Automated Artificial Intelligence To Predict Heart Failure
As Well As Perform Treatment Evaluation
CURENT PROTOCOLS We work with current MRI and CT clinical protocols - there is no change in the workflow.
PATIENT-SPECIFIC From the clinical images, we build a patient-specific cardiac digital twin using Artificial Intelligence.
BIOPHYSIOLOGICAL We extract the biophysiological features from the 4D patient-specific models,
DEEP LEARNING We harness the power of Deep Learning to analyse the geometrical 3D models along with the biophysiological patterns and extract features with high predictive power.
PREDICTION Using the features, we classify and predict the likelihood of an individual's risk of heart failure.
Automatic inspection, segmentation, artefacts correction, and generation of 3D/4D models
Deep learning models are used to provide an evidence-based clinical analysis predicting the risk of
Clinicians upload anonymized MRI or CT images
Results are returned along with the models. Clinicians can check the reports on their desktop or mobile
Clinicians gain advanced information on patient's current and future health outcome, aiding the decision making process
Considerable time is saved through our automated approach, allowing for a higher throughput of patients, in a smarter way
Treatment evaluation can be performed allowing for a cost-effective assessment of the treatment plan
Identification of individuals more prone to benefit from cardiac rehabilitation programs can be obtained from individuals' predicted health outcomes
- Currently only for use for Investigational and Research Purposes -
By 2035, 8 million adults living in the United States will have heart failure, costing a projected $1 trillion*
*American Heart Association