Quantifying Cardiac Tissue Composition using QuPath and Cellpose: An Accessible Approach to Postmortem Diagnosis SAE.

Abstract

Sudden death can be the first symptom of cardiac disease, and establishing a precise postmortem diagnosis is crucial for genetic testing and follow-up of relatives. Arrhythmogenic cardiomyopathy (ACM) is a structural cardiomyopathy that can be challenging to diagnose postmortem because of differences in structural findings and propagation of the disease at the time of death. Cases can have minimal or no structural findings and later be diagnosed according to genotype, known as concealed cardiomyopathy. Postmortem diagnosis often lacks clinical information, whereas antemortem diagnosis is based on paraclinical investigations that cannot be performed after death. However, the entire substrate is available, which is unique to postmortem diagnosis and research and can provide valuable insights when adding new methods. Reactive changes in the heart such as myocardial fibrosis and fat are significant findings. The patterns of these changes in various diseases are not yet fully understood and may be limited by sampling material and conventional microscopic diagnostics. We demonstrate an automated pipeline in QuPath for quantifying postmortem picrosirius red cardiac tissue for collagen, residual myocardium, and adipocytes, by integrating Cellpose into a versatile pipeline. This method was developed and tested using cardiac tissues from autopsied individuals. Cases diagnosed with ACM and age-matched controls were used for validation and testing. This approach is free and easy to implement by other research groups using this as a template. This can potentially lead to the development of quantitative diagnostic criteria for postmortem cardiac diseases, eliminating the need to rely on diagnostic criteria from endomyocardial biopsies that are not applicable to postmortem specimens. We propose that this approach serves as a template for creating a more efficient process for evaluating postmortem cardiac measurements in an unbiased manner, particularly for rare cardiac diseases.