JP2006280571A - Blood pump apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a blood pump apparatus by which the efficiency of electric transmission of power can be controlled to its optimum level according to variations in the load on the pump and the arrangement of coils. <P>SOLUTION: The blood pump apparatus 1 is provided with a pump 5, an external apparatus 7 which is equipped with a power transmission part 71 with a coil 70 for electric power transmission and a voltage detection part 73 for the coil 70 for electric power transmission and the transmission power controller 76 and an internal implantation type apparatus 6 which is equipped with a power supplying portion 61 with a coil 65 for electric power receiving for receiving electric power from the coil 70 for electric power transmission, a load current detection unit 64 for the power supplying portion 61 and a power information transmitting section 62 for transmitting the detected current value to the external apparatus 7. The transmission power controller 76 is provided with an optimum frequency-related relational expression memory part 74, the function of calculating the optimum frequency (an operating section 75) for calculating the optimum frequency of the coil 70 for electric power transmission by using the optimum frequency-related relational expression, an input distance between coils, a load current value to be received and the voltage value of the coil 70 for electric power transmission and the function of control which allows the coil 70 for electric power transmission to supply power at its optimum frequency. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a blood pump apparatus of a type that receives a power supply percutaneously.

As a blood pump device, there is a device constituted by a pump embedded in a living body and an external body device that is not directly connected to the pump and supplies electric power to the pump transcutaneously.
In the control system of the transcutaneous power transmission system, in general, the current or voltage supplied to the load in the body is detected, and the frequency is changed so that the supplied power is maximized. In the case of the PWM control system, the duty ratio and frequency are changed. The method of changing is used. Specifically, the voltage between the coils for power transmission is detected, and the frequency is changed so that this voltage becomes maximum. Furthermore, in order to improve the efficiency, there are a synchronous rectification system and a system controlled by a power factor. In any of the methods, since the displacement of the coil leads to a reduction in efficiency, a detection method thereof has been proposed.
For example, in Patent Document 1 (Japanese Patent Laid-Open No. 11-123244, a method and apparatus for transmitting power to an implantable device), a method for determining a coil position shift from a voltage value (received voltage information) between secondary coils. Has been proposed.

JP 11-123244 A

In an extracorporeal implant type transcutaneous power transmission system, it has been a problem to always supply power with high efficiency against fluctuations in the load on the pump and displacements of opposing coils.
An object of the present invention is to provide a blood pump device that can be controlled so as to optimize the power transmission efficiency according to the load variation of the pump and the arrangement of the coils.

What achieves the above object is as follows.
(1) A pump comprising a housing having a blood inflow port and a blood outflow port, an impeller that rotates within the housing and feeds blood, and an impeller rotational torque generating unit for rotating the impeller, An implantable device embedded in a living body including a power supply circuit for driving the impeller rotational torque generating unit; a power transmission unit including a power transmission coil for transmitting power to the implantable device; and the power transmission A blood pump device comprising a voltage detection unit for detecting a voltage applied to both ends of a coil for use, and an external body device including a transmission power control unit and not connected to the implantable device,
The implantable device includes a power receiving coil that receives power from the power transmitting coil and detects a load current sent from the power source to the power circuit. A current value detection unit, and a power information transmission unit that transmits the current value detected by the current value detection unit to the external device,
The extracorporeal device includes an inter-coil distance input unit that inputs an axial distance between the power transmission coil and the power reception coil, and a power information reception unit that receives information from the power information transmission unit,
The transmission power control unit is an axial distance between the power transmission coil and the power reception coil at an optimal frequency of a number of power transmission coils measured in advance, a load current of the implantable device, The voltage applied to both ends of the power transmission coil is used as input layer data, and the optimal frequency of the power transmission coil related to each input layer data is used as teacher data, and the relational expression between the input layer data calculated using a neural network and the optimal frequency A relational expression storage unit for storing the optimum frequency related relational expression, the optimum frequency related relational expression, the inter-coil distance input to the inter-coil distance input part, the load current value received by the power information receiving part, Optimal frequency calculation function that calculates the optimal frequency of the coil for power transmission using the coil voltage value for power transmission detected by the coil voltage detector for power transmission , The optimum the optimum frequency calculated by the frequency computing function, by which the blood pump device and a frequency control function to perform the power supply by the power transmission coil.

(2) The extracorporeal device includes a display unit, and the transmission power control unit includes the power transmission coil and the power reception in a radial deviation amount between a number of the power transmission coils and the power reception coil measured in advance. The axial distance between the coils for the coil, the load current in the implantable device, and the voltage applied to both ends of the power transmission coil as input layer data, and the radial direction between the coils for each input layer data A relational expression storage unit for storing a relational expression related to a radial deviation amount between coils, which is a relational expression between input layer data calculated using a neural network and a radial deviation quantity between coils, using the deviation amount as teacher data, and the coil The inter-coil direction deviation amount relational expression, the inter-coil distance input to the inter-coil distance input unit, the load current value received by the power information receiving unit, and the transmission coil voltage detection unit The inter-coil radial displacement amount calculation function for calculating the inter-coil radial displacement amount using the detected power transmission coil voltage value, and the inter-coil radial direction calculated by the inter-coil radial displacement amount calculation function. The blood pump device according to (1), wherein the blood pump device has a function of displaying a shift amount on the display unit.
(3) The outside-body device includes an alarm unit, and the transmission power control unit stores an allowable inter-coil radial direction deviation amount, and calculates an inter-coil diameter calculated by the inter-coil radial direction deviation amount calculation function. The blood pump device according to the above (2), which has an alarm function for operating the alarm means when the amount of direction deviation is larger than the allowable amount of deviation in the radial direction between coils stored.

