Deep Learning for Noninvasive Pulmonary Hypertension Diagnosis 

Michał Grzeszczyk

Pulmonary Hypertension (PH) is a condition with a complex etiology that affects approximately 1% of adults and can progress undetected for many years. If untreated it leads to significant morbidity and mortality. The gold standard for PH diagnosis is an invasive Right Heart Catheterization (RHC) that directly measures the mean Pulmonary Artery Pressure (mPAP). RHC is an invasive procedure that carries risks, requires patient preparation, trained staff, highly specialized equipment, it is expensive and time-consuming. It has to be performed at a specialized facility to lower the probability of complications. In this context, the development of a non-invasive, accurate, and reliable technique for mPAP measurement could revolutionize the field, significantly enhancing the quality of life for individuals affected by PH.

Figure: The gold standard for mPAP measurement is the invasive RHC (a)¹. Our goal is to noninvasively estimate mPAP using deep learning and MRI (b)².

¹source: https://www.stanfordchildrens.org/en/topic/default?id=right-heart-catheterization-135-40

²source: https://www.mhvi.com/cardiac-mri/