JP2011062371A - Drip-feed detector, infusion pump and method for controlling the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a low-cost drip-feed detector that precisely detects the drip flow rate, an infusion pump using this detector, and a method for controlling these devices. <P>SOLUTION: The drip-feed detector includes a transparent drip-feed tube, a light emitting part arranged on an external side of the drip-feed tube, and a two-dimensional image sensor located beyond the drip-feed tube and opposite the light emitting part. The image sensor is set up to cover at least a drip nozzle end located within the drip-feed tube and the prescribed distance of drip falling from the drip nozzle. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an infusion detection device, an infusion pump provided with the infusion detection device, and a control method for these devices to perform accurate fluid delivery.

Conventionally, infusion devices connected to infusion pumps, etc., when performing precise infusions, droplets falling inside the infusion cylinder are detected by infrared light, etc., and the number of enemies of the dropped droplets is counted. A technique has been used that includes an infusion detection device that calculates the amount and speed of liquid ejection under the assumption that the size of the liquid droplet is constant, and controls the infusion pump by feedback control.
Furthermore, a method has been considered for measuring the size of a falling droplet using a one-dimensional line sensor or a two-dimensional image sensor to obtain a more accurate liquid feeding amount and liquid feeding speed.

For example, according to the technique of Patent Document 1, the width of a cross section of a drop falling at a certain time using a one-dimensional line sensor is calculated with respect to the gravity drop direction, and the sectional quadrature method is calculated from the change in the cross section width of the drop. A method for determining the volume of a drop that falls using a slab is disclosed.
Further, the technique of Patent Document 2 discloses a method of photographing a flying droplet a plurality of times and calculating the velocity and volume of the moving droplet from the relationship between the distance moved within the unit time and the projected area. Has been.

  In the technique of Patent Document 3, the volume of the droplet attached to the nozzle before the droplet is discharged from the nozzle and the volume of the droplet remaining on the nozzle after the droplet is discharged from the nozzle are photographed. A method of calculating the volume of a droplet ejected from a nozzle by subtracting the volume of the remaining droplet from the volume before ejection is disclosed.

JP-A-8-229119 JP 2000-146993 A JP 2001-327905 A

However, when trying to detect a droplet in a drip tube, the inventors have found that the following characteristics of the droplet are problematic.
As in the case of the method of sequentially scanning with a one-dimensional line sensor as in Patent Document 1, the liquid droplet falling in the drip tube does not maintain a fixed shape, and the falling shape is shown in FIG. 11a. As shown in Fig. 4, it is confirmed that the camera is falling while changing its shape.
Due to the influence of the deformation during the fall, it is impossible to increase the accuracy of the volume to be calculated by the method using the one-dimensional line sensor. In addition, the number of droplets falling at one time is not limited to one, and the droplets may fall into two large and small droplets, which also causes a large error in detecting the liquid amount (FIG. 11b).

  Further, as in Patent Document 2, in the case of a method of photographing a flight or falling using an image sensor and calculating a volume, it is difficult to completely catch a droplet passing through the image sensor. FIG. 12 shows how the shape of the falling liquid droplet changes with time. A liquid droplet that freely falls over time falls in the air while being accelerated by gravitational acceleration with time. For this reason, the moving speed in the middle of dropping is high, and it is very difficult to accurately photograph the way in which the droplet is falling.

That is, in order to completely photograph a droplet, it is not only necessary to use an image sensor capable of photographing at a high speed, but it is also necessary to detect in advance the timing at which the droplet starts to fall. A separate sensor to detect that the droplet has fallen above the image sensor's shooting range (such as triggering the start of shooting by the high-precision sensor to know that the droplet has started dropping) Must be used, which leads to an increase in parts.
Also, as disclosed in Patent Document 3, in the method of calculating the volume from the images before and after being ejected from the nozzle, it is not clear when the droplet falls, so even in an image that does not originally need to calculate the volume, The load for performing image processing increases.

  Therefore, in view of the above problems, the present invention captures an image immediately after the droplet starts to fall in the drip tube after leaving the nozzle, and calculates the volume from the captured droplet shape in advance. A drip detection device capable of obtaining a highly accurate flow rate while suppressing the price of the processor used by reducing the load for performing image processing without requiring a sensor for detecting the falling of the liquid, and an infusion using the same It is an object of the present invention to provide a pump and a method for controlling these devices.