Moreover, what achieves the said objective is as follows.
(4) The method for manufacturing a blood pump device according to any one of (1) to (3) above, wherein the power transmission coil and the power reception coil are measured at the optimum frequency of a large number of power transmission coils measured in advance. The axial distance between the coils, the load current of the implantable device, and the voltage applied to both ends of the power transmission coil are used as input layer data, and the optimum frequency of the power transmission coil related to each input layer data is instructed. Using a neural network as data, a relational expression calculating step for calculating an optimal frequency related relational expression that is a relational expression between the input layer data and the optimal frequency, and an optimal frequency related relation obtained by the optimal frequency related relational expression calculating step A method for manufacturing a blood pump device, comprising: storing a relational expression in the blood pump device so that the transmission power control unit can use the relational expression.

Moreover, what achieves the said objective is as follows.
(5) In the method for manufacturing the blood pump device according to any one of (2) and (3), the radial displacement amount between a plurality of the power transmission coils and the power reception coils measured in advance, The distance between the coils in the axial direction between the power transmission coil and the power reception coil, the load current in the implantable device, and the voltage applied to both ends of the power transmission coil are used as input layer data, and each input layer data Using a neural network as a teacher data for the amount of radial deviation between coils with respect to the coil, a relational expression between the input layer data and the amount of radial deviation between coils is calculated using a neural network. The transmission power control unit can use the radial deviation amount relational expression calculation step and the inter-coil radial deviation amount relational expression calculation step obtained by the inter-coil radial deviation amount relational expression calculation step. A method of manufacturing a blood pump device that performs a storing step of storing the blood pump device.
(6) The method for manufacturing a blood pump device according to (4) or (5), wherein the neural network is a three-layer feedforward neural network including three layers of an input layer, an intermediate layer, and an output layer.

    According to this blood pump device, the optimal frequency calculation function and the control function for supplying power by the power transmission coil using the optimal frequency calculated by the optimal frequency calculation function are provided, so that efficient power supply is performed. be able to.

The blood pump device and the method for producing the blood pump device of the present invention will be described using examples.
FIG. 1 is a block diagram of an embodiment of a blood pump device of the present invention. FIG. 2 is a front view of a pump used in the blood pump device of the present invention. FIG. 3 is a plan view of the pump shown in FIG. FIG. 4 is a longitudinal sectional view of the pump shown in FIG. 5 is a cross-sectional view of the pump of FIG. 2 along the line AA.
The blood pump device 1 of the present invention includes a housing 20 having a blood inflow port 22 and a blood outflow port 23, an impeller 21 that rotates in the housing 20 and feeds blood, and an impeller rotational torque for rotating the impeller. Implantable device 6 that is implanted in a living body that includes a pump 5 that includes the generating unit 3 and a power supply circuit that drives the impeller rotational torque generating unit 3, and power transmission that transmits power to the implanted device. Power transmission unit 71 provided with a coil 70 for power transmission, a voltage detection unit 73 for detecting a voltage applied to both ends of the power transmission coil 70, and a power transmission control unit 76, and an external device not connected to the implantable device 6 7.

The implantable device 6 includes a power receiving coil 65 that receives power from the power transmitting coil and supplies power to the power circuit 52, and a load current sent from the power supply 61 to the power circuit 52. A current value detection unit 64 that detects the current value, and a power information transmission unit 62 that transmits the current value detected by the current value detection unit 64 to the external body device 7. The pump 5 and the implantable device 6 constitute an implantable system 10.
The external device 7 includes an inter-coil distance input unit 77 that inputs an axial distance between the power transmission coil 70 and the power reception coil 65, and a power information reception unit 72 that receives information from the power information transmission unit 62. With.
And the transmission power control part 76 is the optimal frequency of many power transmission coils measured in advance, the axial distance between the power transmission coil and the power reception coil, the load current of the implantable device, the power transmission The voltage applied to both ends of the power coil is the input layer data, the optimal frequency of the power transmission coil for each input layer data is the teacher data, and the optimal expression is the relational expression between the input layer data calculated using a neural network and the optimal frequency Relational expression storage unit 74 for storing frequency-related relational expression, optimum frequency-related relational expression, intercoil distance input to intercoil distance input unit, load current value received by power information receiving unit, and coil voltage detection for power transmission An optimum frequency calculation function (calculation unit 75) for calculating the optimum frequency of the power transmission coil using the power transmission coil voltage value detected by the unit; The optimum frequency calculated by the frequency computing function, and a frequency control function to perform the power supply by the power transmission coil.

  When radial deviation occurs between the power transmission coil and the power reception coil during use, the power transmission efficiency decreases. In this case, by changing the frequency, it is possible to perform power transmission with optimum efficiency in the shifted state. The pump device of the present invention has the above-described calculation of the optimum frequency and the control function based on the calculated frequency. In the present invention, it is not limited to so-called alternating current, and any transmission system in which the frequency affects the power transmission efficiency may be used. For example, it is also effective in frequency control in power transmission by the PWM method.

Further, like the blood pump device 1 as in this embodiment, the external body device 7 includes a display unit 59, and the transmission power control unit 76 has a diameter between a large number of power transmission coils and power reception coils measured in advance. The input layer data includes the axial coil distance between the power transmission coil and the power reception coil, the load current in the implantable device, and the voltage applied to both ends of the power transmission coil in the direction deviation amount. Using the amount of radial deviation between coils in relation to data as teacher data, the relational expression for the relationship between the radial deviations between coils, which is the relational expression between the input layer data calculated using a neural network and the radial deviation between coils, is stored as a relational expression. Stored in the unit 74, the inter-coil radial deviation amount relational expression, the inter-coil distance input to the inter-coil distance input unit, the load current value received by the power information receiving unit 72, the transmission coil An inter-coil radial direction deviation amount calculation function (calculation unit 75) that calculates an inter-coil radial direction deviation amount using the power transmission coil voltage value detected by the voltage detection unit 73, and an inter-coil radial direction deviation amount calculation function. It is preferable to have a function of causing the display unit 59 to display the amount of radial deviation between the coils calculated by the above.
In particular, the external body device 7 includes an alarm unit 78, and the transmission power control unit 76 stores the allowable inter-coil radial direction deviation and calculates the inter-coil radial direction by the inter-coil radial deviation calculation function. It is preferable to provide an alarm function that activates the alarm means 78 when the deviation amount is larger than the memorized allowable radial deviation between the coils.
As the alarm means, a sounding means, a lighting means and the like are suitable. A buzzer is preferable as the sounding means. As the lighting means, a lamp is preferable.

