Our Primary Goal
The core objective of this project is to strengthen our company’s export potential and increase sales volumes in key foreign markets. By implementing targeted export promotion activities, we aim to convert existing sales opportunities into actual contracts and establish new, sustainable connections with potential clients.

Planned Activities

• Product Certification: We are investing in certification services to obtain FDA 510(k) additional clearances for the “Ligence Heart” product. This is a crucial step for the safe and legal placement of our product in the US market.
  
• International Exhibitions: We will present our company and product at international specialized exhibitions abroad to attract new clients and accelerate sales conversions. 

Expected Results

Through these activities, we expect to achieve the following outcomes:

• Present our products at 6 international exhibitions.
• Successfully certify 1 product.
• Significantly increase the export value of our Lithuanian-origin products over the coming years.

This project is implemented with a total budget of €102,552.63. It is co-funded by the European Union, which is providing €51,276.31 in support.

Lithuania, November 26th, 2025 – Ligence announces that its AI-powered echocardiography software, Ligence Heart, has received U.S. Food and Drug Administration (FDA) 510(k) clearance. Ligence Heart, is an AI-powered software that automates measurements and analysis of 2D transthoracic echocardiography (TTE) images and produces a structured report with several key cardiac structural and functional parameters. The clearance enables U.S. healthcare providers to generate reliable and reproducible cardiac measurements, addressing time-consuming tasks in echocardiography analysis.

Why this clearance matters

The number of cardiac diseases each year continues to rise. Manual measurements and reporting consume valuable clinician time, often creating backlogs and adding pressure on already-stretched cardiology teams. Ligence Heart automatically performs key cardiac measurements and generates a clinical report. It is engineered for consistency and operational efficiency with the goal of reducing repetitive manual work and creating a standardised workflow.

Saving clinicians time for actual patient care

“As a cardiologist myself, I know that we spend far too much time on repetitive manual tasks. Instead we should be focusing on increasing the accuracy of heart disease detection, especially heart failure. Automating core TTE measurements is a significant step towards that,” said Arnas Karužas, MD, co-founder and CEO of Ligence. “With FDA clearance, clinicians in the U.S. can now use Ligence Heart to reduce the workload in echo labs and improve overall workflow efficiency.”

Key findings

• Strong correlation between AI and manual measurements across stress stages
• Consistent accuracy in left ventricular volumes and ejection fraction quantification
• Feasibility confirmed for real-time, fully automated AI analysis in stress echocardiography

These results highlight the growing potential of AI to enhance stress echo workflows, enabling faster, more standardized, and objective cardiac function assessments.

Read the full study here


Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

The study evaluated Ligence’s AI-powered software for the automated assessment of left ventricular diastolic function. Using data from 302 patients, the results demonstrated strong alignment with expert evaluations and further confirmed the potential of AI in clinical echocardiography

Key findings

• Strong correlation with expert measurements on key parameters (E velocity r = 0.93, E/A ratio r = 0.94, LAVi r = 0.92)
• High accuracy in distinguishing normal vs severe dysfunction
• Some variability in intermediate grades, highlighting areas for further research and refinement

The study reinforces the feasibility and clinical value of AI-assisted echocardiography while also identifying the next challenges to address.

Read the full article here.

Ligence remains committed to advancing automation in echocardiography to make cardiac diagnostics faster, more consistent, and more accessible worldwide.



Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

The research compared left ventricular ejection fraction (LVEF) measurements from expert-assessed echocardiography and fully automated AI analysis, using cardiac magnetic resonance (CMR) as the gold standard reference.

Key findings

• Strong agreement with both CMR (R = 0.87) and expert measurements (R = 0.88–0.90)
• One Loop AI analysis showed even higher alignment with CMR (R = 0.89, κ = 0.75)
• Results confirm strong clinical potential, though further research is needed

These findings further support the role of AI-driven echocardiography as a reliable and accessible tool to assist clinicians in cardiac diagnostics.

Full study available here


Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

The research, led by Kruno Sveric and colleagues at the University Clinic, Technische Universität Dresden, involved 164 patients with various valvular diseases. The AI model performed complete view selection and LV wall measurements without any operator involvement.

Key findings:

• Strong correlation between AI and human measurements (r = 0.83, MAE = 22 g)
• Higher agreement between AI and CT (r = 0.69, MAE = 28 g) compared to human measurements (r = 0.62, MAE = 37 g)
• Reliable performance across different parameters: IVSD (r = 0.74), EDD (r = 0.73)

The results confirm that AI can accurately and consistently measure LV mass, performing on par with experts and showing even closer agreement with CT. This demonstrates how AI can streamline cardiac imaging workflows and improve measurement consistency in clinical practice.

