Automated assessment of cardiac dynamics in aging and dilated cardiomyopathy Drosophila models using machine learning

This research paper presents a novel machine learning platform for automated assessment of cardiac dynamics in Drosophila models of aging and dilated cardiomyopathy (DCM). The platform uses deep learning to segment optical microscopy images of Drosophila hearts, enabling the quantification of key cardiac parameters. The researchers validated their aging model using experimental datasets, achieving high accuracy in predicting fly age using both video classification and machine learning based on cardiac parameters. Furthermore, they extended their methodology to assess cardiac dysfunction associated with OGDH knockdown, demonstrating its potential for studying DCM. This versatile approach accelerates cardiac assays, promising broader applications in Drosophila and other animal models, including human cardiac physiology.

Yash Melkani, Aniket Pant, Yiming Guo, Girish C. Melkani