The pump 5 shown in FIGS. 2 to 5 includes a housing 20 having a liquid inflow port 22 and a liquid outflow port 23, and a magnetic body 25 therein. The pump 5 rotates within the housing 20 and is rotated by centrifugal force during rotation. A rotor 31 having a centrifugal liquid pump unit 2 having an impeller 21 for feeding liquid, a magnet 33 for attracting the magnetic body 25 of the impeller 21 of the centrifugal liquid pump unit 2, and a motor 34 for rotating the rotor 31. The impeller rotational torque generator 3 provided and the impeller position controller 4 provided with an electromagnet 41 are provided, and the impeller 21 rotates with respect to the housing 20 in a non-contact state.
The pump 5 used in the blood pump device of the present invention is not limited to the type in which the above impeller rotates in a non-contact state. For example, it can be used for a type in which an impeller is joined to a shaft of a motor and the impeller rotates by rotation of the motor.

As shown in FIGS. 2 to 5, the pump 5 includes a housing 20 having a blood inflow port 22 and a blood outflow port 23, and an impeller 21 that rotates in the housing 20 and feeds blood by centrifugal force during rotation. A centrifugal liquid pump unit 2 having an impeller, an impeller rotational torque generating unit (non-control type magnetic bearing component) 3 for the impeller 21, and an impeller position control unit (control type magnetic bearing component) 4 for the impeller 21. With.
As shown in FIG. 4, the impeller 21 is held at a predetermined position in the housing 20 by the action of the non-control type magnetic bearing component 3 and the control type magnetic bearing component 4 and normally does not contact the inner surface of the housing. Rotate.
The housing 20 includes a blood inflow port 22 and a blood outflow port 23, and is formed of a nonmagnetic material. A blood chamber 24 communicating with the blood inflow port 22 and the blood outflow port 23 is formed in the housing 20. An impeller 21 is accommodated in the housing 20. The blood inflow port 22 is provided so as to protrude substantially vertically from the vicinity of the center of the upper surface of the housing 20. As shown in FIGS. 3 and 5, the blood outflow port 23 is provided so as to protrude in a tangential direction from the side surface of the housing 20 formed in a substantially cylindrical shape.

As shown in FIG. 5, a disc-like impeller 21 having a through-hole at the center is housed in a blood chamber 24 formed in the housing 20. As shown in FIG. 4, the impeller 21 is formed between a donut plate-like member (lower shroud) 27 that forms a lower surface, and a donut plate-like member (upper shroud) 28 that opens at the center that forms an upper surface. A plurality of (for example, seven) vanes 18. A plurality (seven) blood passages 26 partitioned by the adjacent vanes 18 are formed between the lower shroud and the upper shroud. As shown in FIG. 5, the blood passage 26 communicates with the central opening of the impeller 21, starts from the central opening of the impeller 21, and extends so that the width gradually increases to the outer peripheral edge. In other words, the vane 18 is formed between the adjacent blood passages 26. In this embodiment, each blood passage 26 and each vane 18 are provided at equal angular intervals and in substantially the same shape.
As shown in FIG. 4, a plurality (for example, 24 pieces) of first magnetic bodies 25 (permanent magnets, driven magnets) are embedded in the impeller 21. In this embodiment, the first magnetic body 25 is embedded in the lower shroud 27. The embedded magnetic body 25 (permanent magnet) attracts the impeller 21 to the side opposite to the blood inflow port 22 by a permanent magnet 33 provided on the rotor 31 of the impeller rotational torque generating unit 3 to be described later, and the rotational torque is rotated by the impeller. It is provided for transmission from the torque generator.

Further, by embedding a certain number of magnetic bodies 25 as in this embodiment, sufficient magnetic coupling with the rotor 31 described later can be ensured. The shape of the magnetic body 25 (permanent magnet) is preferably circular. Alternatively, a ring-shaped magnet may be polarized into multiple poles (for example, 24 poles), in other words, a plurality of small magnets may be arranged in a ring shape with alternating magnetic poles.
Further, the impeller 21 includes a second magnetic body 28 provided in the upper shroud itself or in the upper shroud. In this embodiment, the entire upper shroud is formed of the magnetic body 28. The magnetic body 28 is provided for attracting the impeller 21 to the blood inflow port 22 side by an electromagnet 41 of an impeller position control unit described later. As the magnetic body 28, magnetic stainless steel or the like is used.
The impeller position control unit 4 and the impeller rotational torque generating unit 3 constitute a non-contact magnetic bearing, and the impeller 21 is pulled in an opposite direction so that the appropriate position in the housing 20 that does not contact the inner surface of the housing 20 is obtained. The housing 20 rotates in a non-contact state.

As shown in FIG. 4, the impeller rotational torque generating unit 3 includes a rotor 31 housed in the housing 20 and a motor 34 for rotating the rotor 31. The rotor 31 includes a plurality of permanent magnets 33 provided on the surface on the pump unit 2 side. The center of the rotor 31 is fixed to the rotating shaft of the motor 34. The permanent magnets 33 are provided in plural and at equal angles so as to correspond to the arrangement form (number and arrangement position) of the permanent magnets 25 of the impeller 21.
The impeller rotational torque generating unit 3 is not limited to the one provided with the above-described rotor and motor, and may be composed of, for example, a plurality of stator coils for attracting and rotating the permanent magnet 25 of the impeller 21.
As shown in FIGS. 3 and 4, the impeller position control unit 4 includes a plurality of fixed electromagnets 41 for attracting the impeller magnetic body 28, and a position sensor for detecting the position of the impeller magnetic body 28. 42 is provided. Specifically, the impeller position control unit 4 includes a plurality of electromagnets 41 housed in the housing 20 and a plurality of position sensors 42. The plural (three) electromagnets 41 and the plural (three) position sensors 42 of the impeller position control unit are provided at equiangular intervals, and the electromagnet 41 and the position sensor 42 are also provided at equiangular intervals. ing. The electromagnet 41 includes an iron core and a coil. In this embodiment, three electromagnets 41 are provided. The number of electromagnets 41 may be three or more, for example, four. Three or more are provided, and these electromagnetic forces are adjusted using the detection result of the position sensor 42 to balance the force in the rotation axis (z-axis) direction of the impeller 21 and to be orthogonal to the rotation axis (z-axis). The moment about the x and y axes can be controlled.

