Skip to main content

SINAPSE – probably the most effective AI tool of its kind, already deployed and trusted by leading local medical institutions

AI-powered diagnostic assistant for acute ischemic stroke detection, delivering 93% of accuracy within just 15–30 seconds. It is based on native CT scans without perfusion – the standard scans that are fast and widely available in any stroke unit.

Current barriers to fast and reliable stroke diagnosis

Stroke is the 2nd leading cause of death in Europe, affecting over 1.1 million people annually. ~80% of strokes are ischemic, caused by blood clots blocking blood flow to the brain 1,2

Reference

1 Burden of Stroke in Europe 2 World Stroke Organization 

In ischemic stroke, every minute without treatment costs ~2 million neurons.  Rapid and accurate diagnosis is critical, but hard to achieve under the current conditions 1,2,3

Reference

1 Time is brain–quantified
2 Stroke: Every minute counts
3 CDC Grand Rounds: Public Health Strategies to Prevent and Treat Strokes

Radiologists are overloaded, working under time pressure and fatigue. Stroke-related CT interpretation remains error-prone, with diagnostic accuracy ranging from 45% to 88% 1,2

Reference

1 Work overload and diagnostic errors in radiology
2 Recommendations for Imaging of Acute Ischemic Stroke

Hospitals lack a solution that can reliably, instantly, and accurately assist stroke diagnosis. While research confirms AI’s potential native AI ischemic stroke detection remains a niche with very few viable solutions 1,2

Reference

1 Use of artificial intelligence in the management of stroke: scoping review
2 Accuracy of Automated Computer-Aided Diagnosis for Stroke Imaging: A Critical Evaluation of Current Evidence

SINAPSE: meeting an urgent market need with a solution designed to scale and grow

UNIQUE PRODUCT

SINAPSE is AI-driven diagnostic assistant for stroke detection, delivering 93% of accuracy within just 15–30 seconds, making it one of the most effective AI tools of its kind globally.

ADVANTAGEOUS POSITION

Unique medical data access enables the exclusive development of this and other medical AI tools.

STRONG FOUNDATION

Years of R&D completed and over €500,000 in initial funding already put into the product.

SUPPORTIVE ENVIRONMENT

EU initiatives are promoting the development and adoption of such solutions, making this a strong moment to enter the market and establish an early competitive position.

Advancing stroke care beyond alternatives

Most AI tools focus on contrast enhanced CT scans, which are more complex and costly, and are typically used when the initial diagnosis is already known. SINAPSE analyses the initial native CT scans without perfusion – the standard, non-contrast images available immediately and without extra equipment. This allows earlier stroke detection at a lower cost and with greater accessibility in emergency settings.

ACCURACY

93%

Currently SINAPSE has become one of the most effective AI tools of its kind globally, with an accuracy of 93%.

SPEED

15-30 sec

While doctors and other AI tools can take up to 20 minutes to deliver results, SINAPSE presents result in just 15 to 30 seconds.

INTEGRATION

customized

Unlike competitors who require separate, standalone solutions, SINAPSE integrates directly into existing hospital systems – no new platforms or logins. Intuitive by design.

FOCUS

patient centricity

Real-time AI assistance focusing on the earliest and most critical step in stroke diagnosis. It enables emergency and radiology teams to quickly assess and prioritize critical cases.

Clinically validated, already supporting real-world stroke care

Validated over 5 years in partnership with Pauls Stradins Clinical University Hospital and leading stroke experts in Latvia. System was trained on thousands of anonymized CT scans, curated and annotated by experienced radiologists.

Key results from clinical validation:

±0.62 ASPECTS score difference (vs. ±1.5 for average diagnosis)

93% diagnostic accuracy in stroke detection

15–30 seconds processing time per scan

Fully validated in hospital environment, compatible with existing systems

Continuous refinement through clinician feedback and new patient cases

More than a theoretical model — our clinically validated system is already supporting faster diagnostics and more accurate treatment decisions in real-world stroke care.

Meet the team behind SINAPSE

AGNIS JAKUBOVICS

Co-Founder 
Board Member

  • Founder and Chairman of the Board at APPLY – a company offering custom AI solutions
  • Experienced company manager with a track record in the IT industry
  • Skilled in business development, project planning, and leadership

ELVIJS GULBIS 

Business Dev and Sales
Product Developer

  • 10+ years in sales and marketing roles within healthcare and pharma
  • Held senior positions in the leading healthcare companies in Latvia, including Grindeks and MFD Healthcare Group

LINA BRINGE

Chief Delivery Officer 
AI Researcher

  • 10+ years at APPLY in positions ranging from software development to project management and CDO
  • Expertise in managing the production lifecycle of advanced AI and computer vision solutions.
  • Holding an MSc in Computer Software Engineering

med. Dr. ANDRIS VEISS

Co-Founder 
Radiologist

  • MD at Pauls Stradins Clinical University Hospital
  • Specialises in diagnostic and invasive radiology, invasive oncology
  • Board member of the Latvian Association of Invasive Radiologists

med. PhD Prof. KARLIS KUPCS

Co-Founder 
Medical Lead Radiologist

  • Professor at Riga Stradins University
  • Head of Department at Pauls Stradins Clinical University Hospital
  • Specialises in interventional neuroradiology, cerebral vascular pathology, CT and MRI

SERGEJS IVANOVS

Sr. AI Research and Dev
AI Developer

  • Over 10 years at APPLY as a programmer, now focused on AI research and development
  • Background in Computer Science with expertise in AI, computer vision, and neural networks

Overview

SINAPSE is an AI-powered diagnostic assistant for acute ischemic stroke detection that supports radiologists by identifying abnormalities in CT scans in real-time

Unlike most stroke detection AIs, SINAPSE analyses native (non-contrast) CT scans, requiring no extra equipment and facing very little competition

It processes the data within 15-30 seconds and has reached an accuracy of 93%, making it one of the most effective AI tools of its kind globally

Already in use by two leading medical institutions in Latvia, where it has received positive reviews regarding its quality and usefulness in existing hospital workflows. These partnerships support clinical validation and help refine upcoming features and expanded spectrum of diagnoses.

Interested in driving the next era of patient care with AI?
Connect with our team to explore investment opportunities:

ELVIJS GULBIS 

Business Dev and Sales
Product Developer

TEL: +371 277 309 67
MAIL: Elvijs.Gulbis@sinapsedx.com

    I have read and agree to the website's Privacy Policy.