Classifications

• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01F—MEASURING VOLUME, VOLUME FLOW, MASS FLOW OR LIQUID LEVEL; METERING BY VOLUME
  • G01F1/00—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow
  • G01F1/74—Devices for measuring flow of a fluid or flow of a fluent solid material in suspension in another fluid
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
  • A61M5/00—Devices for bringing media into the body in a subcutaneous, intra-vascular or intramuscular way; Accessories therefor, e.g. filling or cleaning devices, arm-rests
  • A61M5/14—Infusion devices, e.g. infusing by gravity; Blood infusion; Accessories therefor
  • A61M5/168—Means for controlling media flow to the body or for metering media to the body, e.g. drip meters, counters ; Monitoring media flow to the body
  • A61M5/16831—Monitoring, detecting, signalling or eliminating infusion flow anomalies
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
  • A61M5/00—Devices for bringing media into the body in a subcutaneous, intra-vascular or intramuscular way; Accessories therefor, e.g. filling or cleaning devices, arm-rests
  • A61M5/36—Devices for bringing media into the body in a subcutaneous, intra-vascular or intramuscular way; Accessories therefor, e.g. filling or cleaning devices, arm-rests with means for eliminating or preventing injection or infusion of air into body
  • A61M5/365—Air detectors
• • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
  • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
  • F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
  • F04B43/00—Machines, pumps, or pumping installations having flexible working members
  • F04B43/0009—Special features
  • F04B43/0081—Special features systems, control, safety measures
• • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
  • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
  • F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
  • F04B51/00—Testing machines, pumps, or pumping installations
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
  • A61M2205/00—General characteristics of the apparatus
  • A61M2205/33—Controlling, regulating or measuring
  • A61M2205/3306—Optical measuring means
  • A61M2205/331—Optical measuring means used as turbidity change detectors, e.g. for priming-blood or plasma-hemoglubine-interface detection

