阅读说明：本技术 超声设备降功耗的方法和装置 (Method and device for reducing power consumption of ultrasonic equipment ) 是由 顾菊春 殷晨 赵明昌 万永庆 党福德 于 2019-10-22 设计创作，主要内容包括：本发明涉及超声设备技术领域,具体公开了一种超声设备降功耗的方法和装置。所述方法包括：获取超声图像；获取惯性传感器的传感数据；根据所述超声图像、所述传感数据以及图像分类模型,确定超声探头的工作状态；所述图像分类模型为根据样本超声图像和惯性传感器的传感数据训练得到的用于判断超声探头的工作状态的模型；所述样本超声图像包括超声探头检测被检体时的图像,以及所述超声探头闲置时的图像；若所述工作状态为所述超声探头闲置,则控制所述超声设备进入低功耗模式。解决了现有方案中超声探头功耗高,会发热影响超声探头寿命的问题；达到了可以降低超声设备所需耗费的功耗的效果。(The invention relates to the technical field of ultrasonic equipment, and particularly discloses a method and a device for reducing power consumption of ultrasonic equipment. The method comprises the following steps: acquiring an ultrasonic image; acquiring sensing data of an inertial sensor; determining the working state of an ultrasonic probe according to the ultrasonic image, the sensing data and the image classification model; the image classification model is a model for judging the working state of the ultrasonic probe, which is obtained by training according to the sample ultrasonic image and the sensing data of the inertial sensor; the sample ultrasonic image comprises an image when an ultrasonic probe detects a detected body and an image when the ultrasonic probe is idle; and if the working state is that the ultrasonic probe is idle, controlling the ultrasonic equipment to enter a low power consumption mode. The problem that the service life of the ultrasonic probe is influenced due to the fact that the ultrasonic probe is high in power consumption in the existing scheme is solved; the effect of reducing the power consumption required by the ultrasonic equipment is achieved.)

1. A method for reducing power consumption of an ultrasonic device, comprising:

acquiring an ultrasonic image;

acquiring sensing data of an inertial sensor;

determining the working state of an ultrasonic probe according to the ultrasonic image, the sensing data and the image classification model; the image classification model is a model for judging the working state of the ultrasonic probe, which is obtained by training according to the sample ultrasonic image and the sensing data of the inertial sensor; the sample ultrasonic image comprises an image when an ultrasonic probe detects a detected body and an image when the ultrasonic probe is idle;

and if the working state is that the ultrasonic probe is idle, controlling the ultrasonic equipment to enter a low power consumption mode.

2. The method of claim 1, wherein determining the operational state of the ultrasound probe from the ultrasound images, the sensing data, and the image classification model comprises:

extracting image features in the ultrasonic image through a feature extraction layer of the image classification model;

and fusing the image characteristics and the sensing data through a convolution regression layer of the image classification model, and regressing to obtain the working state of the ultrasonic probe.

3. The method of claim 2,

extracting data features in the sensing data through a neural network before fusing the image features and the sensing data;

the fusion of the image features and the sensing data through the convolution regression layer of the image classification model and the regression to obtain the working state of the ultrasonic probe comprise:

and fusing the image characteristics and the data characteristics through a convolution regression layer of the image classification model, and regressing to obtain the working state of the ultrasonic probe.

4. The method of any of claims 1 to 3, further comprising:

when the ultrasonic probe is in the low power consumption mode, acquiring sensing data of the inertial sensor;

controlling the ultrasound device to switch to a high power consumption mode when the sensing data indicates that the ultrasound probe starts to move.

5. The method of any of claims 1 to 3, wherein said controlling the ultrasound device to enter a low power consumption mode comprises:

acquiring the duration of the ultrasonic probe in an idle state;

and controlling the ultrasonic equipment to enter a low power consumption mode when the duration reaches a time threshold.

6. The method of any of claims 1 to 3, further comprising:

and when the ultrasonic equipment is in a low power consumption mode, the step of determining the working state of the ultrasonic probe according to the ultrasonic image and the image classification model is executed again at preset time intervals.

