WO2014052137A1

WO2014052137A1 - Methods and apparatuses for measurement of liquids in motion

Info

Publication number: WO2014052137A1
Application number: PCT/US2013/060475
Authority: WO
Inventors: Perinkulam KASTHURI; Satish Sundar; Krishnan Ramabadran; Daryl James; M.V. Amaresh KUMAR; Lavanya KUMAR
Original assignee: Kasthuri Perinkulam; Satish Sundar
Priority date: 2012-09-25
Filing date: 2013-09-18
Publication date: 2014-04-03

Abstract

An apparatus for accurately determining the flow rate of a fluid includes an illumination source that projects more than two light beams that form flat sheets of light through a conventional transparent drip chamber, and individual optical sensors placed opposite the illumination source that provide timing signals to an external processor corresponding to the times at which each light sheet is first interrupted by a falling drop and at which the falling drop exits each light sheet. The use of more than two light beams allows the optical sensor signals to be advantageously chosen to accurately compute the volume of each drop within the external processor over a wide range of drop sizes. The computed volume can then be used with time measurements also provided within the external processor to compute instantaneous and average flow rates which can be displayed or used for control purposes.

Description

DESCRIPTION

Methods and Apparatuses for Measurement of Liquids in Motion

CROSS-REFERENCE TO RELATED APPLICATIONS

This application Claims Priority to U.S. Provisional Application 61705447. filed on

25 September 2012, titled Methods and Apparatuses for Measurement of Liquids in

Motion, by the same inventors and currently pending.

TECHNICAL FIELD

This invention presents methods and apparatuses for accurately measuring liquid flow rates and the total volume of liquid that flows in the form of drops using optical approaches. The methods presented here are based largely on the accurate measurement of the volume of individual liquid drops. This approach allows for measurement of liquid flows within certain ranges of flow rate with a high degree of accuracy.

BACKGROUND The methods and apparatuses presented here find application in medical applications requiring accurate control and measurement of drug infusion. It is worthwhile noting here that most critical applications in medicine that require accurate control and measurement of flow of drugs require flow rates of less than 250-300 ml/hour. Some of these applications include, but are not limited to:

• Administration of cardiac drugs such as inotropic drugs and chronotropic drugs · Administration of vasodilators to bring down blood pressure and to improve coronary perfusion

Intravenous (IV) sedation

• Fluid replacements

• Anticoagulation drugs

• Intravenous antibiotics · Drugs used in obstetrics such as in those used to make uterus contract in deliveries

Continuous epidural analgesia • Administration of sliding scale infusion

• IV parenteral feeding

If blood can be imaged then infusion of blood products such as plasma and platelets can also be included.

Specific applications of accurate flow rate control include, but are not limited to:

• Administration of digoxin to correct atrial fibrillation in the heart wherein, for example, a clinician might prescribe the use of 1 mg in 60 ml of saline and infuse over one hour.

• To improve kidney function a clinician might prescribe dopamine delivered at 2.5 micrograms per kg body weight per minute. The typical patient weighs approximately 70kg and dopamine ampoules are typically available in lOmg sizes that may be dissolved in 200 ml of saline. Thus 1 ml of the mixture contains 50 micrograms of dopamine and one needs to infuse 3.5 ml per minute, which is 210 ml per hour.

• Administration of adrenaline to the heart may require on the order of 5 micrograms per minute or 300 micrograms per hour. A typical adrenaline ampule contains 1 milligram, or 1000 micrograms and the typical volume of saline for dilution would be 200 ml. Thus one would need to provide an infusion rate of 66 ml/hour.

In all the aforementioned cases, as well as other critical applications not mentioned, overdosing or under dosing of the medication can lead to catastrophic adverse effects. The methods and apparatuses presented here result in high accuracy that can avoid adverse affects for these critical applications. Note that the methods and apparatuses presented here are not restricted to use in medical applications. They may be used in a wide array of other applications requiring accurate measurement of fluid flow. They may also be extended to measurement of the volume and frequency of dispensation of solids such as pellets, coffee beans, rice grains, wheat grains, ball bearings, etc. The various applications that the proposed methods may be applied to are evident to those skilled in the art and the methods and apparatuses presented here are not excluded from those applications - even if not mentioned here - that those skilled in the art may apply them to. Note further that the apparatuses presented here may be used in standalone mode purely for monitoring of flows, or act as accurate feedback devices for pumps such as peristaltic pumps, syringe pumps and the like. The instant invention includes methods and apparatuses for the measurement of fiuid flow in the form of drops. The fluid flow rate (in units of volume/unit time) is simply the product of the drop volume and the drop rate expressed in drops per unit time. These drops typically form at an orifice that is located at the top of a cylindrical chamber, hereinafter referred to as the drip chamber, and fall to the bottom of the drip chamber where they enter tubing that delivers the solution to the target. Drip chambers are commonly used in intravenous delivery systems through which drugs are administered to patients. Drip chambers may of course be used in other applications, including non-medical ones.

