Our Mission
Enhance the Cardiac MRI to promote accuracy for diagnosing heart disease, doing so will improve outcomes at the patient level, and help address the high cost of heart disease management at the payor level.
How we do it
Kardia Processes the Dicom File from a Cardiac MRI. We use Tissue Tagging and DENSE, both standard on a cardiac MRI. We compare the patient's heart to a proprietary DB of the normal heart. We calculate the Z-Score to determine the cardiac muscle viability.
Our founders have published a paper describing this approach:
Machine Learning Outcome Prediction in Dilated Cardiomyopathy Using
Regional Left Ventricular Multiparametric Strain.
https://pubmed.ncbi.nlm.nih.gov/33006006/
Workflow
Kardia processes the cMRI file
Kardia is a SaaS model that can interface throughout the radiology ecosystem via API's.
- Clinicians evaluate patients and order standard diagnostic studies.
- The cMRI meets both CMS Appropriate Use Criteria and American College of Cardiology non-invasive imaging guidelines.
- The imaging center provides the cMRI study. There is no change to the radiology workflow required.
- Kardia processes the cMRI dicom file, generates the output file that can be inserted or linked to the patient's medical record.
- The clinician reviews the Kardia results with the patient during follow up visits.
Kardia addresses the following
Improving diagnostic accuracy and efficiency
- Kardia provides a qualitative assessment at the micro-regional level of the cardiac tissue - we answer the most important question: What is the contractile injury to the left ventricle.
- Kardia provides an easy to interpret video of the cardiac structure, saving the clinician valuable time to diagnose and plan treatments.
- Kardia applies quantitative AI/ML to the study, and provides the ability to measure progression over multiple studies.