Definitions

• the disclosure relates to systems and methods for detecting air in an infusion system.
• a system and method is needed which more accurately detects air in the line of an infusion device.
• a method for detecting air in a chamber of an infusion system is disclosed.
• a plunger is moved against a chamber containing fluid with an actuator device.
• a force acting on the plunger, as it moves against the chamber is detected with a sensor.
• a measurement of the force acting on the plunger is electronically communicated from the sensor to a processor.
• a method for detecting air in a chamber of an infusion system is disclosed.
• a plunger is moved against a chamber containing fluid with an actuator device.
• a force acting on the plunger, as it moves against the chamber is detected with a sensor.
• a measurement of the force acting on the plunger is electronically communicated from the sensor to a processor.
• the processor is used to determine: (1 ) a baseline force profile; (2) a current force profile representing the current force acting on the plunger against the chamber; (3) a difference between the current force profile and the baseline force profile; and (4) that the chamber contains air when the calculated difference crosses a threshold.
• a method for detecting air in a chamber of an infusion system is disclosed.
• a plunger is moved against a chamber containing fluid using an actuator device.
• a force acting on the plunger, as it moves against the chamber is detected with a sensor.
• a measurement of the force acting on the plunger is electronically communicated from the sensor to a processor.
• the processor is used to: (1 ) preprocess a force profile detected by the sensor; (2) extract features from the force profile; and (3) classify the force profile as being an air force profile or a liquid force profile based on the extracted features of the force profile.
• Figure 1 illustrates a block diagram of a drug delivery infusion system under one embodiment of the disclosure
• Figure 2 illustrates a graph plotting a plunger sensor force curve per volume of fluid delivered
• Figure 3 illustrates a corresponding graph to Figure 2 plotting a plunger sensor force negative derivative curve per volume of fluid delivered
• Figure 4 illustrates a corresponding graph to Figures 2 and 3 plotting an in-line sensor ADC curve per volume of fluid delivered;
• Figure 5 illustrates one embodiment of a method, comprising a continuous flow chart, under the disclosure for determining whether air is contained in a chamber of a pump;
• Figure 6 illustrates a graph plotting, for a representative example, an average plunger sensor force curve, a plunger sensor force derivative curve, a baseline, a derivative threshold, a defined baseline range, an expected force differential ⁇ , and a ⁇ delay point threshold;
• Figure 7 illustrates a flowchart for one embodiment of a method for detecting air in a chamber of an infusion system
• Figure 8 illustrates a representative graph for one embodiment plotting a force sensor profile for a liquid curve and an air curve
• Figure 9 illustrates a representative graph for one embodiment of a principal component analysis (PCA) which was done on a plunger force profile
• Figure 10 illustrates a representative graph for one embodiment plotting a plunger force profile
• Figure 1 1 illustrates a representative graph for one embodiment plotting a liquid plunger force curve and an air plunger force curve
• Figure 12 illustrates a representative graph for one embodiment plotting, at an infusion rate of 20 ml/hour, the distribution of the maximum absolute difference between a reference plunger force profile and subsequent profiles comprising an air curve and a liquid curve;
• Figure 13 illustrates a representative graph for one embodiment plotting, at an infusion rate of 550 ml/hour the distribution of the maximum absolute difference between a reference plunger force profile and subsequent profiles comprising an air curve and a liquid curve;
• Figure 14 illustrates a representative graph for one embodiment plotting, at an infusion rate of 20 ml/hr, an air plot and a liquid plot;
• Figure 15 illustrates one embodiment of a method, comprising a continuous flow chart, under the disclosure for determining whether air is contained in a chamber of a pump;
• Figure 16 illustrates a continuation of the flow chart of Figure 1 5;
• Figure 17 illustrates a representative graph for one embodiment plotting a force sensor profile
• Figure 18 illustrates a graph plotting for each cycle of the plunger of Figure 17 six respective difference points representing the measured differences between a baseline, comprising liquid being in the chamber, and the corresponding points of each respective cycle of the plunger;
• Figure 19 illustrates a graph plotting the first six full cycles of the force sensor profile of Figure 17;
• Figure 20 illustrates a graph plotting for each of the first six full cycles of the plunger of Figure 18 six respective difference points representing the measured differences between a baseline, comprising liquid being in the chamber, and the corresponding points of each respective cycle of the plunger;
• Figure 21 illustrates a graph plotting the forty-second through forty-fifth cycles of the force sensor profile of Figure 17;
• Figure 22 illustrates a graph plotting for the forty-second through forty- fifth cycles of the plunger of Figure 18 six respective difference points representing the measured differences between a baseline, comprising liquid being in the chamber, and the corresponding points of each respective cycle of the plunger;
• Figure 23 illustrates a flowchart for one embodiment of a method for detecting air in a chamber of an infusion system
• Figure 24 illustrates one embodiment of a method, comprising a continuous flow chart, for determining whether air is contained in a chamber of a pump based upon a shape of the plunger force profile;
• Figure 25 illustrates a graph plotting air sensor data comprising representative points for each of fluid, air, and a transition
• Figure 26 illustrates a graph plotting average force profiles on the plunger corresponding to the embodiment of Figure 28 for each of fluid, air, and a transition;
• Figure 27 illustrates a graph plotting derivatives of the force profiles on the plunger corresponding to the embodiment of Figures 26 and 28 for each of fluid, air, and a transition;
• Figure 28 illustrates a graph applying a principal component analysis plotting representative points at an infusion rate of 2 milliliters per hour
• Figure 29 illustrates a graph plotting air sensor data comprising representative points for each of fluid, air, and a transition
• Figure 30 illustrates a graph plotting average force profiles on the plunger corresponding to the embodiment of Figure 32 for each of fluid, air, and a transition;
• Figure 31 illustrates a graph plotting derivatives of the force profiles on the plunger corresponding to the embodiment of Figures 30 and 32 for each of fluid, air, and a transition;
• Figure 32 illustrates a graph applying a principal component analysis plotting representative points at an infusion rate of 1 ,000 milliliters per hour;
• Figure 33 illustrates a flowchart for one embodiment of a method for detecting air in a chamber of an infusion system
• Figure 34 illustrates a flowchart of a Bayesian network showing a combination of algorithm sensors and a priori information which may be used to produce an indication of air-in-line or air in a chamber.
• the instant disclosure provides methods and apparatus for determining whether air is present in an infusion system.
• Several types of pumps such as
• SymbiqTM, PlumTM, and GemstarTM pumps sold by Hospira, Inc. involve the use of a cassette with a chamber that is compressed by an actuated plunger to pump fluid at a controlled rate from the drug container to the patient.
• the chamber is surrounded by valves which open and close in a complimentary manner to ensure unidirectional flow.
• the measured force during a pumping cycle is directly related to the type of fluid in the chamber. Fluids are relatively incompressible and generate a higher and different force profile than air.
• a combination of fluid and air in the chamber results in a hybrid shape profile which is indicative of the mixture percentages of both fluid and air.
• the instant disclosure discloses algorithms for utilizing the plunger force to detect the presence of air in the chamber to detect an air embolism prior to its infusion into a patient.
• an event detection algorithm determines a change from fluid to air in the pumping chamber on the basis of a change in the average force exerted against the plunger.
• the algorithm utilizes a derivative spike for event detection and a systematic reduction in the average force to confirm the nature of the change.
• a pattern recognition system/method for recognizing fluid, air, or a mixture thereof in a pumping chamber.
• the system normalizes the force signal/profile acting on the plunger against the chamber to a baseline.
• the system then preprocesses the force signal/profile to smooth and re-samples the x-axis to a standard sampling interval with respect to plunger position.
• the system then extracts features such as the maximum absolute difference between the baseline and each subsequent force profile, or other types of features.
• the system then classifies the force profile as being air, fluid, or a combination thereof using linear discriminate analysis or another type of analysis system/method.
• a varied pattern recognition system/method for recognizing fluid, air, or a mixture thereof in a pumping chamber.
• the system without normalizing to a baseline,
• preprocesses the force signal/profile acting on the plunger against the chamber by applying a low pass filter or by applying another type of preprocessing system/method.