Key to the overall accuracy of the flow rate measured in these systems is an accurate measurement of the volume of each drop. Several factors contribute to variations in drop size. These factors include: the diameter of the orifice, the pressure applied to the fluid, and the physical parameters of the fluid itself, such as its density and surface tension. Thus, drops cannot be simply counted to yield an accurate flow measurement, they must be individually measured. Complicating the drop measurement is the variation in the shape of a drop as it forms and during its fall. It has long been appreciated that drops formed in the manner described above experience oscillations in their shape about an equilibrium shape that is spherical - at least until the drop approaches its terminal velocity, where it flattens substantially. Just before detaching from the forming orifice, drops are typically elongated in the vertical direction. During their subsequent fall, the drops oscillate between vertically (prolate) and horizontally (oblate) extended ellipsoids. Although higher order oscillatory modes exist, the predominant effect is a lowest order oscillatory mode which occurs at a frequency related to the density of the liquid, its surface tension and its volume. The vertical drop height, h(t), and its horizontal width, w(t), vary sinusoidally according to:

lt=dO*[l+e l*e-tT*cosco*t+(p] [1]

wt=dO [l-ew e-tT^sios<x) .t+(p] [2]

Where dO is the diameter of the equilibrium sphere, 8_W and 8d are the fractional amplitudes of the oscillations (0≤ew,d≤l), τ is a time constant associated with the decay of the oscillations, ω is the oscillation frequency in radians/sec, and φ is an initial phase angle. The volume of the drop is given by: V(t)=43nh(t)w2(t) [3]

which must be independent of time after the drop falls from the orifice.

Various methods have been previously proposed for measurement of flows in the form of drops, including capacitive and optical methods. For example, US Patent 6,562,012 describes a capacitive method that uses one electrode in contact with the fluid drop at the nozzle tip, and a second electrode surrounding the drop, wherein measuring the capacitance between the two electrodes gives a measure of the size of the drop. However, this method has the major drawback that it requires accurate a priori knowledge of the dielectric properties of the fluid.

Husser & Hugh (2002) describe an optical method wherein a uniformly diffused light is directed on the falling drop through the transparent wall of the drip chamber, while a 2-dimensional image sensor array located across the drip chamber from the light source captures the image of the shadow cast by the drop as it falls. The authors make the assumption that the drop is axially symmetric about the vertical axis and compute the volume based on sum of partial cylindrical volumes. This method is noninvasive (non-contact) and does not require a special disposable element. A few drawbacks of this method include:

a) At the time of capture of the image of the drop, the drop is not necessarily axially symmetric.

There are higher order modes of oscillations of a drop that are perpendicular to the vertical axis.

b) It requires that the transparent drip chamber within which the drop falls does not introduce any optical distortion so that the system can be calibrated using the known optical properties of the external components. However, a typical cylindrical chamber introduces significant optical distortion in the image of the shadow at the image sensor that is located outside the chamber, thereby compromising the accuracy of the drop volume measurement.

c) A tilt in the chamber with respect to the vertical axis causes an error in the measurement, since the assumption of axial symmetry is no longer valid.

d) Since the method uses diffuse lighting, the size of the image depends on the location of the drop between the light source and the image sensor. An additional source and sensor optical system can be placed orthogonal to the first in order detect the position of the drop, correcting errors introduced by variations in the position of the drop (including tilt) but resulting in additional cost, weight and complexity. This also results in the drip chamber being fully enclosed, thereby preventing easy external viewing of the flow through the orifice.