In order to solve the above problems, the present invention is arranged in a transparent drip tube, a light emitting unit arranged on one side outside the drip tube, and a position facing the light emitting unit and the drip tube. A two-dimensional image sensor, and the image sensor is a drip detection device set to include at least a tip of a drip nozzle in the drip tube and a predetermined drop distance of a drop falling from the drip nozzle. It is characterized by being.
According to the above configuration, the two-dimensional image sensor can capture an image of the drop nozzle tip and a predetermined drop distance of the droplet falling from the drop nozzle in the field of view, so that the shape as close to a sphere as possible can be grasped before it is greatly deformed. In addition, even when the droplet is divided into a plurality of portions, the accurate volume of the droplet can be obtained, so that it is possible to accurately measure the flow rate that is the sum of the droplets during the drip.

In order to solve the above problems, the present invention includes a main body, driving means such as a liquid feeding motor accommodated in the main body, various sensors for infusion operation, and the driving means and the sensor. A controller for connecting, an infusion tube extending from an external infusion device, connected to the main body, and an infusion tube extending into the main body for receiving and holding the liquid in the infusion tube The drip detection device is disposed at a position facing the transparent drip tube, a light emitting unit disposed on one side outside the drip tube, and the light emitting unit sandwiching the drip tube. A two-dimensional image sensor, and the image sensor is an infusion pump that is set to include at least a tip of a dropping nozzle in the drip tube and a predetermined drop distance of the droplet falling from the dropping nozzle. It is characterized by being That.
According to the above configuration, the two-dimensional image sensor can capture an image of the drop nozzle tip and a predetermined drop distance of the droplet falling from the drop nozzle in the field of view, so that the shape as close to a sphere as possible can be grasped before it is greatly deformed. In addition, even when the droplet is divided into a plurality of portions, the accurate volume of the droplet can be obtained, so that it is possible to accurately measure the flow rate that is the sum of the droplets during the drip. For this reason, by connecting the infusion detection device to the infusion pump, the infusion pump can set an accurate flow rate based on the measured droplet volume to perform infusion.

Preferably, in the present invention, the infusion detection device and / or the infusion pump connected to the infusion detection device includes an imaging control unit and a recording unit that stores imaging data, and the imaging control unit includes: The two-dimensional image sensor is controlled to record a state in which a droplet is separated from the tip of the dropping nozzle and falls as a droplet as a series of moving images or shooting data that is a plurality of screen data. An image of a liquid droplet immediately after being dropped from the dropping nozzle of the chemical liquid is determined by determining that a liquid droplet has dropped from the dropping nozzle based on the imaging data. It is characterized in that the data is specified and the volume of the droplet of the image data is calculated.
According to the above-described configuration, it is possible to accurately grasp the droplet that the chemical solution falls off the dropping nozzle, and to obtain the accurate volume of the droplet from the droplet having the shape that is closest to the ideal.

Preferably, according to the present invention, an image immediately after detecting a constriction that occurs when the chemical liquid becomes a liquid droplet is used as a means for determining that the liquid liquid drops from the dropping nozzle. The droplet falling state is characterized in that the volume of the droplet is detected based on an image of the droplet falling state.
According to the above-described configuration, it is possible to accurately grasp the droplet that the chemical solution falls off the dropping nozzle, and to obtain the accurate volume of the droplet from the droplet having the shape that is closest to the ideal.

Preferably, according to the present invention, the chemical liquid held at the tip of the dropping nozzle becomes a liquid droplet within a predetermined time as a means for determining that the liquid droplet has fallen off the dropping nozzle. The determination of the possibility of falling is determined based on the data held in the record, and the volume of the droplet is detected based on the image of the droplet after the lapse of time.
According to the above-described configuration, it is possible to accurately grasp the droplet that the chemical solution falls off the dropping nozzle, and to obtain the accurate volume of the droplet from the droplet having the shape that is closest to the ideal. Furthermore, it is possible to reduce the number of times unnecessary image data is acquired, and to reduce the load for image processing, thereby reducing the price of the processor used and obtaining a highly accurate flow rate.

  In order to solve the above problem, the present invention provides a transparent drip tube, a light emitting unit disposed on one side outside the drip tube, and a position facing the light emitting unit and the drip tube. A two-dimensional image sensor, wherein the image sensor includes at least a tip of a dropping nozzle in the drip cylinder and a device set to include a predetermined drop distance of a droplet falling from the dropping nozzle. The volume of the liquid droplet is calculated in order to calculate and control the liquid feeding amount, and the two-dimensional image sensor is controlled by the imaging control unit so that the liquid droplet is discharged from the tip of the dropping nozzle. The state until it falls and drops and drops is photographed as a series of moving images or photographing data that is a plurality of screen data, and is stored in the recording unit. Fall away And determining the image data of the liquid droplet immediately after being dropped from the dropping nozzle of the chemical liquid based on the determination, and using the liquid droplet of the image data, A method for controlling the apparatus, wherein the volume of a droplet is calculated.