The position sensor 42 detects a gap interval between the electromagnet 41 and the magnetic body 28, and this detection output is a control unit of a control mechanism (in other words, a controller) 6 that controls the current or voltage applied to the coil of the electromagnet 41. 51. Even if a radial force due to gravity or the like acts on the impeller 21, the magnetic flux shearing force between the permanent magnet 25 of the impeller 21 and the permanent magnet 33 of the rotor 31 and the force between the electromagnet 41 and the magnetic body 28. Since the shearing force of the magnetic flux acts, the impeller 21 is held at the center of the housing 20.
As shown in FIG. 1, the implantable device 6 includes a power receiving coil 65 that receives power from the power transmitting coil 70, and supplies a power to the power circuit 52. A current value detection unit 64 that detects a load current sent to the power supply circuit 52 and a power information transmission unit 62 that transmits the current value detected by the current value detection unit 64 to the external body device 7 are provided.

Furthermore, in the blood pump device 1 of this embodiment, the implantable device 6 has a voltage value detection unit 63 that detects the voltage supplied from the power supply unit 61 to the power supply circuit 52, and the power information transmission unit 62 The function of transmitting the voltage value detected by the voltage value detection unit 63 to the external device 7 is provided. The external body device 7 has a function of receiving a signal related to the voltage value detected by the voltage value detection unit 63 transmitted from the power information transmission unit 62, and the operation information calculation unit 58 includes the pump 5. And a function of calculating the power consumption consumed by the implantable device 6 and a function of causing the display unit 59 to display the power consumption.
Further, as shown in FIG. 1, the implantable device 6 includes a power amplifier 52 and a motor control circuit 53 for the magnetic coupling motor 34, a power amplifier 54 for the electromagnet 41, and a sensor 42. A sensor circuit 55, a sensor output monitoring unit (not shown) for monitoring the sensor output, a control unit 51, and a power supply circuit 52 are provided. The control unit 51 has a motor current monitoring function.

Further, as shown in FIG. 1, the external body device 7 detects a voltage applied to both ends of the power transmission unit 71 including the power transmission coil 70 that transmits power to the in-vivo implantable device 6 and both ends of the power transmission coil 70. Information from the voltage detection unit 73, the transmission power control unit 76, the inter-coil distance input unit 77 for inputting the axial distance between the power transmission coil 70 and the power reception coil 65, and the information from the power information transmission unit 62 are received. A power information receiving unit 72. The implantable device 6 is not connected to the external device 7 by connection.
Furthermore, the blood pump device 1 has a communication function different from the communication function related to the power information as shown in FIG. Specifically, the extracorporeal device 7 includes an extracorporeal communication interface 89, and the implantable device 6 includes an intracorporeal communication interface 88 that can communicate with the extracorporeal communication interface 89. Between the two interfaces, the motor current value detected by the implantable device 6 or a related signal is transmitted and received. The communication between the two may be either analog communication or digital communication. A known communication format can be used.
The external body device 7 includes a power supply unit 79, an operation information calculation unit 58, and a blood information input unit (in other words, blood viscosity, specific gravity, or hematocrit value) 57. The operation information calculation unit 58 has a function of calculating the discharge flow rate from the motor current value, the motor rotation value, and the blood information value input from the blood information input unit.

  And the transmission power control part 76 is the optimal frequency of many power transmission coils measured in advance, the axial distance between the power transmission coil and the power reception coil, the load current of the implantable device, the power transmission The voltage applied to both ends of the power coil is the input layer data, the optimal frequency of the power transmission coil for each input layer data is the teacher data, and the optimal expression that is the relation between the input layer data and the optimal frequency calculated using a neural network Relational expression storage unit 74 that stores frequency-related relational expression, optimum frequency-related relational expression, intercoil distance input to intercoil distance input unit, load current value received by power information receiving unit, coil voltage detection unit for power transmission An optimal frequency calculation function (calculation unit 75) for calculating the optimal frequency of the power transmission coil using the power transmission coil voltage value detected by the The optimum frequency that is calculated by the number calculation function, and a frequency control function to perform the power supply by the power transmission coil.

Further, the blood pump device of the present invention may be as shown in FIG.
The difference between the blood pump device 30 of this embodiment and the pump device 1 described above is only the presence or absence of the power information transmitting unit 62 of the in-vivo implanting device 6 and the power information receiving unit 72 of the external body device 7. . The blood pump device 30 is not provided with the power information transmitting unit 62 and the power information receiving unit 72 in the form as in the pump device 1. Instead, the detection signal from the voltage detector 63 is sent to the control unit 51 and transmitted by the communication interface 88. Information regarding the voltage transmitted from the communication interface 88 of the implantable device 6 is received by the communication interface 89 of the external device 7 and sent to the transmission power control unit 76. Therefore, also in this embodiment, the information regarding the voltage value of the power supply unit 61 is sent to the external body device 7, and the blood pump device 30 has a power information transmission function and a power information reception function. Yes.
Next, an optimum frequency related relational expression relational expression (using a sigmoid function) stored in the method for manufacturing the blood pump apparatus of the present invention and the arithmetic unit of the blood pump apparatus of the present invention will be described.