US Patent 5,186,057 describes another optical approach to drop size measurement that measures the time-of -flight of the falling drop as it interrupts two parallel beams of light transmitted through the drip chamber. The intensities of the transmitted beams are independently sensed by separate photodetectors and the signals from the photodetectors are compared to predetermined threshold levels and the resulting signals are used to trigger electronic timers that measure relevant temporal events such as the entry/exit of the drop into/from one of the beams. The method determines the height of the drop and the tilt angle of the drip chamber from vertical from Newton's equations of motion and assumes that the drops have a constant spherical shape wherein the drop diameter is equal to the measured drop height. Although it provides a simple, low cost approach to drop measurement, this method has proven to be highly erroneous in determining drop volume because drops are rarely spherical. As described above, they typically exhibit oscillations between oblate and prolate spheroid shapes during their fall within the drip chamber. Rayleigh's (1892) equation for the lowest order oscillatory mode of a spherical drop is

o2=8apr3 [4]

where ω is the mode frequency in radians per second, σ is the surface tension of the liquid, p is the density of the liquid, and r is the radius of the drop. Evaluation of the oscillation frequency for nominal drop sizes (2 - 5mm) of water shows that, near the orifice, the drop can complete one full oscillatory cycle while it falls less than one drop diameter. The amplitude of the oscillation exhibits a complicated dependence on the shape of the orifice and the properties of the liquid and the force ejecting the drops, but the present inventors have measured variations in drop height exceeding 10% of the nominal value in laboratory tests, corresponding to a 33% variation in the apparent drop volume. Thus the assumption of constant spherical drop shape introduces substantial errors in the calculation of the drop height when using this method. Additionally, since the computed drop diameter must be raised to the third power to compute drop volume, any error in the computed drop diameter is effectively tripled in the volume calculation. Also, the simple threshold triggering of the optically generated electronic signal produces time-based measurement error which is further exaggerated by changes in ambient light and illumination intensity. Three independent time-based measurements are acquired for each drop which are used to solve a set of simultaneous linear equations. Timing errors accumulate in the computation, thereby introducing additional error in the volume calculation.

Thus, there is a need for an apparatus and a method for measuring the volume of liquid drops by a non- contact means that provides the accuracy required for clinical IV drug treatment in a convenient, low cost configuration.

SUMMARY OF THE INVENTION

One embodiment of the present invention employs more than two optical beams in order to allow more accurate measurement of drop height over a wider range of drop sizes, to account for drop oscillations, and to minimize the impact of inevitable system timing errors. The provision of more than two beams allows for selection of a subset of the beams based on the nominal drop size to provide measured drop fall event times that are nearly simultaneous, in order to minimize the variation in drop size that can otherwise occur between the timing events. Additionally, it allows for multiple such samplings of the drop height as the drop falls through the illuminated region, so as to provide a set of height samples that can be numerically fit to the expected drop model in order to achieve higher accuracy in the drop volume calculation. Thus, this embodiment includes at least one light source external to the drip chamber wherein the light passes through a multiplicity of adjacent horizontal slits to form multiple, parallel narrow beams of light that pass through the drip chamber and fall on multiple individual photodetectors located on the other side of the drip chamber. A second set of horizontal slits are placed in front of the photodetectors to provide additional spatial filtering of the light signals to reduce crosstalk between the beams. Cylindrical lenses are advantageously employed between the light source(s) and the drip chamber to form collimated light beams, and in front of the photodetectors to focus the collimated beams onto the active surface of the photodetectors. Nonlimiting examples of light sources include lasers, light emitting diodes, and incandescent lamps. Nonlimiting examples of photodetectors include photodiodes, photoconductors, and one- and two-dimensional solid-state imaging arrays.

Another embodiment of the present invention uses multiple light beams and detectors as described above, but in conjunction with a drop oscillation model which allows for extraction of the parameters of the model and more complete and accurate establishment of drop dimensions. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows a typical apparatus for controlling the flow of a fluid from a reservoir to a target.

Figure 2 is a sectional diagram of a dual beam device. Figure 3 is a timing diagram showing the signals obtained from the device in Fig. 2.

Figure 4 is a diagram of the dual beam flow rate measuring system.

Figure 5 is a perspective illustration of one embodiment of the present invention.

Figure 6 is a sectional diagram of the embodiment shown in Fig. 4.

Figure 7 is a timing diagram showing the signals obtained from the device in Fig. 4. Figure 8 is a sectional diagram of a 6-beam device and small drops.

Figure 9 is a timing diagram showing the signals obtained from the device in Fig. 7.

Figure 10 sectional diagram of a 6-beam device and large drops.

Figure 11 is a timing diagram showing the signals obtained from the device in Fig. 9.

Figure 12 perspective illustration of an embodiment of the present invention using one-dimensional imaging arrays .

