Decision support system for automatic classification of lv strain echocardiograms

Researcher:
Prof. Dan Adam | Biomedical Engineering

Categories:

Information and Computer Science | Medical Devices

The Technology

Evaluation of myocardial contractile function is essential for the diagnosis and treatment of cardiac diseases. Echocardiography is the most widely used non-invasive imaging modality in clinical practice. Echocardiographic strain imaging (developed by Prof. Dan Adam and initially implemented by GE Healthcare), which provides time-dependent strain curves (TSCs), is a promising technique allowing quantification of regional and layer-specific myocardial function and viability.

The main obstacles to application of strain measurements are the lack of consistency between manufacturers and the amount of manual labor required to achieve results. Thus, currently, only assessment of peak global longitudinal strain is employed in clinical practice.

The research team of Prof. Dan Adam has developed a supervised machine learning, physiologically constrained, fully automatic algorithm, trained with labeled data, for classifying the cardiac views, segmenting the LV walls, measuring the strains, and classifying the resultant TSCs into physiologic, pathologic or artifactual classes. The patented method and system may be implemented in a cloud-based decision support system to enable accurate, automatic, objective, and interactive diagnostic metrics of the left ventricle functions, concurrently supporting many users from anywhere in the world.

Advantages

  • Time saving
  • Consistent and objective results, with reliability estimation
  • Clinical decision support

Applications and Opportunities

  • Remote and point-of care echocardiography analysis
  • Automated analysis of echocardiography clips
  • Platform for additional echocardiography analysis services
arrow Business Development Contacts
Ofer Shneyour
Director of Business Development, ICT