  Preferably, according to the present invention, as a method for determining that a liquid droplet has fallen off the dropping nozzle, the image immediately after detecting a constriction generated when the chemical liquid becomes a liquid droplet is used. As a falling state, the volume of a droplet is detected based on an image of the droplet falling state.

  Preferably, according to the present invention, as a method for determining that a liquid droplet has dropped from the dropping nozzle, the chemical liquid held at the tip of the dropping nozzle becomes a liquid droplet within a predetermined time. It is determined based on the dropped data, and after detecting the droplets accumulated at the tip of the dropping nozzle, the volume of the droplet is detected based on the image of the droplet after the lapse of time.

  The present invention detects a drop of a droplet in advance by capturing an image immediately after the droplet leaves the nozzle and starts dropping in the drip tube and calculating the volume from the shape of the captured droplet. A drip detection device capable of obtaining a highly accurate flow rate by reducing the price of the processor used by reducing the load for performing image processing, and an infusion pump using the same. A control method can be provided.

1 is a schematic perspective view of an infusion pump according to an embodiment of the present invention. The block block diagram of the infusion pump of FIG. The schematic block diagram of the drip detection apparatus concerning embodiment of this invention. Explanatory drawing which shows a mode that a droplet is dripped from the dripping nozzle of the drip detection apparatus of FIG. Explanatory drawing of the method of calculating | requiring the volume of the droplet dripped from the dripping nozzle of the drip detection apparatus of FIG. Explanatory drawing of the method of calculating | requiring the volume of the droplet dripped from the dripping nozzle of the drip detection apparatus of FIG. Explanatory drawing of the method of calculating | requiring the volume of the droplet dripped from the dripping nozzle of the drip detection apparatus of FIG. The flowchart which shows an example in the case of calculating | requiring the volume of the droplet dripped from the dripping nozzle of the drip detection apparatus of FIG. The flowchart which shows an example in the case of calculating | requiring the volume of the droplet dripped from the dripping nozzle of the drip detection apparatus of FIG. The flowchart which shows an example in the case of calculating | requiring the volume of the droplet dripped from the dripping nozzle of the drip detection apparatus of FIG. The figure explaining the form of the droplet which falls from the dripping nozzle of a drip detection apparatus. The figure which shows the form of the droplet which falls from the dripping nozzle of a drip detection apparatus with time.

Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings.
The embodiments described below are preferable specific examples of the present invention, and thus various technically preferable limitations are given. However, the scope of the present invention is particularly limited in the following description. Unless otherwise stated, the present invention is not limited to these embodiments.

FIG. 1 is a schematic perspective view of an infusion pump according to an embodiment of the present invention.
As shown in the figure, the infusion pump 11 has a box-shaped hollow main body 15 in which a driving means such as a motor, various sensors for infusion operation, a control device, and the like are accommodated.
That is, the front surface portion of the main body 15 in FIG. 1 is a door 12, and a hinge (not shown) is provided on the left side edge portion so as to be opened and closed as indicated by an arrow.
In this infusion pump 11, an infusion tube 16 suspended from an infusion bag (not shown) via an infusion detection device 40 described later is sandwiched between the door 12 and the main body 15, and the infusion tube 16 is attached to the main body 15. It is housed and held inside. In the main body 15 of the infusion pump 11, the outer peripheral surface of the infusion tube 16 set in a fixed position in the main body 15 is provided with an inner surface of the door 12 by a plurality of fingers individually driven by a pump mechanism built in the main body 15. The peristaltic infusion pump performs liquid feeding by sequentially pressing the outer peripheral surface of the infusion tube sandwiched between the two along the length direction.

  An operation panel 13 is provided on the front surface of the infusion pump 11 constituting the door 12, and the operation panel 13 is provided with a switch button 13 a as a plurality of or many operators. A display unit 14 is formed above the operation panel 13. The display unit 14 can display numbers and images necessary for display by forming a segment display of LED (Light Emitting Diode) (light emitting diode) or a liquid crystal display device, for example. The area of the display unit 14 can be made larger than that shown in the figure based on the variety of display contents. An operation indicator 17 is disposed at the upper end of the infusion pump 11.