DETAILED DESCRIPTION

Referring now to the drawings in detail, in which the reference numerals indicate identical or similar elements among the several views, Figure 1 shows a system 10 for measuring fluid flow rate by measuring drop size in a drip chamber 12. The measurement system comprises an electronic drop sensor 14 mounted to drip chamber 12 and a flow rate computing subsystem 16 that computes the flow rate from measurement of the volume of consecutive drops as provided by signals from the electronic drop sensor 14. Thus, flow rate computing subsystem 16 includes means for establishing and measuring accurate time intervals, digitizing and storing the values of analog signals, processing the digitized analog signals to eliminate external noise and extract pertinent event timing data, and finally computing and displaying the computed flow rate or otherwise outputting signals in formats appropriate to control external instruments. In another embodiment, preprocessing of the image sensor data is done by first subtracting/differencing the background from the field of view. This differencing allows the algorithms to omit any kind of splash of fluid on to the walls of the drip chamber.

Upstream of the drip chamber 12 is the reservoir 18 positioned to feed fluid to the drip chamber through tubing 20. The fluid flows into the drip chamber 12 where an orifice 22 at the upstream end forms drops 24 which fall along a path enclosed by the transparent enclosure of the drip chamber, collecting at the downstream end. Downstream of the drip chamber 12 the fluid is directed through tubing 26 to a valve (not shown) used to establish the desired flow rate, and subsequently to the target.

Figure 2 illustrates the operation of a two-beam optical drop sensor 14. Illumination sources 36 and 38 form beams of light, 40 and 42, respectively, that pass through the transparent enclosure of the drip chamber 12 and fall on optical sensor devices 32 and 34, respectively. Nonlimiting examples of illumination sources include lasers, light emitting diodes, and incandescent lamps. Nonlimiting examples of optical sensor devices include photodetectors including photodiodes, photoconductors, and one- and two-dimensional solid-state imaging arrays. The light beams 40 and 42 have a nominal vertical dimension A and the edges of the adjacent beams are separated by a dimension B. As individual drops24 fall from orifice 22 they interrupt the light beams 40 and 42, thereby inducing changes in the electrical response of the sensor devices 32 and 34. The output signals from sensor devices 32 and 34 are individually processed to form timing signals that are applied to an external flow rate computing subsystem (not shown.)

Figure 3 shows example timing signals derived from the output signals of sensor devices 32 and 34. Falling drop 24 first interrupts beam 40 at initial time TO, shading part of sensor device 32 and reducing the amplitude of its electronic signal. At time Tl, the drop 24 fully emerges from beam 40 and it enters the unilluminated region between the beams so that there is no impact on either sensor signal. The drop 24 then interrupts beam 42 at time T2, subsequently emerging at time T3. At time TO the drop has height hO and velocity V0. At time Tl, the height of the drop is taken to be hi and center of mass of the drop has traveled a distance A+12 0+12 land exhibits a velocity given by

Vl=V0+g(Tl-T0). At time T2 the height of the drop is h2 and the center of mass of the drop has traveled a distance A+B+12h0-12h2 since TO. The velocity of the drop is V2=Vl+g(T2-Tl ). Finally, at time T3 the height of the drop is h3 and the center of mass of the drop has traveled a distance 2A+B+12h0+12h3 since T2. The final velocity is V3=V2+g(T3-T2). Letting tl=(Tl -TO), t2=(T2-T0) and t3=(T3-T0), the equations of motion for the falling drop that relate drop height values, the dimensions of the beams of light, the measured time intervals, and known physical constants can then be summarized as:

A+12h0+12hl=V0 tl+g2tl2 [5] A+B+12h0-12h2=V0 t2+g2t22 [6] 2A+B+12h0+12h3=V0 t3+g2t32 [7]

Thus, there are three equations in the five unknown quantities: hO, hi, h2, h3 and V0. This system of equations is underspecified, and so cannot be solved for all of the unknown quantities. In the prior art, variations in drop height were ignored, so the equations of motion simplify to

A+h=V0 tl+g2tl2 [8]

A+B=V0 t2+g2t22 [9]

2A+B+h=V0 t3+g2t32 [10] from which the unknowns, h and V0, are easily solved to give

V0=g2t32-tl2-t22tl+t2-t3 [11] k=B+g22tl2t2-tl2t3-2tlt22+tlt32+t22t3-t2t32tl+t2-t3 [12]

In fact, this system of equations is overspecified. However, if the drip chamber is tilted by an angle a with respect to the vertical axis, then all elements of the path length of the falling drop are extended by a factor of l/cos(a). The solutions for h and V0 can be substituted into equation [10] resulting in an equation for A+B in terms of the time quantities. This solution then gives cosa=A+Bg2(tl +t2-t3 )tlt22-tl2t2-t22t3+t2t32 [13] which can be used to correct the value of h for tilt. However, variations in drop height cannot be ignored if high accuracy is to be achieved. Equations [8] - [10] can be rewritten as:

A+h+8h01=V0 tl+g2tl2 [14]

A+B+8h02=V0 t2+g2t22 [15]

2A+B+h+8h03=V0 t3+g2t32 [16]

Where 8h01=h02+hl2-h, 8h02=h02-h22 and 8h03=h02+k32-h. Then the solution for h becomes: h=h0-(t2-t3)8h01 Hi (8h02-8h03)tl +t2-t3 [17] where ho is the solution obtained assuming constant drop height. This illustrates how variations in drop height propagate to become significant errors in the calculated quantities. Since, as can be seen from Figure 3, the quantity (t2-t3) is approximately the same as tl, the value of h is approximately: hxh0-8h01+8h02-8h032 [18] and the drop height variations lead directly to proportionate errors in the calculated value of h.

The accuracy of the multi-beam optical approach derives from focusing on measurements involving events where the drop size is nearly constant. For example, in Figure 3 the period from Tl to T2 is the shortest measurement period in the system, resulting when the incident drop falls out of the top beam, crosses the nonilluminated region and enters the bottom beam. As can be appreciated from the diagram in Figure 2, it can be expected that the dimensions of the drop do not change appreciably in this brief period. Combining equations [5] and [6] gives:

B-12hl+h2=V0t2-tl +_g2(t22-tl2) [ 19] and since tl and t2 are nearly time coincident, it can be assumed that the drop heights hi and h2 are nearly equal, hl=h2=h. In fact, if the height of the drop equals the width of the nonilluminated region, this time period vanishes and the drop height is precisely established. Thus, the height of the drop when it is located in the middle of the nonilluminated region may be more accurately estimated as h=B-V0t2-tl-g2(t22-tl2) [20]

However, a value of V0 is still required to complete the calculation of drop height. Equation [6] can be rewritten as

V0=lt2A+B+Ahl2-g2t2 [21]

where Ahl=h0-h2 [22]

Substituting equation [21] into equation [20] yields: h=Btlt2-t2-tlt2(A+g2tlt2+Ahl2) [23]

and it can be seen that the effect of the error term Ahl in the second quantity on the right hand side of equation [23] is diminished as t2 approaches tl . Thus, each pair of slits in a multi-slit system provides one potentially accurate estimate of the drop height. Increasing the number of slits in the device leads to two capabilities that improve drop volume accuracy. First, multiple time samples of drop height allow a description of the time dependence of drop shape that can be fit to numerical models of drop shape as in equations [1] and [2] in order to more accurately establish the parameters of the drop and consequently its volume. Second, for a fixed slit height, the signals from the optical detectors at pairs of slits can be advantageously selected to provide timing events corresponding to tl and t2 in the above argument wherein the corresponding intervals (and errors) are minimized. These manipulations of optical sensor data are executed within the external flow rate computing subsystem 16.

Figure 4 shows a diagram of a dual beam fluid flow rate measuring system 10. Illumination sources 36 and 38 derive operating power from an output port of a microcontroller or microprocessor 168 and direct light beams through drip chamber 12 onto optical sensor devices 32 and 34. The output signals from sensors 32 and 34 are individually amplified by amplifiers 162 within flow rate computing subsystem 16 and are applied to the inputs of an analog multiplexer 164 which sequentially directs the conditioned sensor output signals to the input of an analog-to-digital converter 166. The digital signals representing the digitized sensor output signals are applied to an input port on microcontroller or microprocessor 168 which derives precise timing signals from a frequency standard such as a quartz crystal device 169. Firmware executed within microcontroller or microprocessor 168 processes the digitized sensor output signals based on precise timing signals from frequency standard 169 to establish digital records of the time response of each sensor, such as shown in Figure 3, and to extract the timing intervals such as tl, t2 and t3 in Figure 3 from which the fluid flow rate is established based on equations presented above. Microcontroller or microprocessor 168 is connected to an input device 172 such as a keypad which allows for entry of system constants and administrative data, and an output device 174 such as a liquid crystal display to display the measured fluid flow rate and other parameters of interest. A separate digital output port on the microcontroller or microprocessor 168 is used to control external devices, such as a pump 176 that provides pressurized fluid to the drip chamber 12.

