Air detection system and method for detecting air in a pump of an infusion system
Various systems and methods for detecting air in a chamber of an infusion system are disclosed. In one embodiment, a determination is made that air is contained in the chamber on the basis of a change in the average force exerted against the plunger utilizing a derivative spike for event detection and a systematic reduction in the average force to confirm the nature of the change. In another embodiment, a determination is made that the chamber contains air when a difference between the current force profile and a baseline force profile crosses a threshold. In an additional embodiment, a force profile is classified as being an air force profile or a liquid force profile based on extracted features of the force profile.
1. A method for detecting air in a chamber of an infusion system comprising:
moving a plunger against the chamber containing fluid;
detecting a force acting on the plunger as the plunger moves against the chamber;
preprocessing a force profile based on the detected force acting on the plunger;
extracting features from the force profile; and
classifying the force profile as being an air force profile or a liquid force profile based on an application of machine learning analysis on the extracted features.
2. The method of claim 1 , wherein the classification of the force profile as being the air force profile or the liquid force profile is done without applying signal normalization to normalize to a baseline force profile.
3. The method of claim 1 , wherein the preprocessing comprises: acquiring the force profile; re-sampling the force profile for a set of angles; selecting a sub-set of angles for the force profile; and calculating a derivative of the force profile based on the force profile at the sub-set of angles.
4. The method of claim 1 , wherein the preprocessing comprises: acquiring the force profile; applying a low pass filter to the force profile; re-sampling the force profile for a set of angles; applying a range limit to the force profile; and calculating a difference of the force profile with respect to a baseline force profile.
5. The method of claim 1 , wherein the extraction of features comprises calculating scores of the force profile.
6. A system for detecting air in a chamber of an infusion system, the system comprising one or more hardware processors configured to:
detect a force acting on the plunger as the plunger moves against the chamber containing fluid;
preprocess a force profile based on the detected force acting on the plunger;
extract features from the force profile; and
classify the force profile as being an air force profile or a liquid force profile based on an application of machine learning analysis on the extracted features.
7. The system of claim 6 , wherein the classification of the force profile as being the air force profile or the liquid force profile is done without applying signal normalization to normalize to a baseline force profile.
8. The system of claim 6 , wherein the preprocessing comprises: acquiring the force profile; re-sampling the force profile for a set of angles; selecting a sub-set of angles for the force profile; and calculating a derivative of the force profile based on the force profile at the sub-set of angles.
9. The system of claim 6 , wherein the preprocessing comprises: acquiring the force profile; applying a low pass filter to the force profile; re-sampling the force profile for a set of angles; applying a range limit to the force profile; and calculating a difference of the force profile with respect to a baseline profile.
10. The system of claim 6 , wherein the extraction of features comprises calculating scores of the force profile.