Advantages:
- The deep learning-based approach enables accurate forecasting of left ventricular active contractions, providing early indicators of potential cardiovascular diseases, allowing for timely intervention and improved patient outcomes.
- The technology utilizes clinical metrics and imaging data, avoiding invasive procedures, and reducing patient discomfort while still delivering comprehensive cardiac evaluations.
- The research contributes to a better understanding of mechanical processes and cardiovascular mechanics by accurately representing cardiac structures.
Summary:
Cardiovascular diseases are a significant global health concern, accounting for over 40% of human mortality. The examination of elastic properties of soft biological tissue holds promise for early detection of illnesses or physiological complications. To better understand mechanical processes, accurate representations of cardiac structure using constitutive models are essential. Our researchers developed a cutting-edge constitutive-based deep learning approach to forecast the behavior of left ventricular active contractions by identifying the best-fitting material parameters from patient-specific clinical conditions.
The research yielded promising results, with the deep learning model producing smooth and accurate 𝛄 (gamma) waveforms that precisely reflected the cardiac cycle. Notably, the predicted waveform demonstrated peak active contraction during systole and gradually decreased to zero at end diastole. This breakthrough technology has the potential to enhance our understanding of cardiovascular mechanics and contribute to the early detection and management of cardiovascular diseases, ultimately saving lives and improving global health outcomes.
The Deep-learning Model Approach
Desired Partnerships:
- License
- Sponsored Research
- Co-Development