Figure 5 shows a perspective view of one embodiment of the present invention which includes four light beams. The elements in Figure 54 are spaced more widely apart than might be employed in practice for clarity. In one embodiment, individual illumination sources 41 - 44 are placed on one side of the drip chamber 12 and optical sensors 61 - 64 having identical spacing to the illumination sources are located on the opposite side of the drip chamber 12. Nonlimiting examples of illumination sources include lasers, light emitting diodes, and incandescent lamps. Nonlimiting examples of optical sensors include photodetectors including photodiodes, photoconductors, and one- and two-dimensional solid- state imaging arrays.

A cylindrical lens 51 is located between the illumination sources 41 - 44 in order to collimate the light emitted from the illumination sources 41 - 44 and the collimated light is then passed through an optical mask 52 having slits with specified heights and center-to-center spacing corresponding to the spacing between the illumination sources 41 - 44 to form light beams in the shape of narrow sheets that penetrate the drip chamber 12. Similarly, a corresponding optical mask 53 and cylindrical lens 54 is employed between the optical sensors 61 - 64 and the drip chamber 12 in order to provide a degree of spatial filtering of the incident light beams and focus the captured light onto the active area of each of the sensors. The spatial filtering prevents light that is diffracted by the falling drop 24 from entering adjacent optical sensors, thereby reducing the optical crosstalk between the sensors. Figure 6 shows a cross sectional view of the four-beam device illustrated in Figure 5. It more clearly shows the individual light beams 71 - 74 penetrating the transparent enclosure of the drip chamber 12 and the relative orientation of the illumination sources 41 - 44, the lenses 51 and 54, the optical masks 53 and 53, and the optical sensors 61 - 64. As drops of fluid 24 are formed at the orifice22 and fall through the drip chamber, they sequentially interrupt the well defined light beams 71 - 74 and introduce variations in the signals from the individual optical sensors 61 - 64 which are processed to form timing signals that are applied to an external flow rate computing subsystem (not shown.)

Figure 7 shows example timing signals derived from the output signals of sensor devices 61 - 64. Reference time t=0 is taken to be established by the signal detected when falling drop 24 first interrupts the top beam 71. The remaining time events are developed in the same manner as described for the timing diagram of Figure 3. Taking the height of each optical mask aperture to be A and the spacing between apertures to be B, a set of equations similar to equations [5]-[7] can be developed for this device configuration which provides seven equations in nine unknowns, which is again an underspecified system of equations. In fact, adding each new optical beam to the system provides three new equations, but introduces four new unknowns. However, the additional pairs of beams now provide three accurate time samples of drop height from the intervals t2-tl, t4-t3 and t6-t5: hl=Btlt2-t2-tlt2(A+g2tlt2+Ahl2) [24] h2=Bt3t4-t4-t3t4A+g2t3t4+Ah22 [25] h3=Bt5t6-t6-t5t6(A+g2t5t6+Ah32) [26]

Note that the quantities t2-tl, t4-t3 and t6-t5 are negative in this case, indicating that the drop height is larger than the dimension B. These three values of drop height can be numerically fit to a drop model such as presented in equations [1] and [2] to establish an accurate estimate of the drop volume using equation [3]. The drop rate is conveniently measured by comparing successive absolute time values for one of the timing events, such as the reference t=0 event. The instantaneous fluid flow rate is then determined by multiplying the drop volume by the drop rate. These calculations are executed within the flow rate computing subsystem 16. Figure 8 shows a cross sectional view of a six-beam embodiment. Six individual light beams 71 - 76 now penetrate the transparent enclosure of the drip chamber 12 and the relative orientation of the illumination sources 41 - 46, the lenses 51 and 54, the optical masks 53 and 53, and the optical sensors 61 - 66 are the same as shown in Figure 5. As drops of fluid 24 are formed at the orifice22 and fall through the drip chamber, they sequentially interrupt the well defined light beams 71 - 76 and introduce variations in the signals from the individual optical sensors 61 - 66 which are processed to form timing signals that are applied to an external flow rate computing subsystem (not shown.) The drop height and the height of the nonilluminated regions shown in Figure 8 are shown to be approximately equal to the height of a light beam.

Figure 9 shows example timing signals derived from the output signals of sensor devices 61 - 66. Reference time t=0 is taken to be established by the signal detected when falling drop 24 first interrupts the top beam 71. The remaining time events are developed in the same manner as described for the timing diagram of Figure 3. A set of equations similar to equations [24]-[26] can be developed for this device configuration which provides the potential for five accurate time samples of drop height from the intervals t2-tl, t4-t3, t6-t5, t8-t7, and tl0-t9. Since the drops are the same size as the nonilluminated regions, the time responses don't overlap. This is the ideal case wherein the intervals t2-tl, t4-t3, t6-t5, t8-t7, and tl0-t9 all nearly vanish (except for variations caused by variations in drop height) and any of the intervals can be chosen to provide accurate time samples.