Referring to FIG. 2, the infusion pump 11 has a control unit 20 composed of a CPU (Central Processing Unit) for controlling the entire operation, an infusion probe 21 connected to the infusion tube 16 of FIG. Measuring device 22, door portion detection circuit 23 that detects whether door 12 is opened or closed, blockage detection circuit 24 that detects whether blockage of infusion tube 16 is present, or presence of bubbles (air) of a predetermined length or more in infusion tube 16 A bubble detection circuit 25 having a bubble sensor for detecting presence / absence, a setting / input unit 26 for setting and inputting an infusion flow rate, an infusion rate, and the like, a start switch 27 for instructing the start of infusion, and a power switch 28 are provided. Yes.
Further, the infusion pump 11 includes an administration unit (administration unit), an administration drive unit 30 including a motor included in the administration unit and a motor drive circuit for driving the motor.

  A storage unit 29 is connected to the control unit 20. The storage unit 29 includes a RAM (Random Access Memory), a ROM (Read Only Memory), a hard disk, a flash memory, and the like, and these are connected together with other devices via a bus. The CPU that is the control unit 20 described above performs processing of a predetermined program and controls a ROM and the like connected to the bus. The ROM stores various programs and various information. The RAM has a function as an area for comparing the contents of the memory during the program processing and for executing the program. And the memory | storage part 29 can also comprise a part of infusion apparatus mentioned later.

Further, the infusion pump 11 serves as an alarm unit 32 or an alerting unit, for example, for driving a buzzer driving circuit for driving a buzzer or the like, for example, a display unit 31 formed by a liquid crystal display device or an LED or the like. It has an operation display unit drive circuit (not shown) for driving the display unit drive circuit and the operation indicator, and an inspection device (not shown) which is an external device of the infusion pump 11 is connected via an external terminal or a nurse center. And an external communication unit 33 for transmitting and receiving data to and from a host computer at a site provided outside. A plurality of sets of external communication units and external terminals may be provided so that a plurality of devices can be connected simultaneously. The external connection terminal preferably includes a USB (Universal Serial Bus) terminal. This facilitates connection with an external terminal such as a personal computer or a communication modem.
Furthermore, the infusion pump 11 has a DC / AC power supply 34. Specifically, the infusion pump 11 is driven by a built-in battery or an external battery, is connected to an external AC power supply, or is converted from an AC power supply to an AC-DC converter. Can be driven by direct current.

Next, an embodiment of the drip detection device connected to the flow rate measuring device 22 will be described with reference to FIG.
FIG. 3 shows a schematic configuration of the drip detection device 40 according to the embodiment of the present invention. The drip detection device 40 includes a transparent slightly elongated drip tube 41 made of glass or synthetic resin, and the drip tube. A light emitting unit 42 disposed on one side outside 41, and a two-dimensional image sensor (hereinafter referred to as “image sensor”) 43 disposed at a position facing the light emitting unit 42 across the infusion tube 41. I have.

Here, the image sensor 43 is formed using a CCD (Charge Coupled Device), a CMOS (Metal Oxide Film Semiconductor Device) or the like, and takes in a two-dimensional image by arranging at least each pixel in a matrix (array). The image is widely used to generate binarized image data to be described later.
The light emitting unit 42 is most preferably an LED (light emitting semiconductor) in which a predetermined lens is integrated so as to emit slightly diffusing light. In addition, when a point light source is used, diffusion before the infusion tube 41 is performed. It is preferable to use a collimator lens in the optical path to make it a parallel light beam. A surface light source such as an organic EL can be suitably used, and a light emitting lamp such as a normal bulb may be used. Is preferably used.

The sensor control unit 44 can use an IC circuit centered on the CPU, and is connected to the light emitting unit 42 to issue a lighting command. The sensor control unit 44 is connected to the image sensor 43, acquires a binarized image from the image sensor 43 at a timing described later, and stores it in the storage unit 48. Further, as will be described later, the sensor control unit 44 determines “droplet formation state”, “state before drop drop”, and “droplet drop state” for the image obtained by the image sensor 43, and calculates the calculation unit. The required volume is obtained by instructing 47 and the flow rate is calculated. This point will be described in detail later.
The storage unit 48 is a memory and may also be used as the storage unit 29 in FIG. The timer 49 measures the time from the start of operation of the infusion pump 11 and may be provided integrally with the control unit 20 of the infusion pump 11.
The calculation unit 47 is connected to the sensor control unit 44 and calculates the volume of the droplet from the binarized image stored in the storage unit 48 as described later.
The pump control unit 45 is the control unit 20 of FIG. 2, and the motor 46 and the tube pressing unit 35 constitute the administration drive unit 30 of FIG.