• the system then extracts features from the entire force profile or a subset thereof such as the signal frequency content, the signal phase, the standard deviation or variance, the maximum, the range, the plunger position of critical points, the scores based on a principal component analysis, or extracts other types of features.
• the system then classifies the force profile as being air, fluid, or a combination thereof using linear discriminate analysis, k-nearest- neighbor, support vector machines, or another type of analysis system/method.
• One or more systems/methods of the disclosure include components that are optimized/customized according to a delivery rate of the fluid within the pumping chamber. Some of the existing algorithms fail to account for the profound impact of delivery rate on the observed plunger sensor force profile and the detection electronics. One or more systems/methods of the disclosure provide normalization or clustering states that reduce this impact and thereby improve sensitivity.
• One or more systems/methods of the disclosure may be combined with any existing systems/methods for detecting air in an infusion system to improve the reliability of air detection systems. For instance, many current
• systems/methods use acoustic or ultrasonic sensors to detect the presence of air in tubing segments.
• these systems/methods often do not consider the possibility of an acoustic short circuit or a bubble that is stuck or repetitively passes in front of the sensor.
• Many systems/methods rely on a single air ultrasonic sensor with a fixed threshold which separates the air sensor signal into two regions representing air and fluid. When a voltage is measured that is within the air signal region, the volume of air represented by the signal is accumulated until an alarm condition is met.
• the disclosure allows for the combination of the output of a force sensor signal with one or more air sensors to improve the reliability of existing air detection systems/methods. In doing so, the disclosed system/method does not require additional hardware
• the disclosure does not necessarily require the replacement of existing software modules for air detection but adds an additional safety and/or reliability layer to improve the robustness of existing air detection systems and methods.
• FIG. 1 illustrates a block diagram of a drug delivery infusion system 100 under one embodiment of the disclosure.
• the drug delivery infusion system 100 comprises: a fluid supply container 102; a fluid delivery line 104; an air sensor 105 connected to the fluid delivery line 104; a pump 106 comprising a plunger 107 moveably disposed against a chamber 108; an actuator device 109; a sensor 1 10; a positional sensor 1 1 2; a processing device 1 14; a non- transient memory 1 16 storing programming code 1 1 8; a clock 120; an alarm 122; an input/output device 124; and a delivery/extraction device 126.
• the drug delivery infusion system 100 may comprise a drug delivery infusion system such as the Plum A+TM, GemstarTM, SymbiqTM, or other type of drug delivery infusion system.
• the fluid supply container 1 02 comprises a container for delivering fluid such as IV fluid or a drug to the patient 1 28 through the chamber 1 08 due to movement of the plunger 107 against the chamber 1 08.
• the fluid delivery line 104 comprises one or more tubes, connected between the fluid supply container 102, the pump 1 06, and the delivery/extraction device 126, for transporting fluid from the fluid supply container 102, through the pump 106, through the delivery/extraction device 1 26 to the patient 128.
• the fluid delivery line 104 may also be used to transport blood, extracted from the patient 1 28 using the delivery/extraction device 1 26, as a result of a pumping action of the pump 106.
• the pump 106 comprises a pump for pumping fluid from the supply container 102 or for pumping blood from the patient 128.
• the pump 106 may comprise a plunger based pump, a peristaltic pump, or another type of pump.
• the chamber 108 comprises an inner cavity of the pump 106 into which fluid from the fluid supply container 102 is pumped into and through due to the moveably disposed plunger 107 moving against the chamber 1 08 as a result of the actuator device 109.
• the actuator device 109 may comprise a motor or another type of actuating device for moving the plunger 107 against the chamber 1 08.
• the sensor 1 1 0 is contained within the chamber 1 08 and detects the force acting on the plunger 107 as it moves against the chamber 1 08.
• the sensor 1 10 may comprise a force sensor signal comprising a pressure sensor, an elastic column, a strain gauge, or a piezoelectric crystal force transducer.
• the positional sensor 1 12 is used to determine a position of the plunger 107 against the chamber 108.
• the positional sensor 1 12 may comprise an encoder or may utilize the expected position based upon the commands sent to the actuator.
• the processing device 1 14 is in electronic communication with the pump 106, the actuator device 1 09, the sensor 1 10, the positional sensor 1 12, the non-transient memory 1 1 6 storing the programming code 1 18, the clock 120, the alarm 122, and the input/output device 124.
• the processing device 1 14 comprises a processor for processing information received from the pump 106, the sensor 1 10, the positional sensor 1 12, and the clock 1 20, and for executing a software algorithm, contained in the programming code 1 18 stored in the non- transient memory 1 16, to determine if air, liquid (fluid), or a combination thereof is located in the chamber 108 of the pump 1 06.
• the non-transient memory 1 16 may be located within or outside of the processing device 1 14.
• the clock 120 keeps time of activities of the drug delivery infusion system 100 including the plunger 107, the sensor 1 10, the positional sensor 1 12, and its other components.
• the alarm 1 22, when triggered by the processing device 1 14, is configured to notify the clinician as to the presence of air in the chamber 108, and to stop the pump 106 prior to an air embolism being delivered through the fluid delivery line 104 and the delivery/extraction device 126 to the patient 1 28.
• the input/output device 124 comprises a device which allows a clinician to input information, such as a user-inputted medication infusion program, to the processing device 1 14, and which also outputs information to the clinician.
• the delivery/extraction device 126 comprises a patient vascular access point device for delivering fluid from the fluid supply container 102 to the patient 128, or for extracting blood from the patient 128.
• the delivery/extraction device 1 26 may comprise a needle, a catheter, or another type of delivery/extraction device.
• the drug delivery infusion system [053] In one embodiment of the disclosure, the drug delivery infusion system
• 100 of Figure 1 may determine when air is present in the chamber 108 by analyzing the force on the plunger 107 and the derivative of the force acting on the plunger 107 per delivered volume of the fluid or air exiting the chamber 108.
• FIG. 2-4 illustrate typical data for one embodiment of a single iteration and end-of-bag event in which air is discovered in the chamber of Figure 1 .
• Figure 2 illustrates a graph plotting a plunger sensor force curve
• the Y-axis represents pounds of force on the plunger detected by a plunger force sensor and the X-axis represents volume in milliliters of the fluid delivered from the chamber.
• Figure 3 illustrates a corresponding graph to Figure 2 plotting a plunger sensor force negative derivative curve 127 per volume of fluid delivered.
• the Y-axis represents a derivative of the average force on the plunger of Figure 2 in pounds per unit volume and the X-axis represents volume in milliliters of the fluid delivered from the chamber.
• Figure 4 illustrates a corresponding graph to Figures 2 and 3 plotting an in-line air sensor ADC curve 129 per volume of fluid delivered.
• Y-axis represents an ADC count (also referred to as Analog-to-Digital-Count) of the fluid in-line as detected by an air sensor and the X-axis represents volume in milliliters of the fluid delivered from the chamber.
• ADC count also referred to as Analog-to-Digital-Count
• the X-axis represents volume in milliliters of the fluid delivered from the chamber.
• Figure 5 illustrates one embodiment of a method 132, comprising a continuous flow chart, under the disclosure for determining whether air is contained in a chamber of a pump.
• the method 132 may be implemented using the drug delivery infusion system 100 of Figure 1 with the plunger being moved with the actuator device against the chamber containing fluid, the sensor detecting a force acting on the plunger as it moves against the chamber, the processor processing the force measurements taken by the sensor and implementing programming code stored in a non-transient memory in order to determine whether air is contained in the chamber using the algorithm set forth in method 132, and the alarm being turned on if the processor determines that air is contained in the chamber which may trigger the pump being shut down.
• the method 132 may utilize the clock of the drug delivery infusion system 100 of Figure 1 to keep time of activities of the plunger or the sensor, and may use the positional sensor to determine a position of the plunger.
• the method 1 32 may utilize varying components to implement the method.
• step 134 the method starts. After step 134, in step 136 a