7. The method of claim 6, further comprising:

acquiring a target department to which the ultrasonic probe is currently applied;

and inquiring the preset time interval corresponding to the target department according to the corresponding relation between departments and time intervals.

8. A classification model training method, the method comprising:

acquiring a sample ultrasonic image, wherein the sample ultrasonic image comprises an image of an ultrasonic probe when the ultrasonic probe detects a detected object and an image of the ultrasonic probe when the ultrasonic probe is idle;

acquiring sensing data of an inertial sensor in an ultrasonic probe when a sample ultrasonic image is obtained;

and training an initialization model according to the sample ultrasonic images and the sensing data corresponding to each sample ultrasonic image to obtain an image classification model, wherein the image classification model is used for determining the working state of the ultrasonic probe according to the ultrasonic images and the sensing data.

9. An apparatus for reducing power consumption of an ultrasound device, comprising a memory and a processor, wherein at least one program instruction is stored in the memory, and wherein the processor is configured to load and execute the at least one program instruction to implement the method of any one of claims 1 to 7, or 8.

10. A computer storage medium having stored therein at least one program instruction which is loaded and executed by a processor to implement a method as claimed in any one of claims 1 to 7, or 8.

Technical Field

The invention relates to the technical field of ultrasonic equipment, in particular to a method and a device for reducing power consumption of ultrasonic equipment.

Background

An ultrasound probe is an important component in an ultrasound apparatus, and is used for transmitting sound waves, receiving echoes of the sound waves after passing through a subject, and then transmitting the echoes to an ultrasound host for processing. In the existing scheme, when the ultrasonic equipment is started, the ultrasonic probe is always in a standby state, the power consumption of the ultrasonic probe is high, certain electric energy is consumed, and when the ultrasonic probe is in the standby state for a long time, the ultrasonic probe is also heated, so that the service life of the ultrasonic probe is influenced.

Disclosure of Invention

The invention aims to at least solve one of the technical problems in the prior art, and provides a method and a device for reducing power consumption of ultrasonic equipment, so as to solve the problems that the service life of an ultrasonic probe is influenced due to heating caused by high power consumption of the ultrasonic probe in the prior scheme; the effect of reducing the power consumption required by the ultrasonic equipment is achieved.

The invention provides a method for reducing power consumption of an ultrasonic device, which comprises the following steps:

acquiring an ultrasonic image;

acquiring sensing data of an inertial sensor;

determining the working state of an ultrasonic probe according to the ultrasonic image, the sensing data and the image classification model; the image classification model is a model which is obtained according to the training of the sample ultrasonic image and is used for judging the working state of the ultrasonic probe; the sample ultrasonic image comprises an image when an ultrasonic probe detects a detected body and an image when the ultrasonic probe is idle;

and if the working state is that the ultrasonic probe is idle, controlling the ultrasonic equipment to enter a low power consumption mode.

Optionally, the determining the working state of the ultrasound probe according to the ultrasound image, the sensing data and the image classification model includes:

extracting image features in the ultrasonic image through a feature extraction layer of the image classification model;

and fusing the image characteristics and the sensing data through a convolution regression layer of the image classification model, and regressing to obtain the working state of the ultrasonic probe.

Optionally, before fusing the image features and the sensing data, extracting data features in the sensing data through a neural network;

the fusion of the image features and the sensing data through the convolution regression layer of the image classification model and the regression to obtain the working state of the ultrasonic probe comprise:

and fusing the image characteristics and the data characteristics through a convolution regression layer of the image classification model, and regressing to obtain the working state of the ultrasonic probe.

Optionally, the method further includes:

when the ultrasonic probe is in the low power consumption mode, acquiring sensing data of the inertial sensor;

controlling the ultrasound device to switch to a high power consumption mode when the sensing data indicates that the ultrasound probe starts to move.