Here, the image sensor 43 is disposed on the opposite side of the light emitting unit 42 with the infusion tube 41 interposed therebetween. At this time, the light emitting unit 42 and the image sensor 43 are arranged so that the dropping nozzle 51 existing above the drip tube 41 is reflected.
The image sensor 43 emits light from the light emitting unit 42 that is a light source, and receives light reaching the image sensor 43 without causing interference by the dropping nozzle 51 or reflection by the droplet. Therefore, the portion blocked by the dropping nozzle 51 and the droplet 60 becomes a shadow, and as a result, the image sensor 43 can reflect the shape of the droplet 60 and the dropping nozzle 51 (particularly the shape of the lower end).
At this time, it appears as shown in FIG. However, depending on the chemical solution, the portion where the liquid surface faces perpendicularly to the direction of light transmits light, so that only the contour of the chemical solution becomes a shadow and may be projected as shown in FIG. In any case, it is possible to recognize the presence and shape of the dripping nozzle 51 and the chemical solution by processing the contour.

When infusion is performed using a natural drop or the infusion pump 11, the chemical solution passing through the drip tube 41 falls from the drip nozzle 51 in the drip tube 41 into the air and is stored below the drip tube 51. At this time, the chemical solution does not drop immediately due to the influence of the surface tension, but is stored in a state of being held by the dropping nozzle 51.
When the amount of the chemical liquid held by the dropping nozzle 51 increases, a liquid pool is formed in the nozzle. When the amount of the liquid pool exceeds a certain level, the gravity becomes larger than the surface tension, so that the chemical liquid separates from the nozzle and becomes a droplet. Fall. At this time, the state in which the chemical solution is stored at the lower end of the dropping nozzle 51 and becomes a liquid pool is referred to as a “droplet forming state”.

  When the liquid pool stored in the dripping nozzle 51 grows and becomes a liquid droplet 60 and leaves the nozzle, the chemical liquid becomes narrower in the diameter of the cross section above the liquid pool, that is, near the dripping nozzle 51. 61 is formed. The constriction 61 becomes thinner with time, and is divided into a droplet 62 that is cut and dropped and a residue 63 remaining in the dropping nozzle 51. The state until the constriction 61 is generated in the liquid pool and the falling chemical liquid is formed is referred to as a “state before dropping the liquid drop”. A state in which the liquid droplet is falling away from the nozzle is referred to as a “liquid droplet falling state”. During infusion, the drip tube 41 repeats the droplet formation state, the pre-droplet drop state, and the drop drop state. In general, since the time in which the liquid drop is falling is short, a high shooting speed is required, but in the other two conditions, the time is longer than in the liquid drop state and the change is slow, so even at a slow shooting speed. It is possible to capture changes.

Next, an embodiment of a method for accurately controlling the infusion flow rate of the infusion based on the above-described image by the infusion detection device 40 of FIG. 3 or the infusion pump 11 incorporating the same will be described.
Please refer to FIG.
In FIG. 5, among the matrix-like pixels of the image sensor 43 described in FIG. 3, predetermined horizontal locations (one horizontal row of predetermined locations) are respectively selected at the upper and lower positions, and these are selected. Are used as line sensors.
Specifically, among the array sensors specified along the horizontal direction of the image sensor 43, reference numeral 43-1 is an upper detection sensor (close to the lower end of the dropping nozzle 52 in FIG. 3 and slightly below the residue 63. And a lower detection sensor 43-2 (a region lower than the dripping nozzle of FIG. 3 and slightly lower than the constricted portion 61). The detection images of the upper detection sensor 43-1 and the lower detection sensor 43-2 are used.

  The position of the upper detection sensor 43-1 is set to a height at which a constriction 61 is produced before a liquid pool grows and drops as a droplet. The width of the detection region is set to be larger than the length of the shadow generated on the droplet or the dropping nozzle 52, and the height is set sufficiently smaller than the size of the droplet. For example, the horizontal width is the same as the width of the image, and the height is the height of one pixel. The length shadowed by the chemical in each region is obtained, and the shadow length in the upper detection sensor 43-1 is Y1, and the shadow length in the lower detection sensor 43-2 is Y2. In the state before the droplet is dropped, the upper detection sensor 43-1 is narrowed by the constriction 61, and Y1 <Y2. However, since the chemical is connected to the droplet nozzle 52, both Y1 and Y2 have values greater than zero.