• step 136 determines whether the end-of-bag (EOB), or equivalent situation in which the chamber contains air, has been detected. If the answer to the determination in step 136 is 'yes' and the end of the bag has been detected, the method proceeds to step 138 and an end-of-bag alarm is turned on to indicate that air is in the chamber. This end-of-bag (EOB) event may pause the pump infusion or be used by another algorithm to qualify an air-in-line alarm. If the answer to the determination in step 136 is 'no' and the end of the bag has not been detected, the method proceeds to step 140 in which a determination is made as to whether there is a previously confirmed peak.
• EOB end-of-bag
• step 140 determines whether a trigger event has occurred in which the current negative derivative (of the average force) D(i) of the force of the plunger per delivered volume of the fluid exiting the chamber exceeds a derivative threshold D t which indicates the beginning of a possible end-of-bag (EOB) event signifying that air may have entered the chamber.
• D t the derivative threshold D t is flow dependent.
• the derivative threshold D t may be set to 1 .5 for a flow rate of the fluid below 200 milliliters per hour and to 3.0 for a flow rate of the fluid above 200 milliliters per hour. In other embodiments, the derivative threshold D th may be varied as a direct function of flow rate.
• step 1 50 determines whether the force of the plunger versus volume of the fluid delivered must exceed a derivative threshold D th in order to indicate a possible end-of-bag (EOB) event signifying that air may be in the chamber.
• step 1 50 determines whether the force of the plunger versus volume of the fluid delivered must exceed a derivative threshold D th in order to indicate a possible end-of-bag (EOB) event signifying that air may be in the chamber.
• the baseline B represents the average force during infusion when the chamber is filled with fluid.
• the baseline B comprises the average force acting on the plunger over a defined baseline range occurring up to the trigger event.
• the defined baseline range comprises the immediately previous 100 micro liters of average force data on the plunger taken immediately previous and up to the trigger event.
• the baseline range may comprise multiple cycles of average force data. In other embodiments, the baseline range may vary.
• the trigger event comprises the point at which the negative derivative force D(i) acting on the plunger per the delivered volume of the fluid exiting the chamber first exceeds the derivative threshold D th so long as subsequently the number of consecutive measured points P of the cycle of the plunger from the trigger event, in which the negative derivative force D(i) acting on the plunger per the delivered volume of the fluid exiting the chamber continues to exceed the derivative threshold D t , exceeds the consecutive point threshold P t .
• the trigger event may vary.
• step 142 the method proceeds to step 1 54 and makes a
• the current average force ⁇ ( ⁇ ) comprises the current average force on the plunger taken over a certain number of points of the cycle up to the current point of the plunger.
• the current average force on the plunger may be calculated based on two cycles of the plunger immediately preceding and up to the current point of the plunger.
• the current average force on the plunger may be taken over a varied range.
• the expected force differential ⁇ comprises .15 pounds of force. In other embodiments, the expected force differential ⁇ may vary.
• step 154 If the answer to the determination in step 154 is 'yes,' the method proceeds to step 156, confirms that an end-of-bag (EOB) or equivalent event has occurred, and proceeds through steps 146, 136, and 138 to turn on the end-of-bag alarm to indicate that air is in the chamber. This end-of-bag (EOB) event may turn off the pump. If the answer to the determination in step 154 is 'yes,' the method proceeds to step 156, confirms that an end-of-bag (EOB) or equivalent event has occurred, and proceeds through steps 146, 136, and 138 to turn on the end-of-bag alarm to indicate that air is in the chamber. This end-of-bag (EOB) event may turn off the pump. If the answer to the determination in step 154 is
• the method proceeds to step 158 and makes a determination as to whether the delay points DP is greater than a ⁇ delay point threshold.
• the ⁇ delay point threshold comprises a defined delay range, starting from the point of the trigger event, over which the differential between the baseline B and the current average force ⁇ ( ⁇ ) must exceed the expected force differential ⁇ in order to determine that an end-of-bag (EOB) event has occurred.
• the ⁇ delay point threshold comprises 200 micro liters of delivered fluid. In other embodiments, the ⁇ delay point threshold may vary.
• Figure 6 illustrates a graph plotting, for a representative example, an average plunger sensor force curve 162, a plunger sensor force (negative) derivative curve 164, a baseline 166, a derivative threshold 168, a defined baseline range 170, an expected force differential ⁇ 172, and a ⁇ delay point threshold 174.
• the right-most Y-axis represents average pounds of force on the plunger detected by a plunger force sensor
• the left-most Y-axis represents a derivative (average) of the force on the plunger of Figure 2 in pounds per milliliter
• the X-axis represents volume in milliliters of the fluid delivered from the chamber.
• the average plunger sensor force curve 162 comprises the average force delivered with each circle representing a measured point of the cycle of the plunger with measured data point 1 being the first circle shown in the graph.
• the plunger sensor force derivative curve 164 comprises the derivative force per volume delivered with each triangle representing a measured point of the cycle of the plunger with measured derivative data point
• the baseline 166 comprises a horizontal line 166.
• the derivative threshold 168 comprises a horizontal line.
• the defined baseline range 170 comprises a horizontal distance which in this example is 100 micro liters.
• the expected force differential ⁇ 172 comprises a vertical distance.
• the ⁇ delay point threshold 174 comprises a horizontal distance which in this example is 200 micro liters.
• the method of Figure 5 may be applied to the example of Figure 6 as follows to determine if air is contained in the chamber.
• step 134 the method starts.
• the method then proceeds to step
• the method then proceeds to step 146, increments variable i, and the method then repeats step 136.
• the method then proceeds to step 148 and increments variable P from 0 to 1 .
• step 1 50 determines that variable P which currently equals 1 is not greater than the consecutive point threshold P t of 1 .
• the method then proceeds to step 146, increments i to 18, and repeats steps 136, 140, and 144.
• the method then proceeds to step 148 and increments variable P from 1 to 2.
• the method then proceeds to step 150 and determines that variable P which currently equals 2 is greater than the consecutive point threshold P th of 1 .
• step 146 increments variable i to 19 and proceeds back to and repeats step 136.
• step 136 a determination is made that the end-of-bag (EOB) has not been detected for measured point 19.
• the method then proceeds to step 142 and increments the delay points DP to 1 .
• step 1 54 determines that the differential between the baseline B and the current average force G(i) for measured point i
• step 158 determines that the delay points DP of 1 is not greater than the ⁇ delay point threshold comprising the number of measured points in the cycle of the plunger, starting from the trigger event, within 200 micro liters of fluid delivered from the chamber.
• the method then proceeds to step 146, increments i and proceeds back to and repeats step 136.
• step 1 56 confirms that an end-of-bag (EOB) event has occurred, and proceeds through steps 146, 136, and 138 to turn on the end-of-bag alarm to indicate that air is in the chamber.
• the end-of-bag alarm being turned on may further comprise pausing the infusion.
• FIG. 7 illustrates a flowchart for one embodiment of a method 180 for detecting air in a chamber of an infusion system.
• the method 180 may be implemented using the drug delivery infusion system 100 of Figure 1 with the plunger being moved with the actuator device against the chamber containing fluid, the sensor detecting a force acting on the plunger as it moves against the chamber, the processor processing the force measurements taken by the sensor and implementing programming code stored in a non-transient memory in order to determine whether air is contained in the chamber using the algorithm set forth in method 180, and the alarm being turned on if the processor determines that air is contained in the chamber which may trigger the pump being shut down.
• the method 180 may utilize the clock of the drug delivery infusion system 100 of Figure 1 to keep time of activities of the plunger or the sensor, and may use the positional sensor to determine a position of the plunger, with each being in electronic communication with the processor. In other embodiments, the method 180 may utilize varying components to implement the method.
• step 182 a plunger is moved with an actuator device against a chamber containing fluid.
• step 184 a force acting on the plunger is detected with a sensor as the plunger moves against the chamber.
• step 1 86 a measurement of the force is electronically communicated from the sensor to a processor.
• step 187 a determination is made, with the processor, that the chamber contains air when: (1 ) a trigger event occurs in which a change in force acting on the plunger per delivered volume of the fluid exiting the chamber exceeds a threshold; and (2) subsequent to the trigger event a differential between a baseline average force acting on the plunger and a current average force acting on the plunger exceeds an expected force differential within a defined delay range.