The method for manufacturing a blood pump device according to the present invention includes a distance between coils in the axial direction between the power transmission coil and the power reception coil at an optimum frequency of a number of power transmission coils measured in advance, Using the load current and the voltage applied to both ends of the power transmission coil as input layer data, the optimal frequency of the power transmission coil related to each input layer data as teacher data, using a neural network, the input layer data and the optimal frequency A relational expression calculating step for calculating an optimal frequency related relational expression that is a relational expression of the above, and an optimal frequency related relational expression obtained by the optimal frequency related relational expression calculating step so that the transmission power control unit can use the relational expression And a storage step of storing the blood pump device.
In the blood pump device of this embodiment, the power transmission coil stored in the transmission power control unit 76 of the extracorporeal device 7 and the amount of radial deviation between the many power transmission coils and the power reception coil measured in advance are The distance between the coils in the axial direction between the power receiving coils, the load current in the implantable device, and the voltage applied to both ends of the power transmitting coil are used as input layer data, and the radial deviation between the coils for each input layer data Using the neural network as a teacher data, the inter-coil radial deviation amount relational expression calculating step for calculating the inter-coil radial deviation amount related relation expression, which is a relational expression between the input layer data and the inter-coil radial deviation amount, Further, the inter-coil radial deviation amount relational expression obtained by the inter-coil radial deviation amount relational expression calculation step is used so that the transmission power control unit can use it. It has become to perform the step of storing the blood pump.
Therefore, in the embodiment to be described, the calculation of the optimum frequency, the control of the power transmission unit and the amount of radial deviation between the coils are calculated.
Each of the above relational expressions is calculated using a neural network.

As the neural network, it is preferable to use a three-layer feedforward network with three inputs and two outputs. The three layers are composed of an input layer, an intermediate layer, and an output layer. The input layer receives the axial coil distance between the power transmission coil and the power reception coil, the load current in the implantable device, and the voltage applied to both ends of the power transmission coil. The optimum frequency of the coil for use and the radial deviation between the coils are output. The intermediate layer is composed of a plurality of artificial elements (nonlinear elements). A three-layer neural network can approximate all functions that satisfy either the "continuity condition" or the "square integrable condition" with high accuracy. In order to increase the output accuracy of the neural network, the number of elements in the intermediate layer must be increased, but the amount of calculation increases accordingly.
The neural network is configured by using a plurality of artificial elements that simulate biological neurons. Here, the artificial element used in the neural network is not an exact model of a biological neuron, but a simplified engineering model that extracts a specific function. A deterministic analog model is used as the artificial element. This is a model in which the output y is determined as an analog value in the range of 0 ≦ y ≦ 1, and this output is deterministically determined from the input. The definitive analog model is shown in FIG.

シグモイド関数のゲインαを大きくすると、シグモイド関数はステップ関数に近づく。シグモイド関数とゲインαの関係を図８に示す。このモデルの出力計算手続きは次式により表される。

ここでθは、細胞内電圧の閾値である。 First, another element or an externally applied signal is input to the model. These values are written as {x ₁ , x ₂ , x ₃ ,… x _N }. This is multiplied by weights {w ₁ , w ₂ , w ₃ ,... W _N } and added to obtain an amount s corresponding to the intracellular potential. The weight corresponds to the synaptic connection weight. When obtaining the output y from the intracellular potential s, a continuous output is obtained using a sigmoid function. The sigmoid function is expressed by Equation 1 below.

When the gain α of the sigmoid function is increased, the sigmoid function approaches a step function. FIG. 8 shows the relationship between the sigmoid function and the gain α. The output calculation procedure of this model is expressed by the following equation.

Here, θ is a threshold value of the intracellular voltage.

を得る。これを重み付け総和と呼ぶ。次に、s_j ^（１）をシグモイド関数に通して、素子の出力

を得る。 The optimum frequency and coil deviation amount calculation unit of the present invention is configured by a feedforward type neural network. This is a neural network configured so that a signal flows in one direction from the input side to the output side. The neural network used in the present invention is a feedforward neural network composed of three layers: an input layer, an intermediate layer, and an output layer. The intermediate layer is preferably one layer, but may be two or more layers. The input layer is composed of three elements, and the axial distance between two opposing coils, the load current to the pump, and the voltage applied to both ends of the external coil are input. The intermediate layer is composed of a plurality of elements and is coupled to each element of the input layer and the output layer. In order to increase the calculation accuracy of the optimum frequency and the amount of deviation of the coil, it is necessary to increase the number of elements in the intermediate layer. However, since the calculation becomes complicated if the number of elements is excessively increased, the number of elements is preferably 3 to 10. An example of a feedforward type neural network used in the present invention is shown in FIG. The j-th intermediate layer element of the intermediate layer adds the connection weight w _ij ⁽¹⁾ to the input x _i and sums up for i = 1, 2, 3, and first, the intracellular potential (weighted sum)

Get. This is called a weighted sum. Next, let s _j ⁽¹⁾ pass through the sigmoid function and output the element

Get.

を得る。続いて、s_k ^（2）をシグモイド関数に通して、素子の出力

を得る。ここで、ｙ_k ^（2）(k＝1,2)は、最適周波数とコイルの径方向のずれ量である。具体的には、ｙ₁ ^（2）は、最適周波数であり、ｙ₂ ^（2）は、コイルの径方向のずれ量である。 When there are N intermediate layer elements j = 1, 2,..., N, the output is obtained by the above procedure for all of them. After obtaining the output of the intermediate layer element in this way, the kth output layer element is assumed that the output y _j ⁽¹⁾ of the intermediate layer element and y _k ⁽²⁾ (k = 1, ^{2) of the} output layer element This takes y _j ⁽¹⁾ (j = 1, 2,..., N) as its input, sums it with the “weight” w _jk ⁽²⁾ of the connection,

Get. Subsequently, s _k ⁽²⁾ is passed through the sigmoid function and the output of the element

Get. Here, y _k ⁽²⁾ (k = 1, 2) is the optimum frequency and the amount of deviation in the radial direction of the coil. Specifically, y ₁ ⁽²⁾ is the optimum frequency, and y ₂ ⁽²⁾ is the amount of deviation in the radial direction of the coil.

