Leveraging multimodal data to analyze cardiomyopathy

Overview

Multimodal data that includes images, structured and unstructured clinical data, genomic data, and electrocardiograms allow researchers to adopt a comprehensive approach to study important diseases like cardiomyopathy.

Challenge

Cardiomyopathy (CM) is a broad term for diseases that adversely impact heart muscle, which can disrupt oxygen flow to the body and cause heart failure (HF). Like other diseases, research into the causes of CM, patient outcomes, and interventions can be significantly enhanced using different kinds of real-world clinical, genomic, imaging, and electrocardiogram (ECG) data. ECG and cardiac imaging data are particularly important for CM, as ECGs are often used for diagnosis and images can define remodeling underlying CM.

NashBio worked with a client to provide a variety of multimodal CM data to better define how cardiac tissue is remodeled during CM and determine genetic CM factors.

At a Glance

ECG waveforms

NYHA

classification from notes

320

DNA samples with 124k genotyped

imaging studies

Our Impact

We initially worked with our client to refine a cohort that would be ideal for their multimodal research, ensuring that the cohort had imaging, genetic, as well as structured and unstructured EHR data.

During the cohort selection process, our clinical team performed chart reviews to identify New York Heart Association Functional Classes of Heart Failure as well as cardiovascular family history. Imaging data included cardiac echocardiograms and MRIs, while genomic data included both array and WGS sequences.

Using this data, our client is working to determine the genetic causes of CM and how CM impacts heart tissue. Our data and expertise would be applicable for similar types of research into other diseases, especially those that benefit from multimodal data.