Ventilaor_Pressure_Prediction

Inspiration

To help patients who have trouble breathing

What it does

Overcomes the cost barrier of developing new methods for controlling mechanical ventilators

How we built it

Using Python, we built a machine learning model that takes in five parameters:

1. R - lung attribute indicating how restricted the airway is (in cmH2O/L/S). Physically, this is the change in pressure per change in flow (air volume per time). Intuitively, one can imagine blowing up a balloon through a straw. We can change R by changing the diameter of the straw, with higher R being harder to blow.
2. C - lung attribute indicating how compliant the lung is (in mL/cmH2O). Physically, this is the change in volume per change in pressure. Intuitively, one can imagine the same balloon example. We can change C by changing the thickness of the balloon’s latex, with higher C having thinner latex and easier to blow.
3. time_step - the actual time stamp.
4. u_in - the control input for the inspiratory solenoid valve. Ranges from 0 to 100.
5. u_out - the control input for the exploratory solenoid valve. Either 0 or 1.
6. Using which we can predict the required pressure.

Built With

• flask
• googlecolab
• python