• the processor turns on an alarm when the processor determines that the chamber contains the air. Step 188 may further comprise shutting down the pump when the alarm is turned on.
• step 187 may further comprise for a consecutive number of measured points of a cycle of the plunger from the trigger event the derivative force acting on the plunger per the delivered volume of the fluid exiting the chamber continuing to exceed the derivative threshold for more than a threshold number of the measured points of the cycle of the plunger against the chamber.
• the baseline average force of step (2) may comprise the average force acting on the plunger over a defined baseline range occurring up to the trigger event. The baseline average force may further represent the chamber being filled with the fluid.
• any of the steps of method 180 may be altered, not followed, or additional steps may be added.
• the drug delivery infusion system 100 of Figure 1 may determine when air is present in the chamber 108 by analyzing a shape of the force profile on the plunger 107 and determining that air is contained in the chamber 1 08 when the shape of the force profile on the plunger 107 changes significantly from a baseline shape of a force profile representing liquid being in the chamber 108. This is because it has been discovered that when air reaches the chamber 108, the shape of the force profile on the plunger 107 during a stroke or cycle of the plunger 107 changes in a consistent manner when and after the transition is made from fluid being in the chamber 108 to air being in the chamber 108.
• the shape of the force profile on the plunger 107 can be used as for detecting air-in-line by discriminating the force profile shapes associated with air and fluid.
• the characteristics of the shape of the force profile depend on the delivery rate of the fluid being delivered from the chamber 1 08 with some variability related to mechanism, set, fluid type, and distal and proximal pressure.
• Figure 8 illustrates a representative graph for one embodiment plotting a force sensor profile for a liquid curve 190 and an air curve 192.
• the Y-axis represents pounds of force on the plunger detected by a plunger force sensor and the X-axis represents a sample number collected at a rate of 250 Hz.
• Liquid curve 190 represents liquid being disposed in the chamber.
• Air curve 192 represents air being disposed in the chamber. As shown, the liquid curve 190 has substantially higher forces on the plunger than the air curve 192 during the expansion portion e of each cycle c of the plunger, while the difference between the curves 190 and 192 during the retraction phase r is significantly less.
• Figure 9 illustrates a representative graph for one embodiment of a principal component analysis (PCA) which was done on a plunger force profile with points 194 representing liquid being disposed in the chamber and points
• the X-axis represents a first score and the Y-axis represents a second score.
• the points 194 representing liquid being disposed in the chamber have a higher first score than the points 196 representing air being disposed in the chamber.
• Figure 10 illustrates a representative graph for one embodiment plotting a plunger force profile 198.
• the Y-axis represents pounds of force on the plunger detected by a plunger force sensor and the X-axis represents sample number for each cycle c of the plunger.
• six points p1 - p6, comprising a 6 point vector pattern are sampled at specific plunger positions during the expansion portion e of each cycle (or stroke) c of the plunger. No points are sampled during the retraction portion r of each cycle c of the plunger.
• This force sampling of each cycle may be used to determine whether air or liquid is contained in the chamber based on the shape of the measured force profile.
• the determination may be made using principle component analysis (PCA) to determine the correlation between the pattern variance of fluid versus air being in the chamber.
• PCA principle component analysis
• Pre-processing may be applied to normalize the patterns across sets/needle heights and varying mechanisms. Separate analysis is performed for each separate fluid infusion rate or ranges of infusion rates. In other embodiments, a varying number of points per cycle of the plunger may be utilized, and the determination may be made using varying types of analysis.
• Figure 1 1 illustrates a representative graph for one embodiment plotting a liquid plunger force curve 200 and an air plunger force curve 202.
• the Y-axis represents pounds of force on the plunger detected by a plunger force sensor and the X-axis represents a sample number of a cycle of the plunger.
• Liquid plunger force curve 200 represents liquid being disposed in the chamber.
• Air plunger force curve 202 represents air being disposed in the chamber. As shown, the liquid plunger force curve 200 has substantially higher forces on the plunger than the air plunger force curve 202.
• Figure 12 illustrates a representative graph for one embodiment plotting, at an infusion rate of 20 ml/hr, the distribution of the maximum absolute difference between a reference plunger force profile and subsequent profiles comprising an air curve 204 and a liquid curve 206.
• the Y-axis represents the number of profiles and the X-axis represents the difference associated with the point of the maximum absolute difference between the measured force profile and the baseline profile.
• Air curve 204 represents air being disposed in the chamber and liquid curve 206 represents liquid being disposed in the chamber. As shown, the liquid curve 206 has a substantially lower difference from the (liquid) baseline than the air curve 204.
• Figure 1 3 illustrates a representative graph for one embodiment plotting, at an infusion rate of 550 ml/hr, the distribution of the maximum absolute difference between a reference plunger force profile and subsequent profiles comprising an air curve 208 and a liquid curve 210.
• the Y-axis represents the number of profiles and the X-axis represents the difference associated with the point of the maximum absolute difference between the measured force profile and the baseline profile.
• Air curve 208 represents air being disposed in the chamber and liquid curve 210 represents liquid being disposed in the chamber. As shown, the liquid curve 210 has a substantially lower difference from the (liquid) than the air curve 208.
• Figures 12 and 13 demonstrate a significant difference between air and fluid across varying infusion rates after the maximum difference calculation is applied for feature extraction.
• Figure 14 illustrates a representative graph for one embodiment plotting, at an infusion rate of 20 ml/hr air depiction 212 and liquid depiction 214.
• the Y- axis represents the difference between the observed force and the and the X- axis represents two groups: (1 ) the group of differences associated with liquid in the plunger chamber; and (2) the difference with air in the plunger chamber
• Air depiction 212 represents air being disposed in the chamber and liquid depiction
• the air depiction 212 has a substantially lower (more negative) difference from the liquid baseline while the fluid 212 has a difference that is close to zero from the liquid baseline.
• the separation between the two groups provides the basis for a method for discriminating force measurements associated with air from those associated with fluid.
• each force shape profile of the plunger is compared to a baseline force profile, a point-by-point difference between the force shape profile and the baseline force profile is determined, and when the minimum (most negative) difference between the force shape profile and the baseline force profile drops below a threshold a determination is made that the chamber contains air.
• the baseline force profile may represent liquid being in the chamber.
• varying algorithms may be implemented to determine when air is contained in the chamber based on the force shape profile of the plunger.
• Figures 15 and 16 illustrate one embodiment of a method 220, comprising a continuous flow chart, under the disclosure for determining whether air is contained in a chamber of a pump.
• the method 220 may be implemented using the drug delivery infusion system 100 of Figure 1 with the plunger being moved with the actuator device against the chamber containing fluid, the sensor detecting a force acting on the plunger as it moves against the chamber, the processor processing the force measurements taken by the sensor and implementing programming code stored in a non-transient memory in order to determine whether air is contained in the chamber using the algorithm set forth in method 220, and the alarm being turned on if the processor determines that air is contained in the chamber which may further shut down the pump.
• the method 220 may utilize the clock of the drug delivery infusion system 100 of Figure 1 to keep time of activities of the plunger or the sensor, and may use the positional sensor to determine a position of the plunger. In other embodiments, the method 220 may utilize varying components to implement the method.
• step 222 the method starts. After step 222, the method proceeds through location step 224 to step 226.
• the force profile X(k) comprises a vector comprising the six forces on the plunger at each of the six positions/points of the plunger during the k cycle of the plunger. In other embodiments, the force profile may be acquired with a varying number of positions of the plunger.
• step 230 determines that the profile count PC is less than or equal to the number of initial cycles of the plunger to ignore Ni then the method proceeds back to and repeats steps 224, 226, 228, and 230 until the profile count PC is not less than or equal to the number of initial cycles of the plunger to ignore Ni at which point the method proceeds to step 232.