  Therefore, with respect to Y1, in FIG. 5A, the reference value “a” in the state of the constriction 61 is used, and when 0 <Y1 <a and Y2> a, the “state before droplet drop” is defined. To do. Then, as shown in FIG. 5B, when the liquid pool becomes large and is separated into the falling droplet 62 and the residue 63 remaining in the dropping nozzle 52, the upper detection sensor 43-1 detects the droplet 62 and the dropping nozzle 52. Because it will be in the middle, the chemical will not be reflected. Therefore, Y1 = 0, and Y2> 0 because a shadow is caused by the falling droplet 62. Therefore, the droplet drop state is set when Y1 = 0 and Y2> 0. The other state shown in FIG. 5C is a “droplet formation state”.

Next, an example of a technique that can be determined as “droplet falling state” will be described.
In FIG. 6, the line sensor unit used for detection is moved from the upper end of the image downward (X direction in FIG. 6) so as to change to the positions of 43-1, 43-2, and 43-3, and sequentially moved. The shadow formed by the droplet nozzle 52 or the chemical solution is measured by one line sensor unit that is moved (scanned). The length (width dimension) of the shadow at this time is Y (X). The state of the image is determined by a change in Y (X) when X is moved downward.
In the state before the droplet shown in FIG. 6A falls, a constriction 61 occurs in the chemical solution, and Y (X) once decreases as X increases and becomes a value equal to or less than the reference value a. The reference value a is a threshold value that is set to a degree slightly exceeding the size Y of the constriction. However, before Y (X) = 0, Y (X) increases again due to the chemical before dropping as a droplet. Thereafter, it decreases and Y (X) = 0. Therefore, as X increases, Y (X) decreases until <Y (X) becomes 0 <Y (X) <a, and then increases and decreases. State ". In the “droplet falling state”, Y (X) = 0 once and then increases again, and Y (X) = 0. Depending on the conditions under which droplets can be formed, Y (X) = 0, and a change in which Y (X) increases again can occur a plurality of times. Therefore, once Y (X) = 0, the increasing change occurs once more, and when Y (X) = 0 at the lower end of the image (X is maximum), the “droplet falling state” is set. .
When the above conditions are not satisfied, the “droplet formation state” is set.

Next, an example of a method for calculating the volume from the binarized image will be described.
FIG. 7 illustrates a method for calculating the volume using the droplet 62 which is an image determined as “droplet falling state”.
As shown in FIG. 7, it is assumed that the image of the droplet 62 is divided into n rectangles having a minute height h, and the actual droplets 62 are stacked with n disks each having a diameter of the rectangle. To do. Then, the volume of the n disks is calculated from the width and height of the rectangle, and the total is set as the volume of the droplet (that is, the volume is calculated by the division quadrature).

Next, a first embodiment of a method for calculating a flow rate using an image acquired as described above will be described with reference to FIGS. The following operations are performed using the calculation unit 47 in accordance with instructions from the sensor control unit 44 in FIG. The sensor control unit 44 and the calculation unit 47 can also be used by the control unit 20 and the storage unit 29 in FIG.
8, in step 1, the image sensor 43 described in FIG. 3 acquires an image (ST1). Preferably, the acquired image is subjected to a binarization process (ST2) so that it can be easily processed later (ST2).
In the next step, the state of the liquid droplets or the chemical solution projected on the image is determined (ST3).
Whether or not a drop state of the droplet is detected is detected (ST4). If not, the process returns to ST1 after a predetermined time S (for example, 4 ms).
When a droplet is detected in ST4, the state of the previous image is examined (ST5). If the captured image is in the “droplet falling state” and the previous image is in the “droplet falling state”, the volume is calculated using the captured image in this step ( ST6), the flow rate is calculated (ST7). This series of processes is repeated every certain time S (for example, 4 ms) (ST8).

FIG. 9 shows a calculation method of the second embodiment.
In step 11, the image sensor 43 acquires (ST11). This acquired image is binarized so that it can be easily processed later (ST12) (it is not necessary if it is not necessary).
Then, in the next step (ST13), the state of the liquid droplet or the chemical liquid projected on the image is determined. That is, it is determined whether or not the state of the image is a state before the drop of the liquid droplet (ST14). If it is not the state before the drop of the liquid droplet, the process returns to ST11 after a certain time S1 (for example, 30 ms) (ST23).