Optionally, the controlling the ultrasound probe to enter the low power consumption mode includes:

acquiring the duration of the ultrasonic probe in an idle state;

and controlling the ultrasonic equipment to enter a low power consumption mode when the duration reaches a time threshold.

Optionally, the method further includes:

and when the ultrasonic probe is in a low power consumption mode, the step of determining the working state of the ultrasonic probe according to the ultrasonic image and the image classification model is executed again at preset time intervals.

Optionally, the method further includes:

acquiring a target department to which the ultrasonic probe is currently applied;

and inquiring the preset time interval corresponding to the target department according to the corresponding relation between departments and time intervals.

In a second aspect, a classification model training method is provided, the method including:

acquiring a sample ultrasonic image, wherein the sample ultrasonic image comprises an image of an ultrasonic probe when the ultrasonic probe detects a detected object and an image of the ultrasonic probe when the ultrasonic probe is idle;

acquiring sensing data of an inertial sensor in an ultrasonic probe when a sample ultrasonic image is obtained;

and training an initialization model according to the sample ultrasonic images and the sensing data corresponding to each sample ultrasonic image to obtain an image classification model, wherein the image classification model is used for determining the working state of the ultrasonic probe according to the ultrasonic images and the sensing data.

In a third aspect, there is provided a power consumption reduction apparatus for an ultrasound device, the apparatus comprising a memory and a processor, the memory having at least one program instruction stored therein, and the processor implementing the method according to the first or second aspect by loading and executing the at least one program instruction.

In a fourth aspect, there is provided a computer storage medium having stored therein at least one program instruction which is loaded and executed by a processor to implement a method according to the first or second aspect.

The current working state of the ultrasonic probe is judged through the image classification model, and then the ultrasonic probe is controlled to enter a low-power-consumption mode when the ultrasonic probe is idle, so that the power consumption of ultrasonic equipment is reduced, and the problem that the service life of the ultrasonic probe is influenced due to the fact that the ultrasonic probe is high in power consumption in the existing scheme is solved; the effect of reducing the power consumption required by the ultrasonic equipment is achieved.

Drawings

FIG. 1 is a flow chart of a method for reducing power consumption of an ultrasound device according to the present invention. Fig. 2 is an ultrasonic image of an ultrasonic echo signal when the propagation medium of the present invention is air.

Fig. 3 is an ultrasonic image of an ultrasonic echo signal when the propagation medium of the present invention is human tissue.

Fig. 4-6 are schematic diagrams of the training principle of the image classification model of the present invention.

FIG. 7 is a flowchart illustrating a control flow of the power consumption reduction method according to the present invention.

FIG. 8 is another schematic diagram of the training principle of the image classification model of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the following detailed description and accompanying drawings. Wherein like elements in different embodiments are numbered with like associated elements. In the following description, numerous details are set forth in order to provide a better understanding of the present application. However, those skilled in the art will readily recognize that some of the features may be omitted or replaced with other elements, materials, methods in different instances. In some instances, certain operations related to the present application have not been shown or described in detail in order to avoid obscuring the core of the present application from excessive description, and it is not necessary for those skilled in the art to describe these operations in detail, so that they may be fully understood from the description in the specification and the general knowledge in the art. Furthermore, the features, operations, or characteristics described in the specification may be combined in any suitable manner to form various embodiments. Also, the various steps or actions in the method descriptions may be transposed or transposed in order, as will be apparent to one of ordinary skill in the art. Thus, the various sequences in the specification and drawings are for the purpose of describing certain embodiments only and are not intended to imply a required sequence unless otherwise indicated where such sequence must be followed. The numbering of the components as such, e.g., "first", "second", etc., is used herein only to distinguish the objects as described, and does not have any sequential or technical meaning.

Referring to fig. 1, a method for reducing power consumption of an ultrasound apparatus according to an embodiment of the present application is shown, as shown in fig. 1, the method includes:

step 101, an ultrasound image is obtained.