Full study available here


Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

They pulled data from 1,245 patients and ran it through a fully automated pipeline. The AI selected the echo views on its own, then measured the key aortic parameters — the ones clinicians usually spend the most time double-checking. Once the outputs were in, the team compared them to clinician-measured values.

The correlation wasn’t just “acceptable,” it was strong across the board (R roughly 0.81 to 0.98). That’s the kind of alignment where you stop asking whether the model works and start asking where it fits into the workflow.

The value here isn’t replacing anyone. It’s standardization. Aortic stenosis grading swings a lot between operators. If an automated model can reduce that variability and tighten the workflow, clinicians get more reliable inputs and fewer borderline cases fall into diagnostic limbo.

The full study breaks down the dataset, methods, and validation details if you want to go deeper.
Read the full study here


Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

The setup is straightforward. The model first identifies and selects the correct echo views on its own. No manual steering. Once it has the right clips, it extracts the diastolic parameters automatically. Think E/e’, e’ velocities, mitral inflow metrics, the usual suspects. Then the AI outputs are compared directly against expert clinician measurements.

The interesting part is the agreement. The correlation sits in the range you’d actually trust in practice. It’s not hand-wavy alignment, it’s tight enough that you could fold this into a clinical workflow without feeling like you’re gambling on a black box.

The bigger story here is standardization. Diastolic assessments jump around because humans interpret them differently on different days. If an automated system can bring consistency to that workflow, you cut noise, reduce repeat scans, and free clinicians to focus on the non-routine cases.

If you want the specifics, the full study breaks down the methodology, dataset, and validation metrics:
Read the full study here


Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

A recent JACC Journals study, led by Sveric Kruno et al., compared AI-based echocardiography with expert evaluations for left ventricular (LV) mass assessment, validated against CT imaging.

The findings: AI delivered better accuracy, diagnostic performance, and repeatability compared to manual expert assessment.

Key Results

• 20% reduction in measurement error versus experts
• Higher diagnostic accuracy (AUC 0.78 vs 0.71 for linear; 0.84 vs 0.77 for hybrid methods)
• Highest test–retest reliability achieved with AI echocardiography

These results reinforce the growing role of AI in making cardiac imaging faster, more objective, and less dependent on operator variability-paving the way for more consistent patient outcomes.

Read the full study on JACC Journals


Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

The latest issue of the European Heart Journal – Cardiovascular Imaging showcases groundbreaking research in echocardiography software, with Ligence prominently featured. Four abstracts co-authored by our team have been published, marking a historic milestone: for the first time, AI has surpassed human performance in echocardiographic analysis. However, this is not about AI being better—it is about AI learning from the very best conditions cardiologists can work under and bringing that expertise into real-world clinical practice. This achievement underscores Ligence’s position as a leader in AI-powered echocardiography and its transformative impact on clinical workflows.

Highlights of the Published Research

Our four abstracts address key advancements in echocardiography software, showcasing the impact of AI-assisted analysis on clinical workflows and diagnostic accuracy. Here are the highlights:

1. AI Surpassing Human Performance in EF: This landmark study proves that AI-driven software provides more precise left ventricular function measurements than human cardiologists, setting a new standard in echocardiographic diagnostics. Read more
2. Another study demonstrates strong agreement between AI-based and expert cardiologist measurements of aortic valve hemodynamics, highlighting the potential of automated tools to streamline echocardiographic assessment of aortic stenosis. Read more
3. A multicenter study reveals that echocardiographic image quality significantly impacts the accuracy and yield of machine learning–based automated left heart measurements, underscoring the importance of high-quality imaging in AI-assisted diagnostics. Read more
4. Final study show how deep learning enables automated assessment of left ventricular function. Read more

The Impact of This Breakthrough

The publication of four Ligence abstracts in a single issue of EHJ – Cardiovascular Imaging marks a turning point in cardiac imaging. AI-driven echocardiography does not replace clinicians—it supports them by providing consistent, expert-level analysis even in time-pressured clinical environments. This milestone sets the stage for AI’s widespread adoption in echocardiographic diagnostics, ensuring every patient benefits from enhanced precision and efficiency.

As demand for accuracy and automation increases, Ligence remains at the forefront, delivering innovative solutions that empower cardiologists and improve patient outcomes.

Stay Updated

For the latest insights into AI-driven echocardiography, follow our updates and explore how Ligence is shaping the future of cardiac imaging. To learn more about our solutions, visit Ligence.io.

The implementation of these new guidelines highlights the need for precision and consistency in echocardiographic assessments. Advanced echocardiography software, like the solutions offered by Ligence, can play a crucial role in ensuring that strain analysis and LV diastolic function measurements adhere to the latest standards. By integrating cutting-edge software into your practice, you can enhance diagnostic accuracy and improve patient outcomes.


Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Innovation Council and SMEs Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.