This document discusses using computational models and machine learning techniques to analyze cardiac shape and function from medical images. Key points include:
1) Models can regularize and extract diagnostic metrics from clinical data to provide clinical support and patient stratification.
2) Examples of metrics that can be extracted include cardiac shape, diastolic performance, and non-invasive blood pressure measurements.
3) There is an opportunity for these techniques to have a clinical impact but they must also demonstrate robustness.