• step 232 a determination is made as to whether the baseline count BS_LN_CNT is less than the baseline ready variable BS_LN_RDY. The baseline count
• BS_LN_CNT is not less than BS_LN_RDY than the method proceeds through location step 234 of Figure 15, through location step 236 of Figure 16, to step
• step 232 If in step 232 a determination is made that the baseline count
• BS_LN_CNT is less than BS_LN_RDY than the method proceeds to step 240.
• step 240 a determination is made as to whether the Analog-To-Digital-Count
• ADC Analog-To-Digital-Count
• step 242 determines whether the baseline count BS_LN_CNT is greater than 0. In other embodiments, the lag number of cycles used may vary. If step 242 determines that the baseline count BS_LN_CNT is not greater than 0 then the method proceeds back to location step 224 to step 226 and continues the loop. If step 242 determines that the baseline count BS_LN_CNT is greater than 0 then the method proceeds through location step 234 of Figure 15, through location step 236 of Figure 16, to step 238 of Figure 16 which is discussed later on.
• step 240 If in step 240 the determination is made that either the Analog-To-Digital- Count (ADC) read by an air sensor downstream of the chamber is less than the primary threshold for fluid TPRI (which means that liquid is in the chamber), and that the profile count PC is greater than the number of initial cycles of the plunger to ignore plus 2 (indicating that the lag of 2 cycles, due to the air sensor being located downstream of the chamber, has been completed) represented by PC being greater than Ni + 2, then the method proceeds to step 244.
• ADC Analog-To-Digital- Count
• the equation for Xt may be varied.
• step 246 the method proceeds to step 248 and determines the baseline force profile Xm, expressed as a 6 point vector, using the equation Xm
• the baseline force profile Xm represents the baseline force vector for a situation in which liquid (fluid) is contained in the chamber 2 cycles prior to the current cycle due to the air sensor being located downstream of the chamber. In other embodiments, the baseline force profile Xm may be calculated using varying equations.
• step 250 determines whether D is greater than a threshold for a given infusion rate Trate which is set to -.3.
• Trate may be set to a varying number depending on the infusion rate or other factors, such as the signal variance. Additionally, more than one value for Trate may be used to provide regions of high probability versus low probability. If a determination is made in step 250 that the minimum distance D between the current vector force profile of the plunger and the baseline force vector is not greater than the threshold for a given infusion rate Trate, which indicates that air is in the chamber, then the method proceeds to step 252.
• step 254 determines whether Nw count is greater than or equal to Nw with Nw representing the threshold number of consecutive observed air cycles of the plunger after which an air alarm will be turned on indicating that air is contained in the chamber. If a determination is made in step 254 that Nw count is not greater than or equal to Nw than the method proceeds through location step
• step 256 back to location step 224 of Figure 15 to step 226 of Figure 15 and repeats the loop. If a determination is made in step 254 that Nw count is greater than or equal to Nw than the method proceeds to step 258, sets Flag Delta to 1 indicating that air is present in the chamber, turns on an alarm to indicate that air is present in the chamber, and proceeds through location step 256 back to location step 224 of Figure 15 to step 226 of Figure 15 and repeats the loop.
• Step 258 may further comprise shutting down the pump.
• step 250 If a determination is made in step 250 that the minimum distance D between the current vector force profile of the plunger and the baseline force vector is greater than the threshold for a given infusion rate Trate, indicating that liquid is contained in the chamber, then the method proceeds to step 260.
• step 260 Nw count is reset to 0 with Nw count representing the current number of observed air cycles, and Flag Delta is also reset to 0 with Flag Delta representing that air is present in the chamber.
• the baseline ready variable BS_LN_RDY may be set to other values.
• step 262 If a determination is made in step 262 that the baseline count
• BS_LN_CNT is not greater than or equal to the baseline ready variable
• step 264 If a determination is made in step 262 that the baseline count BS_LN_CNT is greater than or equal to the baseline ready variable BS_LN_RDY then the method proceeds to step 264.
• the forgetting rate a may be set to .1 . In other embodiments, the forgetting rate a may be set to varying values.
• the adaptive baseline may be determined in alternate manners such as a moving average or Kalman filter.
• Step 264 comprises an adaptive baseline step which allows the user to assert control over the baseline force profile Xm by controlling the forgetting rate a.
• the forgetting rate a may be pre-programmed.
• varying ways may be used to calculate the baseline force profile Xm.
• the method proceeds through location step 256 back to location step 224 of Figure 1 5 to step 226 of Figure 15 and repeats the loop.
• one or more steps of the method 220 may be modified, not followed, or one or more additional steps may be added.
• any of the variables of the method 220 may be either user set, using an input device, or pre-set into the processor.
• Figure 17 illustrates a representative graph for one embodiment plotting a force sensor profile 266.
• the Y-axis represents pounds of force on the plunger detected by a plunger force sensor and the X-axis represents time in seconds.
• Six points p1 - p6 are calculated during the expansion portion e of each cycle c of the plunger. No points are sampled during the retraction portion r of each cycle c of the plunger.
• Line 268 represents the point, during the forty- fifth cycle of the plunger, at which an air alarm is turned on due to air being in the chamber when the method 220 of Figures 15 and 16 is applied which is discussed more thoroughly below.
• Figure 18 illustrates a graph plotting for each cycle c of the plunger of
• Figure 17 six respective difference points dp representing the measured differences between a baseline, comprising liquid being in the chamber, and the corresponding points of each respective cycle of the plunger.
• the Y-axis represents the differences and the X-axis represents time.
• the circled points cp represent the minimum difference for each cycle of the plunger between the six respective difference points dp of each cycle of the plunger and the baseline.
• Line 270 represents the threshold for a given infusion rate Trate which is set to -
• the method determines that liquid is contained in the chamber during the first forty-three cycles of the plunger, determines that air is in the chamber during the forty-fourth cycle of the plunger, and after line 268, as it does in Figure 17, turns on an air alarm during the forty-fifth cycle of the plunger representing that air is in the chamber.
• Figure 19 illustrates a graph plotting the first six full cycles C1 - C6 of the force sensor profile 266 of Figure 17.
• the Y-axis represents pounds of force on the plunger detected by a plunger force sensor and the X-axis represents time in seconds.
• Six points p1 - p6 are calculated during the expansion portion e of each cycle c of the plunger. No points are sampled during the retraction portion r of each cycle c of the plunger.
• Figure 20 illustrates a graph plotting for each of the first six full cycles C1
• Figure 21 illustrates a graph plotting the forty-second through forty-fifth cycles C42 - C45 of the force sensor profile 266 of Figure 1 7.
• the Y-axis represents pounds of force on the plunger detected by a plunger force sensor and the X-axis represents time in seconds.
• Six points p1 - p6 are calculated during the expansion portion e of each cycle c of the plunger. No points are sampled during the retraction portion r of each cycle c of the plunger.
• Figure 22 illustrates a graph plotting for the forty-second through forty- fifth cycles C42 - C45 of the plunger of Figure 18 six respective difference points dp representing the measured differences between a baseline, comprising liquid being in the chamber, and the corresponding points of each respective cycle of the plunger.
• the Y-axis represents the differences and the X-axis represents time.
• the circled points cp represent the minimum difference for each cycle of the plunger between the six respective difference points dp of each cycle of the plunger and the baseline.
• Line 270 represents the threshold for a given infusion rate Trate which is set to -.3.
• the method determines that liquid is contained in the chamber during the first forty-three cycles of the plunger, determines that air is in the chamber during the forty-fourth cycle of the plunger, and after line 268, as it does in Figure 17, turns on an air alarm during the forty-fifth cycle of the plunger representing that air is in the chamber.
• Xm (.574252, 1 .192627, 1 .990768, 2.551261 , 3.144921 ,
• FIG 23 illustrates a flowchart for one embodiment of a method 280 for detecting air in a chamber of an infusion system.
• the method 280 may be implemented using the drug delivery infusion system 100 of Figure 1 with the plunger being moved with the actuator device against the chamber containing fluid, the sensor detecting a force acting on the plunger as it moves against the chamber, the processor processing the force measurements taken by the sensor and implementing programming code stored in a non-transient memory in order to determine whether air is contained in the chamber using the algorithm set forth in method 280, and the alarm being turned on if the processor determines that air is contained in the chamber which may trigger the pump being shut down.