If it is determined in ST14 that the state is “before droplet drop”, i = 0 (ST15), and the image sensor 43 captures an image in the same manner as above (ST16). One image is captured (ST17). Each time i is increased by 1 (ST18). In ST19, it is determined whether or not i has reached a certain number i ′ (for example, 20). If it has not reached yet, it returns to ST16 after a certain time S2 (for example, 4 ms, a value sufficiently smaller than S1) (ST20). , Import images. In other words, if it is determined in ST14 that the state is before the drop, the image sensor capturing time interval is changed from S1 to S2, and i ′ images are captured. In ST19, i = i ′, that is, when i ′ imaging is completed, the process proceeds to the volume calculation process (ST21), and after a predetermined time S2 has elapsed (ST22), the process returns to ST11.
Details of the volume calculation process ST21 are shown in FIG. In the volume calculation process, the images are binarized according to the recorded order (ST31), and the state of the image is determined (ST32). It is determined whether or not a droplet as a binarized image is in a droplet falling state (ST33), and if a positive result is obtained, the state of the previous image is examined (ST34). Then, it is determined whether or not the state of the previous image is a state before the drop of the liquid droplet (ST35). If an affirmative result is obtained, the volume of the determined image is calculated (ST36) and the flow rate is calculated (ST36). ST37). When a negative result is obtained in each determination unit of ST33 and ST35, the presence or absence of the next image is confirmed (ST38), and if an image exists, the image is updated (ST39), and the process returns to ST31. If there is no next image, the volume calculation process is terminated (ST40), and the process waits until it is started again.

FIG. 10 shows a third embodiment. In this embodiment, each time an image is captured, the state is determined, and based on the determined state, the flow rate calculation and the time interval until the next image capture are determined.
First, the image sensor 43 acquires an image (ST51). The acquired image is binarized so that it can be easily processed later (ST52) (unnecessary if not necessary). In the next step, the state of the liquid droplets or the chemical solution projected on the image is determined (ST53).
It is determined whether or not the presence or absence of a drop drop state is detected (ST54). If an affirmative result is obtained, that is, if the droplet is in a drop state, the process proceeds to the step of calculating the volume (ST56). Based on the obtained volume, the flow rate and speed are calculated (ST57). Then, after a predetermined time (for example, 30 ms) has elapsed, the process returns to ST51 which is an image acquisition step.
If no droplet is detected in ST54, it is determined whether or not the droplet is in a state before the droplet is dropped (ST58). If it is in the state before the droplet is dropped, a predetermined time S3 (for example, 4 ms) elapses. Later (ST60), the process returns to the image capturing step (ST51), and a series of operations are repeated. If the determination result in ST58 is “droplet formation state”, a series of operations are repeated from the image capturing step after a predetermined time S4 (for example, 30 ms) has elapsed. At this time, the time S4 is sufficiently longer than the time S3.

(Example of flow rate calculation method)
After the calculation of the volume of the droplet, in the above-described image acquisition, an ID associated with the detected droplet order is attached and the time when the image of the droplet is captured and the calculated volume are recorded in the storage unit 48 of FIG. When a new recording is performed, the time of the droplet to be recorded this time is compared with the time of the last recorded droplet. If the difference is less than a certain time (for example, within 50 ms), the two are the same droplet. Shall be recorded.
After the volume and time of the droplet are recorded, the flow rate is calculated. If the recorded droplet is the Nth droplet from the start of detection, and the volume of the Nth droplet is V (N) and the time is T (N), the flow rate Q is
Q = V (N) / (T (N) -T (N-1))
Is required. Also using multiple K droplets
Q = (V (N) + V (N−1) +... V (N−K + 1)) / (T (N) −T (N−K))
It is also possible to ask for.

  As described above, according to the present embodiment, when capturing a droplet that freely falls in the drip tube due to gravity, it is possible to reliably capture the falling droplet even at a relatively low speed. In addition, since a sensor for detecting that a droplet has fallen in advance is not required, volume calculation can be performed while suppressing the price of a sensor that performs imaging. In addition, by adjusting the shooting interval according to the size of the chemical liquid held in the nozzle, it is possible to reduce the number of times of shooting and to reduce power consumption.

The present invention is not limited to the embodiments described above.
As a further configuration example of the present invention, instead of the pump control unit 45, the motor 46, and the tube pressing unit 35, the tube is pressure-closed, and the adjustment valve for controlling the liquid feeding speed due to natural fall and the pressure-closing degree of the adjustment valve are set. A method of controlling the liquid feeding speed by adjusting the pressure closing degree of the regulating valve based on the calculated liquid feeding speed may be provided.
A part of the configuration of each embodiment described above can be omitted, or the individual configurations of each embodiment can be interchanged and combined. It can also be combined with other configurations not described above.

