Biomechanic simulations for quantification of the ventilation/perfusion ratio in COVID-19 patients

Biomechanic simulations for quantification of the ventilation/perfusion ratio in COVID-19 patients is a PRACE-supported project led by Dr Simone Melchionna from the National Research Council (CNR), Italy. The project aims at a prognostic judgement of patient management based on the joint usage of pulmonary reconstruction, biomechanical simulations, physiological modelling, machine learning (ML), and artificial intelligence (AI). The project is a result of a collaboration between academic researchers, AI experts, a private entity, medical doctors from radiology, and intensive care units (ICU). The expected main outcome is to generate predictive values for oxygen and carbon dioxide levels in different ventilation operating scenarios, based on acquired time series, ventilator operating conditions, postures, age, habits, etc. Doctors could evaluate ventilation efficacy specifically to treat severe cases. Participants in the project consider that the expected results could provide quantitative guidance for ICUs pre-admission and post-admission evaluation, informing clinicians about those patients with comorbidities that require special attention in terms of ventilation operation conditions and manoeuvring. Simultaneously, results of the project could help hospitals quickly set up the new prognostic system, promote the standardisation of work, rationalise the workflow, and improve the efficiency of treatment, as well as medical safety. To make this project a success, PRACE awarded 30 000 000 core hours on Hawk, hosted by GCS at HLRS, Germany.