Automation of Mechanical Ventilation for Optimal Pressure Predictions Using Machine Learning

The onset of Covid-19 has brought a lot of concerns in the new era. This has affected a majority of people across the globe and has resulted to a significant rise in the number of patients who are also suffering from acute respiratory distress syndrome (ARDS), severe acute respiratory syndrome (SARS), and other types of respiratory failures which calls for the need of mechanical ventilation. The accurate prediction of the settings for mechanical ventilation along with the response of the lungs of the patient toward the adjustments of pressure will be helpful in improving the efficiency of the medical care treatment. A particular kind of sequence prediction problem is the prediction of lung pressure. The most effective technique for solving the sequence prediction problem is called Long Short-Term Memory (LSTM). LSTM has an advantage over standard Recurrent Neural Network (RNN), since it can selectively recall patterns over time. Thus, the proposed algorithmic approach addresses the challenges of pressure prediction in mechanical ventilation and depicts healthcare assistance with the optimal pressure predictions.