Brain: JBS-01K


Ischemic Stroke - TOAST
Ischemic stroke subtype(TOAST) classification solution


Also available JBS-04K:



  • Quantitative analysis results assist physicians in decision making
  • Resolving difficulties in MR image interpretation, which is highly dependent on doctors experi-ence by quantitative analysis
  • Access management to enhance the security of the system
  • improvement of the system peiformance by constantly reflecting knowledge of experts
  • Convenient patient data management through PACS
  • Web-based solution: Users can use without installation


Based on the patient's brain MR imaging data and clinical data, the solution automatically classifications the subtype of ischemic stroke(TOAST) and assists medical staff in diagnosis of ischemic stroke.

JBS-01K is an AI medical system that diagnoses a subtype of ischemic stroke using patient’s MR image and atrial fibrillation (AF) information.
This system performs lesion detection and TOAST (Trial of ORG 10172 in Acute Stroke Treatment) classification of ischemic stroke using 3D hybrid artificial neural networks. JBS-01K provides classification probabilities by analyzing input MR images of 4 sequences (DWI, FLAIR, T1, T2) and clinical information data.

This system assists physicians to decide ischemic stroke subtype by quantitative analysis result. JBS-01K is currently the only medical device classified as ‘Computer Aided Detection and Diagnosis Software’ (A26430.14) and approved by the Ministry of Food and Drug Safety in Korea through a large- scale and multi-center clinical trial for safeness and effectiveness.


  • Auto-detection and visualization of brain stroke lesion
  • Automatic MR images matching with standard brain template (MIA Montreal Neurological Institute)
  • Registration of brain MR images matched with standard brain templates (MN!) • The probabilities of 4 ischernic stroke subtypes • According to the final decision of physician (or doctor, diagnosis results arc determined • Annotation tool for lesion marking and correction
  • Registration and visualization of patient clinical information • Convenient patient data management through PACS
  • Web-has. user interface • Analysis result report