Next, a method for determining the values of the connection weight w of the artificial neural network mounted on the optimum frequency and the radial deviation calculation unit of the coil will be described.
A sequential update learning method is used to determine w. In the present invention, the connection weight is finely adjusted every time training data is given, using an error back-propagation method (back propagation) which is one of successive update learning methods.
Training data related to input / output to be realized is given as a plurality of input / output pairs. In the present invention, the training data corresponds to three inputs of the axial distance between opposing coils used as input, the current consumed by the pump and the implantable device (load current), and the voltage applied to both ends of the extracorporeal coil. This is the optimum frequency and the amount of deviation in the radial direction of the coil. The optimum frequency and the amount of deviation in the radial direction of the coil are given as a learning teacher signal.
Now, assume that there are L pieces of training data, and the m-th of them requires an output y ^(m) for an input (x ₁ ^(m) , x ₂ ^(m) , x ₃ ^(m) ). In this method, before learning, the radial distance between opposing coils, the current consumed by the pump and the implantable device (load current), the voltage applied to both ends of the extracorporeal coil, the optimum frequency, and the coil diameter It is necessary to collect calibration data in advance regarding the amount of deviation in direction. Specifically, for each of the axial distances 6, 8, 10,..., 20 mm (in 2 mm increments) between the opposing coils, the radial deviation amounts 0, 2, 4,..., 16 mm (in 2 mm increments) ), And determine the optimum frequency and voltage applied to both ends of the external coil at load currents of 0.2, 0.4, ..., 2.4A (in increments of 0.2A). Here, the optimum frequency is a frequency at which the transmission efficiency is highest for a given coil position and load. The number of calibration data points is 8 × 9 × 12 = 864 points because there are 8 axial distances between opposing coils, 9 radial displacements, and 12 load currents.

Here, the calibration data is acquired with an axial distance between the opposing coils of 2 mm intervals, a radial deviation of the coils of 2 mm intervals, and a load current of 0.2 A intervals. In consideration of increasing the frequency accuracy, it is desirable to make the measurement interval as fine as possible. However, if the amount of calibration data (the number of measurement points) increases, the calibration requires a considerable amount of time, so the above conditions are appropriate for the measurement interval.
FIG. 10 shows a learning procedure (learning routine) of the error reverse carrying method. The iteration loop inside FIG. 10 is a loop related to training data. When training data is selected each time (Step 1 below), parameters are corrected so as to reduce an error evaluation scale for the training data (see below). Step 2, 3). The outer loop is repeated until the error rating scale of all training data is sufficiently small.

Hereinafter, the error back propagation method will be described in detail.
<Step 1> Selection of Training Data In the inner loop of FIG. 10, the m-th training data is selected as the first procedure, and the following two steps are executed on this data. The m-th input is expressed as (x ₁ ^(m) , x ₂ ^(m) , x ₃ ^(m) ), but the right shoulder index is used to fix the training data during the two steps. Since (m) is not necessary, it is omitted. Instead, the variable's right shoulder index (m) is used to identify the layer. That is, the subscript (1) on the right side of the variable indicates that the variable is for the first layer (intermediate layer), and the subscript (2) indicates that the variable is for the second layer (output layer). The input layer is the 0th layer.

<Step 2> Output Calculation FIG. 11 shows a feedforward type neural network composed of three layers: an input layer, an intermediate layer, and an output layer. The input layer consists of nodes (signal relay points) indicated by black dots. The intermediate layer and the output layer are composed of elements (neuron engineering models) indicated by white circles. In the back propagation method, a deterministic analog model is used as an element.
The operation of this element is shown in FIG. If the input of the element is x ₀ , x ₁ ,..., X _N , the output is y, the coupling weight is w ₀ , w ₁ ,..., W _N , and the weighted sum (intracellular voltage) is s, the operation of the element is It can be formulated as follows.

を得る。続いて、s_j ^（1）をシグモイド関数に通して、素子の出力

を得る。中間層素子がj＝1,2,…,Nの N個あるとき、これらの全てについて上述の手続きにより中間層素子の出力を求める。出力層素子の出力y_k ⁽²⁾（k＝1,2）は、中間層素子の出力y_j ⁽¹⁾（j＝1,2,…,N）をその入力として受け取り、結線の「重み」w_jk ^（2）をかけて、重み付け総和

を得る。 Note that random values generated by random numbers or the like are assigned as initial values to all the connection weights in the network. However, the initial value must not be all zero. After applying the input to the input node, the j-th intermediate layer element adds the connection weight w _ij ⁽¹⁾ to the input x _i (= y _i ⁽⁰⁾ ) and sums up for i = 1, 2, 3 First, the weighted sum

Get. Next, s _j ⁽¹⁾ is passed through the sigmoid function and the output of the element

Get. When there are N intermediate layer elements j = 1, 2,..., N, the output of the intermediate layer element is obtained for all of them by the above procedure. The output y _k ⁽²⁾ (k = 1,2) of the output layer element receives the output y _j ⁽¹⁾ (j = 1, 2,..., N) of the intermediate layer element as its input, and the “weight” of the connection "W _jk ⁽²⁾ _multiplied by weighted sum

Get.

を得る。これがネットワークの出力b_k （＝y_k ^(２)）となる。
ここで、このネットワークの出力を学習の目標出力と比較する。目標出力をt_kとすると，誤差評価尺度Eは、

と表される。勾配法の原理により、この評価関数を小さくするように、ネットワーク内部の重みを修正する。 Subsequently, s _k ⁽²⁾ is passed through the sigmoid function to output the element

Get. This is the network output b _k (= y _k ⁽²⁾ ).
Here, the output of this network is compared with the target output of learning. When the target output is t _k , the error evaluation scale E is

It is expressed. Based on the principle of the gradient method, the weight inside the network is corrected so as to reduce this evaluation function.

を求めたあと、メモリに保持されているyを用いてα(1-y)yを計算し、これにuを掛け、

を出力する。zは左側の層の素子への入力となる。 <Step 3> Correction of the connection weights The procedure for correcting the connection weights is as follows. First, FIG. 12 shows a network flow when the connection weight is corrected (error back propagation mode). The number of nodes, the number of elements, the connection structure between elements, and the weight of the connection are the same as in FIG. 6 used for output calculation (output calculation mode), but the function of the elements and the direction of signal flow are different.
FIG. 13 shows element functions in the error back propagation mode. There is a memory inside the element, and its output y (= y _j ^(m) , hereinafter, subscript is omitted and abbreviated as y) is held in Step 2. Input _{z 1, z} 2 given from the right side of the device, ..., the z _N, the weight of the right connection of the elements _{w 1, w} 2, ..., multiplied by w _N, weighted sum

After calculating α (1-y) y using y held in the memory, multiply this by u,

Is output. z is the input to the left layer element.

