLK’s stroke AI solution to medical professionals around the world.” He added, “I hope that the utilization of stroke AI solutions will become more active, as they can relieve the burden and fatigue faced by medical professionals when diagnosing difficult cases and provide patients with appropriate treatment and rehabilitation plans.”
- See more related articles
You must be logged in to post a comment.