When the ultrasonic device is started, the ultrasonic probe can transmit an ultrasonic signal at a certain frequency. The frequency of the ultrasonic signal is determined by the ultrasonic equipment and the current examination mode of the ultrasonic equipment. When the current examination mode of the ultrasonic equipment is the liver, the frequency of the ultrasonic signal is 4.7Mhz (harmonic wave) and 4.4Mhz (fundamental wave); when the examination mode is heart, the frequency of the ultrasonic signal is 3.3Mhz (harmonic wave), 2.7Mhz (fundamental wave); this embodiment is not limited to this.

Accordingly, after the ultrasonic signal is emitted, the ultrasonic probe may receive the ultrasonic echo signal. After the ultrasonic echo signal is received, an ultrasonic image can be generated according to the ultrasonic echo signal, and the step can obtain the ultrasonic image.

Step 102, acquiring sensing data of the inertial sensor.

When the ultrasonic probe is provided with the inertial sensor, sensing data of the inertial sensor can be acquired in real time.

103, determining the working state of the ultrasonic probe according to the ultrasonic image, the sensing data and the image classification model; the image classification model is a model for judging the working state of the ultrasonic probe, which is obtained by training according to the sample ultrasonic image and the sensing data of the inertial sensor; the sample ultrasound image includes an image when an ultrasound probe detects a subject and an image when the ultrasound probe is idle.

Wherein, the current working state of the ultrasonic probe is as follows: the ultrasonic probe is used for drawing, or the ultrasonic probe is idle.

After obtaining the ultrasound image and the sensing data, the ultrasound device may input the ultrasound image and the sensing data into the image classification model and determine an output of the image classification model as a current operating state of the ultrasound probe. In practical implementation, after the ultrasound image and the sensing data are input into the image classification model, the image classification model may output a probability, such as a probability that the ultrasound probe is being imaged or a probability that the ultrasound probe is idle; and correspondingly, the working state of the ultrasonic probe can be determined according to the probability output by the image classification model. For example, when the output result of the image classification model is that the probability of mapping is 40%, the working state of the ultrasound probe can be determined to be idle.

Of course, in order to avoid that the ultrasound apparatus cannot be used normally due to misjudgment of the image classification model, in actual implementation, the working state may be determined as idle when the probability of the ultrasound probe being idle exceeds a certain threshold, for example, exceeds 80%, and otherwise, the ultrasound probe is still determined as working. For example, when the probability of the idle ultrasound probe is 45%, the working state of the ultrasound probe is still determined to be working.

The image classification model is a model for judging the working state of the ultrasonic probe, which is obtained by training according to the sample ultrasonic image and the sensing data. The training method of the image classification model comprises the following steps:

first, a sample ultrasound image is acquired, which includes an image when an ultrasound probe detects a subject and an image when the ultrasound probe is idle.

In practical implementation, in order to enable the trained image classification model to determine the working state of the ultrasound probe when the ultrasound probe maps different parts, the ultrasound image for detecting the object includes images of the ultrasound probe when the ultrasound probe detects different parts in the object, and in consideration of the difference of the objects to which the ultrasound probe is applied, the ultrasound image in this embodiment is an image for detecting different parts in the object to be detected by the ultrasound probe. For example, if the ultrasound probe is a probe for animals, the ultrasound image includes images for detecting different parts of different animals; if the ultrasound probe is a probe for detecting a human body, the ultrasound image includes images for detecting different parts in the human body. In practice, different types of ultrasound probes can be connected to the same ultrasound device, so that the selected sample ultrasound image can cover ultrasound images of different types of ultrasound probes, which is not described herein again.

Referring to fig. 2, an ultrasound image corresponding to an ultrasound echo signal is shown when propagating in air. When the ultrasound probe detects the object, the ultrasound echo signals are greatly different at different positions under the influence of the tissues in the object, for example, refer to fig. 3, which shows a possible ultrasound image corresponding to the ultrasound echo signals when the object is a human body.