• the method 280 may utilize the clock of the drug delivery infusion system 100 of Figure 1 to keep time of activities of the plunger or the sensor, and may use the positional sensor to determine a position of the plunger, with each being in electronic communication with the processor. In other embodiments, the method 280 may utilize varying components to implement the method.
• step 282 a plunger is moved, with an actuator device, acting against a chamber containing fluid.
• a sensor is used to detect a force acting on the plunger as it moves against the chamber.
• step 286 a measurement of the force is electronically communicated from the sensor to a processor.
• the processor determines: (1 ) a baseline force profile; (2) a current force profile representing the current force acting on the plunger against the chamber;
• step 290 the processor turns on an alarm when the processor determines that the chamber contains the air.
• Step 290 may further comprise shutting down the pump when the alarm is turned on.
• the baseline force profile represents the chamber being filled with the fluid.
• the processor determines the baseline force profile by taking force measurements at a plurality of plunger positions over a cycle of the plunger against the chamber.
• the processor determines the baseline force profile by averaging force measurements taken over a plurality of cycles of the plunger against the chamber.
• the processor determines the baseline force by additionally taking into account the current force profile acting on the plunger during a current cycle of the plunger against the chamber.
• the processor further applies a forgetting rate, moving average or Kalman filter which controls what portion of the updated baseline force profile is made up of the average or estimated baseline force measurements and what portion of the updated baseline force profile is made up of the current force profile.
• the processor determines the current force profile by taking force measurements at a plurality of plunger positions over a current cycle of the plunger against the chamber.
• the processor calculates the difference between the current force profile and the baseline force profile by calculating respective differences between a plurality of points of the current force profile relative to a respective plurality of points of the baseline force profile, and determining a minimum difference of the respective differences or an absolute maximum difference of the respective differences.
• the processor determines that the chamber contains the air when the minimum difference is less than the threshold. In still another embodiment, the processor determines that the chamber contains the air when the calculated difference is below the threshold. In other embodiments, any of the steps of the method 280 may be altered, not followed, or additional steps may be added.
• Figure 24 illustrates one embodiment of a method 300, comprising a continuous flow chart, under the disclosure for determining whether air is contained in a chamber of a pump based upon a shape of the plunger force profile.
• the method 300 may be implemented using the drug delivery infusion system 100 of Figure 1 with the plunger being moved with the actuator device against the chamber containing fluid, the sensor detecting a force acting on the plunger as it moves against the chamber, the processor processing the force measurements taken by the sensor and implementing programming code stored in a non-transient memory in order to determine whether air is contained in the chamber using the algorithm set forth in method 300, and the alarm being turned on if the processor determines that air is contained in the chamber which may trigger the pump being shut down.
• the method 300 may utilize the clock of the drug delivery infusion system 100 of Figure 1 to keep time of activities of the plunger or the sensor, and may use the positional sensor to determine a position of the plunger. In other embodiments, the method 300 may utilize varying components to implement the method.
• step 302 the method 300 starts. After step 302, the method proceeds through location step 304 to step 306. In step 306 a force profile over one cycle of a plunger of the chamber is acquired using the sensor. In one embodiment, as shown in box 308, the sampling frequency may be 62.5 Hz. In other embodiments, varying parameters may be used. After step 306, the method proceeds to step 310 and re-samples the force profile for the cycle of the plunger at uniform increments with respect to position or at specific positions. In one embodiment, as shown in box 312, the re-sampling may take place over a set of angles and may be performed using linear, quadratic or cubic
• step 310 the method proceeds to step 314 and selects a sub-set of angles (i.e. one or more ranges).
• the sub-set of angles may comprise a range of angles based on the infusion rate.
• varying parameters may be used.
• step 318 the method proceeds to step 318 and calculates a derivative.
• this step may comprise simultaneously applying a smoothing operation.
• this step may comprise applying varying parameters.
• Steps 306 through 318 comprise acquisition and preprocessing steps.
• N 8.
• the scores may be calculated using varying parameters.
• step 326 the method proceeds to step 330 and determines a classification based on the result of the linear discriminate analysis. This step may also consider class means by infusion rate as shown in box 328.
• step 330 the method proceeds to step 332 and determines whether air is in the chamber based on the classification.
• step 332 determines that air is contained in the chamber then the method proceeds to step 334 and sounds an air alarm during which the pump may be shut down. If step 332 determines that air is not in the chamber based on the classification then the method proceeds back to location step 304.
• step 346 may be followed.
• step 338 a force profile of the plunger over one cycle of a plunger of the chamber is acquired using the sensor.
• the sampling frequency may be 62.5 Hz.
• varying parameters may be used.
• step 340 a low pass filter is applied.
• step 342 a re-sampling is done. In one embodiment, as shown in box 31 2, the re-sampling may take place over a set of angles.
• varying parameters may be used.
• step 344 a range limit is applied. In one embodiment, as shown in box 316, a sub-set of angles comprising a range of angles based on the infusion rate. In other embodiments, varying parameters may be used.
• step 346 a difference is calculated. In one embodiment, this difference may comprise determining differences in points of the force profile. In other embodiments, this difference may use varying parameters.
• Figure 25 illustrates a graph plotting air sensor data comprising representative points for each of fluid 348, air 350, and transition 352.
• the Y- axis represents an ADC count of the fluid of the chamber measured by a sensor and the X-axis represents sample number.
• the graph provides the observed air sensor ADC readings versus sample number over 1 0 aggregated runs. Each run ends with a transition from fluid to air.
• the fluid readings (hashed symbols) 348, are clearly differentiated from those associated with air (open symbols) 350. Points that are close to or on a transition region from air to fluid are marked by solid symbols 352.
• Figure 26 illustrates a graph plotting force average profiles on the plunger corresponding to the embodiment of Figure 25 for each of fluid 348, air 350, and transition 352.
• the graph demonstrates systematic differences in the average force associated with the three states (fluid 348, air 350, and transition 352) of Figure 25.
• the Y-axis represents force and the X-axis represents an angular position of the motor powering the plunger.
• Figure 27 illustrates a graph plotting derivatives of the force profiles on the plunger corresponding to the embodiment of Figures 26 and 28 for each of fluid 348, air 350, and transition 352.
• the Y-axis represents a derivative of the force and the X-axis represents an angular position of the motor powering the plunger.
• the graph demonstrates that the systematic differences between the three states of Figure 28 can be enhanced and differentiated from mechanism specific variation through the application of the first derivative.
• Figure 28 illustrates a graph applying a principal component analysis to plot representative points at an infusion rate of 2 milliliters per hour, with hashed symbols representing fluid points 348, open symbols representing points associated with air 350, and solid points representing transitional
• Figure 29 illustrates a graph plotting air sensor data comprising representative points for each of fluid 348, air 350, and transition 352.
• the Y- axis represents an ADC count of the fluid of the chamber measured by a sensor and the X-axis represents sample number.
• the graph provides the observed air sensor ADC readings versus sample number over 1 0 aggregated runs. Each run ends with a transition from fluid to air.
• the fluid readings (hashed symbols) 348, are clearly differentiated from those associated with air (open symbols) 350. Points that are close to or on a transition region from air to fluid are marked by solid symbols 352.
• Figure 30 illustrates a graph plotting force average profiles on the plunger corresponding to the embodiment of Figure 32 for each of fluid 348, air 350, and transition 352.
• the graph demonstrates systematic differences in the average force associated with the three states (fluid 348, air 350, and transition 352) of Figure 29.