Figure 10 shows another cross sectional view of the six-beam device, but now the drop 24 is much larger, approximately six times the height of a light beam. Figure 11 shows the corresponding timing signals derived from the output signals of sensor devices 61-66, and it is immediately recognized that the timing signals overlap extensively because of the size of the drops. However, in this case the intervals tl-t6, t3-t8 and t5-tl0 can be used to develop three equations for drop height: hi =(2A+3B)tl t6-t6-tl t6(A+g2tl t6+A hl2) [27] tl2=(2A+3B)t3t8-t8-t3t8A+g2t3t8+Ah22 [28] h3=(2A+3B)t5tl 0-tl 0-t5tl 0(A+g2t5tl 0+Δ h32) [29] whereas if there had been only four beams only one equation would have obtained.

Consideration of Figures 8-11 leads to a set of design parameters for the optical elements of the fluid flow rate measuring system 10. Given a specified range of drop heights, hmin≤h≤hmax, a minimum value of the beam height dimension, A, as established by available and/or cost effective optical elements and a desired minimum number of accurate drop height measurements, N, for fitting to the drop model equations such as equations [1] and [2]; then the beam edge spacing dimension, B, is given by:

B=A+2hmin [30] and the number of independent optical beams,Nbeams, is given by:

Nbeams≥hmax-hminA+B+(N-l) [31]

where it is assumed that dimensions A and B are both uniform throughout the optical array.

Figure 12 shows a perspective view of another embodiment of the present invention which includes four light beams. The elements in Figure 12 are spaced more widely apart than might be employed in practice for clarity. In one embodiment, individual illumination sources 41 - 44 are placed on one side of the drip chamber 12 and optical sensors 61 - 64 having identical spacing to the illumination sources are located on the opposite side of the drip chamber 12. A cylindrical lens 51 is located between the illumination sources 41 - 44 in order to collimate the light emitted from the illumination sources 41 - 44 and the collimated light is then passed through an optical mask 52 having slits with specified heights and center-to-center spacing corresponding to the spacing between the illumination sources 41 - 44 to form light beams in the shape of narrow sheets that penetrate the drip chamber 12. Similarly, a corresponding optical mask 53 and cylindrical lens 54 is employed between the optical sensors 61 - 64 and the drip chamber 12 in order to provide a degree of spatial filtering of the incident light beams and focus the captured light onto the active area of each of the sensors. The spatial filtering prevents light that is diffracted by the falling drop 24 from entering adjacent optical sensors, thereby reducing the optical crosstalk between the sensors.

In this embodiment the optical sensors 61-64 are solid-state linear imager arrays and the sensors are repositioned along the optical axis to focus the horizontal extent of the spatially filtered scene on the active imaging regions of the devices. Rather than measuring a continuous scalar signal from the optical sensors used in Figure 5, each of the imager arrays can output a sampled vector signal consisting of analog or digital values corresponding to the intensity of light incident on each pixel. Thus, each linear imager array captures a slice of the shadow of the drop as it falls through the drip chamber 12 and a separate measurement of the width of each drop 24 can be acquired corresponding to each snapshot of the drop so acquired. The drop width values so obtained from each linear imager array can be combined with drop height values obtained using the previously described method, but employing a statistical metric, such as the mean of the values of all of the pixels, as the (sampled) scalar signal. In this way, both the height and the width of the drop can be accurately established at several points along the trajectory of each drop, thereby allowing a more accurate calculation of drop volume.

Summary

A method and an apparatus for accurately measuring drop volume of a fluid within a drip chamber involves deriving timing signals from optical sensors that sense more than two optical beams projected through the transparent housing of the drip chamber in order to allow more accurate measurement of drop height over a wider range of drop sizes, to account for drop oscillations, and to minimize the impact of inevitable system timing errors. The provision of more than two beams allows for selection of a subset of the beams based on the nominal drop size to provide measured drop fall event times that are nearly simultaneous, in order to minimize the variation in drop size that can otherwise occur between the timing events. Additionally, it allows for multiple such samplings of the drop height as the drop falls through the illuminated region, so as to provide a set of height samples that can be numerically fit to the a fluid drop model in order to achieve higher accuracy in the drop volume calculation.