And secondly, acquiring sensing data of an inertial sensor in the ultrasonic probe when the sample ultrasonic image is obtained.

Thirdly, training an initialization classification model according to the sample ultrasonic image and the sensing data to obtain an image classification model, wherein the image classification model is used for determining the current working state of the ultrasonic probe according to the input ultrasonic image.

Optionally, the sample ultrasound image may be pre-processed, the pre-processing including at least one of desensitization, cropping, scaling, and enhancement. And before the classification model is initialized in the sample ultrasonic image village, normalization processing can be carried out on the sample ultrasonic image.

The initialized classification model may be a multi-layer convolutional neural network, and the multi-layer convolutional neural network may be a classification neural network. The shallow layer of the classification neural network is used for extracting the image characteristics of the ultrasonic image, and the deep layer is used for fusing the image characteristics of the shallow layer network and the sensing data to regress the working state of the ultrasonic probe, namely, the output of the trained image classification model is the working state of the ultrasonic probe.

And inputting the normalized ultrasonic image into the multilayer convolution neural network, and stopping training until loss is minimum.

Specifically, the deep layer part of the multilayer neural network is used for fusing the characteristics extracted by the shallow layer network and the sensing data of the inertial sensor, so as to obtain the working state of the ultrasonic probe. Referring to fig. 4 and 5, schematic diagrams of image classification models are shown.

Correspondingly, when the working state of the ultrasonic probe is determined through the trained image classification model, the image features of the ultrasonic image are correspondingly extracted through the feature extraction layer of the image classification model, and the image features are fused and regressed to obtain the working state of the ultrasonic probe.

In practical implementation, please refer to fig. 6, before the image features and the sensing data are fused, the data features of the sensing data may be extracted through a neural network, so as to fuse the image features and the data features, and the working state of the ultrasonic probe is obtained through regression. Correspondingly, when the working state of the ultrasonic probe is determined through the trained image classification model, the image characteristics of the ultrasonic image are correspondingly extracted through the characteristic extraction layer of the image classification model, the data characteristics of the sensing data are extracted through the neural network, and the image characteristics and the sensing data are fused and regressed to obtain the working state of the ultrasonic probe.

And 104, if the working state is that the ultrasonic probe is idle, controlling the ultrasonic equipment to enter a low power consumption mode.

The output of the image classification model is the working state of the ultrasonic probe, and when the output result is that the ultrasonic probe is idle, the ultrasonic probe can be controlled to enter a low power consumption mode in order to reduce the power consumption required by ultrasonic equipment.

And when the output of the image classification model is that the working state of the ultrasonic probe is that the ultrasonic probe is drawing, the ultrasonic equipment can be controlled to be kept in a high power consumption mode.

In practical implementation, medical personnel can switch the working state of the ultrasonic probe at will according to the use requirements of the medical personnel, for example, when the ultrasonic probe is in a low power consumption mode, the medical personnel can trigger thawing, so that the ultrasonic equipment enters a high power consumption mode; similarly, when the ultrasound device is in a high power mode, the medical personnel may also trigger a freeze, thereby causing the ultrasound device to enter a low power mode. That is, please refer to fig. 4, which shows a control flow chart of the method for reducing power consumption of the ultrasonic apparatus.

And when the working state is that the ultrasonic probe is drawing, the ultrasonic equipment is controlled to enter a high power consumption mode.

In summary, the current working state of the ultrasonic probe is judged through the image classification model, and then when the ultrasonic probe is idle, the ultrasonic equipment is controlled to enter a low-power mode, so that the power consumption of the ultrasonic equipment is reduced, and the problem that the service life of the ultrasonic probe is influenced due to the fact that the ultrasonic probe is high in power consumption in the existing scheme is solved; the effect of reducing the power consumption required by the ultrasonic equipment is achieved.