なる関係で、シグモイド関数の微分が求まることから出てきたものである。以上のように動作する素子を図１２のように接続する。図１２の結線構造と結線の重みは図１１と全く同様である。図１２の右から、Step２で求めたネットワークの出力ｂ_jと目標出力ｔ_jとの誤差ｂ_j−ｔ_jを加える。この入力に基づき、各層で図１４に示した素子機能が働き、素子出力ｚ_j ^(m) が次々と定まる。こうして、誤差逆伝搬モードで各層の素子出力ｚ_j ^(m) を求めた後、これと出力計算モード（Step2）で求めた各層の素子出力y_j ^(m) を用いて、次式により結線の重みを修正する。

w_ij ^（m+1）は、第m層の第i素子と第 (m+1)層の第j素子を結ぶ結線の重みである。 Where α (1-y) y is

This is because the derivative of the sigmoid function is obtained. The elements operating as described above are connected as shown in FIG. The connection structure and connection weight in FIG. 12 are exactly the same as in FIG. From the right side of FIG. 12, an error b _j −t _j between the network output b _j obtained in Step 2 and the target output t _j is added. Based on this input, the element function shown in FIG. 14 works in each layer, and the element output z _j ^(m) is determined one after another. Thus, after obtaining the element output z _j ^(m) of each layer in the error back propagation mode, using this and the element output y _j ^(m) of each layer obtained in the output calculation mode (Step 2), Correct the weight.

w _ij ^{(m + 1)} is the weight of the connection line connecting the i-th element in the m-th layer and the j-th element in the (m + 1) -th layer.

The connection weight w _ij ^{(m + 1)} is corrected by the above method. Learning repeats the loop outside the learning procedure shown in FIG. 10 until the error evaluation scale (Equation 14) of all training data becomes sufficiently small.
The weight w _ij ^{(m + 1)} of the connection between each element of the artificial neural network in the optimum frequency and coil radial direction deviation calculation unit of the present invention is finally implemented by the learning method described above. .
In this way, the connection weight w _ij ^{(m + 1)} between the elements of the artificial neural network is finally determined by the learning method described above. Thus, the relational expression calculating step for calculating the optimal frequency related relational expression and the radial deviation amount relational expression of the coil ends.
Then, a relational expression including the following expression 10 including the connection weight term determined as described above and other expressions 11 to 13 for final calculation is stored in the relational expression storage unit 74. Note that individual relational expressions are stored for the optimum frequency and the amount of deviation in the radial direction of the coil.

In the blood pump device 1 of the present invention, the relational expression storage unit 74 stores optimum frequency relational expressions (10) to (13), and also relates to a deviation amount relational expression (10) to (13) in the radial direction of the coil. (13) (The formula is the same, but the connection weights are different) is stored.
Then, the transmission power control unit 76 of the blood pump device 1 determines the axial distance between the power transmission coil and the power reception coil that is actually applied, the load current value in the pump and the implantable device, the power transmission coil Using the voltage values applied to both ends and the above-described relational expressions (10) to (13) set, a function for calculating the optimum frequency and the amount of deviation in the radial direction of the coil is provided.

続いて、このs_j ^（1）を下記式のシグモイド関数を用いて、中間層素子の出力ｙ_j ^（１）を求める。

Specifically, using the axial coil distance between the power transmission coil and power reception coil, the load current value in the pump and the implantable device, and the voltage value applied to both ends of the power transmission coil From the equation (10), the weighted sum s _j ⁽¹⁾ of the intermediate layer element is calculated.

Subsequently, the output y _j ⁽¹⁾ of the intermediate layer element is obtained from this s _j ⁽¹⁾ using the sigmoid function of the following equation.

そして、下記式１３を用いて、出力層素子の出力y_k ^（２）（k＝1,2）を算出する。

Then, the weighted sum s _k ⁽²⁾ (k = 1, ^{2) of the} output layer elements is obtained using the following equation (12).

Then, the output y _k ⁽²⁾ (k = 1, ^{2) of} the output layer element is calculated using the following equation (13).

Therefore, this blood pump device receives as input the axial coil distance between the power transmission coil and the power reception coil, the load current value in the pump and the implantable device, and the voltage value applied to both ends of the power transmission coil. It has a function to calculate and output an optimum frequency and a deviation amount in the radial direction of the coil from a relational expression obtained using a layer feedforward artificial neural network.
In the embodiment described above, both the optimum frequency and the amount of deviation in the radial direction of the coil are calculated, but only the optimum frequency may be calculated.
By the above method, the three-layer feedforward that calculates and outputs the optimum frequency and the deviation in the radial direction of the coil by inputting the axial distance between the opposing coils, the load current, and the voltage applied to both ends of the extracorporeal coil. It is possible to provide an artificial heart power transmission system including an artificial neural network. The artificial heart power transmission system of the present invention further includes a display unit that specifically displays the amount of deviation in the radial direction of the coil calculated and output by the artificial neural network as a numerical value. In addition, since it is assumed that the power transmission efficiency is lowered and the overall operation of the system is affected as the amount of deviation in the radial direction of the coil increases, the power transmission system for artificial heart of the present invention has the amount of deviation of the coil. Informing means for informing that when the value of becomes larger.