  DESCRIPTION OF SYMBOLS 11 ... Infusion pump, 22 ... Flow rate measuring device, 40 ... Drip apparatus, 41 ... Drip tube, 42 ... Light emission part, 43 ... Image sensor, 44 ... Sensor control part

Claims (8)

A transparent drip tube,
A light emitting portion disposed on one side of the outside of the infusion tube;
A two-dimensional image sensor disposed at a position facing the light emitting unit and the infusion tube,
The drip detection apparatus, wherein the image sensor is set to include at least a tip of a drip nozzle in the drip tube and a predetermined drop distance of a drop falling from the drip nozzle. The body,
Drive means such as a liquid feeding motor accommodated in the main body;
Various sensors for infusion operation,
A control device to which the driving means and the sensor are connected;
An infusion tube extending from an external drip detection device and connected to the main body,
Containing and holding an infusion tube extending into the main body, and a pump unit for feeding the liquid in the infusion tube, and
The drip detection device is
A transparent drip tube,
A light emitting portion disposed on one side of the outside of the infusion tube;
A two-dimensional image sensor disposed at a position facing the light emitting unit and the infusion tube,
The infusion pump, wherein the image sensor is set so as to include at least a tip of a dropping nozzle in the drip tube and a predetermined drop distance of a droplet falling from the dropping nozzle.   The infusion detection device and / or the infusion pump connected to the infusion detection device includes an imaging control unit and a recording unit for storing imaging data, and the imaging control unit controls the two-dimensional image sensor. Then, the state until the droplets are separated from the tip of the dropping nozzle and fall as droplets is photographed as a series of moving images or photographing data which is a plurality of screen data, stored in the recording unit, and the photographing Based on the data, by determining that a liquid droplet has fallen off the dropping nozzle, the image data of the liquid droplet immediately after being dropped from the dropping nozzle of the chemical liquid is specified, The infusion pump according to claim 2, wherein the volume of the droplet of image data is calculated.   As a means for discriminating that a liquid droplet has fallen off the dropping nozzle, the image immediately after detecting a constriction generated when the chemical liquid becomes a liquid droplet is set as a liquid droplet falling state. 4. The infusion pump according to claim 3, wherein the volume of the droplet is detected based on an image in a falling state.   As a means for discriminating that a liquid droplet has dropped from the dropping nozzle, the chemical liquid held at the tip of the dropping nozzle can be discriminated as a droplet within a predetermined time. The infusion pump according to claim 3, wherein the infusion pump is configured to detect the volume of the droplet based on an image of the droplet after the lapse of time, based on data held in the record. A transparent infusion tube, a light emitting portion arranged on one side outside the infusion tube, and a two-dimensional image sensor arranged at a position facing the light emitting portion with the infusion tube interposed therebetween, The image sensor uses the apparatus set to include at least the tip of the dropping nozzle in the drip tube and a predetermined drop distance of the droplet falling from the dropping nozzle to calculate and control the liquid feeding amount. A method for determining the volume of a droplet,
The imaging control unit controls the two-dimensional image sensor, and the state from when the droplet leaves the tip of the dropping nozzle and falls as a droplet is taken as a series of moving images or shooting data that is a plurality of screen data. Take a picture and save it in the recording part,
Based on the imaging data, it is determined that a liquid droplet has fallen off the dropping nozzle,
Based on the determination, the image data of the liquid droplet immediately after being dropped from the dropping nozzle of the chemical solution is specified, and the volume of the liquid droplet is calculated using the liquid droplet of the image data. A method for controlling the apparatus.   As a method for determining that a liquid droplet has fallen off the dropping nozzle, the image immediately after detecting a constriction generated when the chemical liquid becomes a liquid droplet is set as a liquid droplet falling state. The method for controlling the apparatus according to claim 6, wherein the volume of the droplet is detected based on an image in a falling state.   As a method for determining that a liquid droplet has fallen off the dropping nozzle, the chemical liquid held at the tip of the dropping nozzle is determined based on data dropped as a droplet within a predetermined time. The apparatus according to claim 6, wherein after detecting the droplet accumulated at the tip of the dropping nozzle, the volume of the droplet is detected based on an image of the droplet after the lapse of time. Control method.

Images