• the Y-axis represents force and the X-axis represents an angular position of the motor powering the plunger.
• Figure 31 illustrates a graph plotting derivatives of the force profiles on the plunger corresponding to the embodiment of Figures 30 and 32 for each of fluid 348, air 350, and transition 352.
• the Y-axis represents a derivative of the force and the X-axis represents an angular position of the motor powering the plunger.
• the graph demonstrates that the systematic differences between the three states of Figure 32 can be enhanced and differentiated from mechanism specific variation through the application of the first derivative.
• Figure 32 illustrates a graph applying a principal component analysis to plot representative points at an infusion rate of 1 ,000 milliliters per hour, with hashed symbols representing fluid points 348, open symbols representing points associated with air 350, and solid points representing transitional
• Figure 33 illustrates a flowchart for one embodiment of a method 360 for detecting air in a chamber of an infusion system.
• the method 360 may be implemented using the drug delivery infusion system 100 of Figure 1 with the plunger being moved with the actuator device against the chamber containing fluid, the sensor detecting a force acting on the plunger as it moves against the chamber, the processor processing the force measurements taken by the sensor and implementing programming code stored in a non-transient memory in order to determine whether air is contained in the chamber using the algorithm set forth in method 360, and the alarm being turned on if the processor determines that air is contained in the chamber which may trigger the pump being shut down.
• the method 360 may utilize the clock of the drug delivery infusion system 100 of Figure 1 to keep time of activities of the plunger or the sensor, and may use the positional sensor to determine a position of the plunger, with each being in electronic communication with the processor. In other embodiments, the method 360 may utilize varying components to implement the method.
• step 362 a plunger is moved, with an actuator device, against a chamber containing fluid.
• step 364 a sensor is used to detect a force acting on the plunger as it moves against the chamber.
• step 366 a measurement of the force is electronically communicated from the sensor to a processor.
• the processor (1 ) preprocesses a force profile detected by the sensor; (2) extracts features from the force profile; and (3) classifies the force profile as being an air force profile or a liquid force profile based on the extracted features of the force profile.
• the processor turns on an alarm when the processor determines that the chamber contains the air. Step 370 may further comprise shutting down the pump when the alarm is turned on.
• the processor classifies the force profile as being the air force profile or the liquid force profile without applying signal normalization to normalize to a baseline force profile. In another embodiment, the processor further applies a signal normalization to normalize the force profile relative to a baseline force profile. In an additional embodiment, the processor
• the processor preprocesses the force profile detected by the sensor by: 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.
• the processor preprocesses the force profile detected by the sensor by:
• the processor extracts the features from the force profile by at least one of calculating scores of the force profile or applying a linear discriminate analysis to the force profile.
• the processor calculates the scores of the force profile by multiplying a derivative of the force profile by a set of eigenvectors, and applies the linear discriminate analysis by multiplying the scores by weights.
• the processor classifies the force profile as being the air force profile or the liquid force profile based on means of a linear discriminate analysis applied to the force profile.
• any of the steps of the method 360 may be altered, not followed, or additional steps may be added.
• features of the force profile are determined preferably on the basis of force changes versus displacement or position but may also be calculated on the basis of time.
• the features are a characteristic of the profile that is related to the presence of air or other condition that is desired to be known.
• features may include: the scores from an abstract factor analysis, such as principal components analysis (PCA); the peak magnitude of the force profile; the phase shift with respect to time or position of the force profile; the maximum or minimum value of the first derivative with respect to position; the correlation coefficient of the force profile with exemplary profiles representing air and fluid; the distance (e.g., Euclidean or Mahalanobis distance) between the observed profile and a set of template profiles; ratios and/or differences between one or more points or averaged regions in the force profile; the correlation between the force profile and additional sensor readings (e.g., proximal and distal pressure); variance of the force profile from the mean; and a difference of the force profile from the mean.
• PCA principal components analysis
• the features may be viewed as a set of residuals which represent the difference between the force profile or the derivative of the force profile and the expected value.
• the expected value may be determined using adaptive filtering, such as Kalman filtering, or as a moving or exponentially weighted moving average.
• a set of channels are defined which represent the observed force profile at a particular position through time.
• One or more channels are subjected to analysis through time to detect changes in their expected level on the basis of a model, an averaged profile, and/or a problematic network. When either the residual level exceeds a pre-determined threshold or the probability of an air/fluid transition increases beyond a set level, air is indicated in the pumping chamber.
• an alternate embodiment involves a series of channels as describe above. Each channel is separately filtered through time using a moving average, spike rejection filter and/or a low- pass filter. This provides a multiplicity of signals that vary through time. The set of signals is then subject to the derivative based algorithm in which change detection occurs using an event detection and change confirmation method, as described previously. Since each channel provides an indication of the fluid chamber status, a method is employed to combine the indicators and provide one final indicator. The preferred method is to always utilize the channel that provides the reading that is most associated with air. For example, this may comprise the channel that experienced the high derivative and greatest change through time. Alternately, aggregation of the signals can occur using a voting algorithm, fuzzy logic, decision trees, support vector machines or Bayesian networks.
• the multiple channels described above may be subjected to an N-th order Kalman filter and used to generate a residual from an expected value. A change is detected when the residual exceeds a pre-set threshold. In other embodiments, other methods may be utilized.
• Figure 34 illustrates a flowchart of a Bayesian network showing a combination of algorithm sensors and a priori information which may be used to produce an indication of air-in-line or air in a chamber.
• any of the following air devices, tests, or algorithms may be utilized individually or collectively in different numbers or weights to identify air-in-line 380 or air in the chamber 382 to sound an air alarm 384: a recent proximal pressure change
• 61/525,587 The systems, methods, and algorithms/tests of any of the listed incorporated by reference patents may be utilized in conjunction with the systems, methods, and algorithms/tests of the instant disclosure.
• the air indicator or air alarm as described herein may be used to qualify alarms from other sensors and thereby reduce the probability of nuisance alarms.
• One or more systems/methods of the disclosure more accurately detects air in the line of an infusion device than many current systems and methods.
• the systems/methods of the disclosure may be combined with existing systems/methods for detecting air in an infusion system to improve the reliability of air detection systems.
• the disclosure allows for the combination of the output of a force sensor signal with one or more air sensors to improve the reliability of existing air detection systems/methods. In doing so, the disclosed system/method does not require additional hardware modifications but instead leverages the acquired force signal. Additionally, the disclosure does not necessarily require the replacement of existing software modules for air detection but adds an additional safety layer to improve the robustness of existing air detection systems and methods.

Landscapes

• Health & Medical Sciences (AREA)
• Engineering & Computer Science (AREA)
• Life Sciences & Earth Sciences (AREA)
• Animal Behavior & Ethology (AREA)
• Veterinary Medicine (AREA)
• Vascular Medicine (AREA)
• Anesthesiology (AREA)
• Biomedical Technology (AREA)
• Heart & Thoracic Surgery (AREA)
• Hematology (AREA)
• Mechanical Engineering (AREA)
• General Engineering & Computer Science (AREA)
• General Health & Medical Sciences (AREA)
• Public Health (AREA)
• Emergency Medicine (AREA)
• Physics & Mathematics (AREA)
• Fluid Mechanics (AREA)
• General Physics & Mathematics (AREA)
• Infusion, Injection, And Reservoir Apparatuses (AREA)
• External Artificial Organs (AREA)

Priority Applications (6)

Applications Claiming Priority (4)

Publications (1)

Family

ID=49236139

Family Applications (1)

Country Status (7)

Cited By (1)

Families Citing this family (67)

Citations (5)

Family Cites Families (1479)