FIG. 1 is a block diagram of an embodiment of a blood pump device of the present invention. FIG. 2 is a front view of an embodiment of the blood pump device of the present invention. FIG. 3 is a plan view of the blood pump device shown in FIG. 4 is a longitudinal sectional view of the blood pump device of the embodiment shown in FIG. FIG. 5 is a cross-sectional view of the blood pump device of FIG. FIG. 6 is a block diagram of another embodiment of the blood pump device of the present invention. FIG. 7 is an explanatory diagram for explaining a definitive analog model. FIG. 8 is an explanatory diagram for explaining the sigmoid function and the relationship between the intracellular potential and the impulse frequency. FIG. 9 is an explanatory diagram for explaining a feedforward neural network. FIG. 10 is an explanatory diagram for explaining the learning procedure of the sequential update learning method. FIG. 11 is an explanatory diagram for explaining a feedforward type neural network that learns by the error back propagation method. FIG. 12 is an explanatory diagram for explaining the signal flow in the error back-propagation mode. FIG. 13 is an explanatory diagram for explaining the operation of the element in the output calculation mode. FIG. 14 is an explanatory diagram for explaining the element function in the error back-propagation mode.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Blood pump apparatus 2 Blood pump part 3 Impeller rotational torque generation part 4 Impeller position control part 5 Pump 6 Implantation side apparatus 7 Outside body apparatus 21 Impeller 25 Magnetic body 31 Rotor 34 Motor 41 Electromagnet 59 Display part 60 Relational expression memory | storage part 61 Power supply unit 64 Current value detection unit 62 Power information transmission unit 74 Relational expression storage unit 76 Transmission power control unit

Claims (6)

A pump including a housing having a blood inflow port and a blood outflow port; an impeller that rotates in the housing and feeds blood; and an impeller rotational torque generator for rotating the impeller; and the impeller rotational torque An in-vivo implantable device embedded in a living body comprising a power supply circuit for driving the generator, an electric power transmission unit comprising an electric power transmission coil for transmitting electric power to the in-vivo implantable device, and A blood pump device comprising a voltage detection unit for detecting a voltage applied to both ends, an external body device that includes a transmission power control unit and is not connected to the implantable device,
The implantable device includes a power receiving coil that receives power from the power transmitting coil and detects a load current sent from the power source to the power circuit. A current value detection unit, and a power information transmission unit that transmits the current value detected by the current value detection unit to the external device,
The extracorporeal device includes an inter-coil distance input unit that inputs an axial distance between the power transmission coil and the power reception coil, and a power information reception unit that receives information from the power information transmission unit,
The transmission power control unit is an axial distance between the power transmission coil and the power reception coil at an optimal frequency of a number of power transmission coils measured in advance, a load current of the implantable device, The voltage applied to both ends of the power transmission coil is used as input layer data, and the optimal frequency of the power transmission coil related to each input layer data is used as teacher data, and the relational expression between the input layer data calculated using a neural network and the optimal frequency A relational expression storage unit for storing the optimum frequency related relational expression, the optimum frequency related relational expression, the inter-coil distance input to the inter-coil distance input part, the load current value received by the power information receiving part, Optimal frequency calculation function that calculates the optimal frequency of the coil for power transmission using the coil voltage value for power transmission detected by the coil voltage detector for power transmission , The optimum the optimum frequency calculated by the frequency computing function, blood pump apparatus characterized by comprising a frequency control function to perform the power supply by the power transmission coil. The extracorporeal device includes a display unit, and the transmission power control unit includes a plurality of previously measured power transmission coils and power reception coils between the power transmission coil and the power reception coil in a radial deviation amount between the power transmission coils and the power reception coil. The distance between the coils in the axial direction, the load current in the implantable device, and the voltage applied to both ends of the power transmission coil are used as input layer data, and the radial deviation between the coils related to each input layer data is As a teacher data, a relational expression storage unit for storing a relational expression related to the amount of radial displacement between coils, which is a relational expression between the input layer data calculated using a neural network and the amount of radial deviation between coils, and the radial direction between the coils Deviation relational expression, inter-coil distance input to the inter-coil distance input unit, load current value received by the power information receiving unit, detected by the transmitting coil voltage detection unit The inter-coil radial displacement amount calculation function for calculating the inter-coil radial displacement amount using the transmitted coil voltage value, and the inter-coil radial displacement amount calculated by the inter-coil radial displacement amount calculation function. The blood pump device according to claim 1, further comprising: a function of displaying on the display unit. The outside body device includes an alarm unit, and the transmission power control unit stores an allowable amount of radial deviation between coils and calculates an amount of radial deviation between coils calculated by the radial deviation amount calculation function between coils. The blood pump device according to claim 2, further comprising a warning function that activates the warning means when the value is larger than a stored allowable radial deviation between coils. The method for manufacturing a blood pump device according to any one of claims 1 to 3, wherein an axial coil interval between the power transmission coil and the power reception coil is measured at an optimal frequency of a large number of power transmission coils measured in advance. The distance, the load current of the implantable device and the voltage applied to both ends of the power transmission coil are used as input layer data, and the optimal frequency of the power transmission coil related to each input layer data is used as teacher data, A relational expression calculating step for calculating an optimal frequency related relational expression that is a relational expression between the input layer data and the optimal frequency, and the optimal frequency related relational expression obtained by the optimal frequency related relational expression calculating step. A method of manufacturing a blood pump device, comprising: storing the blood pump device so that the control unit can be used. 4. The method for manufacturing a blood pump device according to claim 2, wherein the power transmission coil and the power reception in a radial deviation amount between a number of the power transmission coils and the power reception coil measured in advance. The axial distance between the coils for the coil, the load current in the implantable device, and the voltage applied to both ends of the power transmission coil as input layer data, and the radial direction between the coils for each input layer data Using the neural network as the deviation amount as the teacher data, the inter-coil radial deviation amount relational expression for calculating the inter-coil radial deviation amount relational expression, which is the relational expression between the input layer data and the inter-coil radial deviation amount The calculation step and the inter-coil radial deviation amount relational expression obtained in the inter-coil radial deviation amount relational expression are calculated so that the electric power control unit can use the blood. Manufacturing method of a blood pump apparatus characterized by performing a storage step of storing in the pump device. 6. The method of manufacturing a blood pump device according to claim 4, wherein the neural network is a three-layer feedforward neural network including three layers of an input layer, an intermediate layer, and an output layer.

Images