Claims

What is claimed is:

1. A method for computing a volume of a falling drop of liquid comprising: a) generating at least N beams of light having predetermined shapes, vertical and horizontal dimensions, and center-to-center spacings that are directed across a path of, and perpendicular to a vertical axis of motion of a falling drop of liquid, wherein N is an integer greater than two; b) measuring a time of each event at which the falling drop first intercepts each beam of light; c) measuring a time of each event at which each drop fully exits each beam of light; d) calculating time intervals between said events; e) selecting a set of M pairs of events having the smallest intervals, wherein M is an integer

smaller than N; f) calculating a set of M drop height values using mathematical relationships between the shapes, dimensions and spacings of the beams of light, the set of M pairs of events and their respective time intervals, and known physical constants; and g) calculating the volume of the drop based on the values of and relationships among the M drop height values.

2. The method of claim 1 further comprising determining said number, N, and said shapes, vertical and horizontal dimensions, and center-to-center spacings of said at least N beams from a predetermined range of dimensions for said falling drop of liquid.

3. The method of claim 1 further comprising calculating a fluid flow rate using the calculated drop volume and the measured event times.

4. The method of claim 1 wherein the calculating steps are performed using a microcontroller or a microprocessor.

5. The method of claim 1 further comprising displaying the calculated volume.

6. The method of claim 1 further comprising generating electronic output signals to control external devices based on the calculated volume.

7. An apparatus for determining a fluid flow rate in a fluid drop forming system comprising: a) light beam generation devices for directing at least N light beams across a path of a falling fluid drop with each light beam having a predetermined shape, vertical and horizontal dimensions, and center-to-center spacing from other beams, wherein N is an integer greater than two; b) light beam detection devices for receiving the N light beams and for generating an independent signal for each beam in response to the fluid drop entering or exiting from each beam; and c) a flow rate computing system for: i) receiving signals from the light beam detection devices, ii) establishing a time measure for each event at which each drop first intercepts and

subsequently just exits each light beam, iii) calculating time intervals between said events and selecting a set of M pairs of events

having the shortest intervals, wherein M is an integer smaller than N, iv) calculating a set of M drop height values using mathematical relationships between the shapes, dimensions and spacings of the beams of light, the set of M pairs of events and their respective time intervals, and known physical constants, v) calculating a volume of the drop based on the values of and relationships among the M drop height values, and vi) calculating a fluid flow rate using the calculated drop volume and the measured event times.

8. The method of claim 7 further comprising determining said number, N, and said shapes, vertical and horizontal dimensions, and center-to-center spacings of said at least N beams from a predetermined range of dimensions for said falling drop of liquid.

9. The apparatus of claim 7 further comprising a display unit for displaying volume and flow rate from the flow rate computing system.

10. The apparatus of claim 7 further comprising a control signal developed by the flow rate computing system for controlling an external device based on the calculated volume and flow rate.

11. An apparatus for determining a fluid flow rate in a fluid drop forming system comprising: a) light beam generation means for directing at least N light beams across a path of a falling fluid drop with each light beam having a predetermined shape, vertical and horizontal dimensions, and center-to-center spacing from other beams, wherein N is an integer greater than two; b) light beam detection means for receiving the N light beams and for generating an independent signal for each beam in response to the fluid drop entering or exiting from each beam; and c) processing means for: i) receiving signals from the light beam detection means, ii) establishing a time measure for each event at which each drop first intercepts and

The method of claim 11 further comprising determining said number, N, and said shapes, vertical and horizontal dimensions, and center-to-center spacings of said at least N beams from a predetermined range of dimensions for said falling drop of liquid.

The apparatus of claim 11 further comprising display means for displaying volume and flow rate from the processing means.

The apparatus of claim 11 further comprising a control means for controlling an external device based on the calculated volume and flow rate.

In an IV administration system having a fluid source fluidly connected to a drop forming orifice adapted to direct fluid drops from the fluid source to fall through a drip chamber connected to an IV tube connected to a patient, a system for measuring fluid flow comprising: a) light beam generation means for directing at least N light beams across a path of a falling fluid drop with each light beam having a predetermined shape, vertical and horizontal dimensions, and center-to-center spacing from other beams, wherein N is an integer greater than two; b) light beam detection means for receiving the N light beams and for generating an independent signal for each beam in response to the fluid drop entering or exiting from each beam; and c) processing means for: i) receiving signals from the light beam detection means, ii) establishing a time measure for each event at which each drop first intercepts and

16. The method of claim 15 further comprising determining said number, N, and said shapes, vertical and horizontal dimensions, and center-to-center spacings of said at least N beams from a predetermined range of dimensions for said falling drop of liquid.

17. The system as in claim 15 further comprising a holding means for mounting the light beam

generation means and light beam detection means to the drip chamber.