In actual implementation, when the ultrasonic probe is provided with the inertial sensor, sensing data acquired by the inertial sensor can be acquired, and medical personnel can pick up the ultrasonic probe when using the ultrasonic probe, namely the ultrasonic probe can move, so that the acquired sensing data indicate that the movement mode of the ultrasonic probe accords with the movement mode of the ultrasonic probe when the medical personnel takes pictures, namely the sensing data indicate that the ultrasonic probe is ready to take pictures, and the ultrasonic equipment can be controlled to enter a high-power-consumption mode.

It should be added that, the above is exemplified by determining the working state of the ultrasound probe according to the ultrasound image and the sensing data of the inertial sensor only by the image classification model, when the image classification model is obtained according to the ultrasound image training in practical application, the ultrasound device may also directly input the ultrasound image to the image classification model and obtain the working state, specifically, referring to fig. 8, the image feature in the ultrasound image may be extracted by the feature extraction layer in the image classification model, and then the working state of the ultrasound probe may be obtained according to the image feature regression. That is, the working state can be obtained only by acquiring the ultrasound image, and the detailed description is omitted here. And the result determined by the image classification model according to the ultrasonic image and the result detected by the inertial sensor can be combined to obtain the working state of the ultrasonic probe. For example, when it is determined that the ultrasonic probe is idle according to the image classification model, but the sensing data detected by the inertial sensor indicates that the motion state of the ultrasonic probe does not conform to the motion state of the ultrasonic probe when the ultrasonic probe is idle, the working state of the ultrasonic probe is determined to be idle. And in order to prevent the ultrasonic probe from being incapable of being normally used due to misjudgment, in actual implementation, the working state of the ultrasonic probe is determined as mapping only when the output result of the image classification model and the sensing data of the inertial sensor both represent that the ultrasonic probe is mapping, and otherwise, the working state of the ultrasonic probe is determined as idle.

In addition, in order to switch from the low power consumption mode to the high power consumption mode in time when the medical staff needs to use the ultrasound probe, after the ultrasound device is in the low power consumption mode, the step of acquiring the ultrasound image may be performed at preset time intervals, that is, the above steps 101 to 104 are performed again.

The preset time interval can be a time interval preset by a system or a time interval self-defined by medical staff. For example, the preset time interval is a time interval defined by medical staff by a user, when the ultrasonic probe is used for examining a fetus, because a pregnant woman needs to get off the bed after the examination is finished, a next pregnant woman gets on the bed and preparation before the examination needs a certain time, after the ultrasonic probe enters the low power consumption mode, the ultrasonic probe can emit ultrasonic signals again after 2 minutes.

Optionally, the preset time interval may be further obtained by:

first, a target department to which the ultrasound probe is currently applied is acquired.

Such as gynecology, obstetrics, nephrology, etc.

Secondly, inquiring the preset time interval corresponding to the target department according to the corresponding relation between departments and time intervals.

By the scheme, when medical personnel prepare to print pictures, the ultrasonic equipment can be awakened in time, and normal use of the ultrasonic equipment is guaranteed.

In addition, since the ultrasound device is generally used in a hospital, and the working hours of the hospital are generally fixed, for example, 8:30-12:00,13:00-17:30, the ultrasound device may also perform the step of transmitting the ultrasound signal only during the working time period, and the above-mentioned procedure is not performed during other time periods, thereby further reducing the power consumption of the ultrasound device. The working time period is a preset time period, and can also be a self-defined time period for medical staff. For example, medical personnel can customize summer work and winter work and rest according to hospital regulations.

An embodiment of the present application further provides a power consumption reduction apparatus for an ultrasound device, the apparatus includes a memory and a processor, the memory stores at least one program instruction, and the processor loads and executes the at least one program instruction to implement the method described above.

A fourth aspect of the present invention provides a computer-readable storage medium, in which a computer program is stored, and the computer program is used for implementing the steps of reducing the power consumption of the ultrasound apparatus when being executed by a processor.

A fourth aspect of the invention provides a computer readable storage medium having stored thereon a computer program for performing the steps of reducing power consumption of an ultrasound device when executed by a processor.

Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to examples, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.