CN113627492A

CN113627492A - Method for determining size of scanning object, scanning method, scanning device and electronic equipment

Info

Publication number: CN113627492A
Application number: CN202110834182.2A
Authority: CN
Inventors: 逄岭; 郑凌
Original assignee: Neusoft Medical Systems Co Ltd
Current assignee: Neusoft Medical Systems Co Ltd
Priority date: 2021-07-20
Filing date: 2021-07-20
Publication date: 2021-11-09

Abstract

The disclosure relates to a method for determining the size of a scanned object, a scanning method, a scanning device and electronic equipment, and relates to the technical field of electronic information processing, wherein the method comprises the following steps: the method comprises the steps of obtaining plain film information obtained by scanning a target scanning object by scanning equipment according to the direction of a first coordinate axis, wherein the plain film information is plain film images or plain film scanning data, and determining first length information of the target scanning object in the direction of the first coordinate axis through a pre-trained size recognition model according to the plain film information. The method and the device can accurately identify the length of the scanning object in the scanning direction according to the plain film information of the scanning object on the premise of not increasing the scanning dose and extra measuring equipment.

Description

Method for determining size of scanning object, scanning method, scanning device and electronic equipment

Technical Field

The present disclosure relates to the field of electronic information processing technologies, and in particular, to a method for determining a size of a scanned object, a scanning method, an apparatus, and an electronic device.

Background

With the continuous development of image processing technology, X-ray scanning devices are widely used in the medical field, for example: CT (english: Computed Tomography, chinese: Computed Tomography) equipment, CR (english: Computed Radiography, chinese: Computed Radiography) equipment, DR (english: Digital Radiography, chinese: Digital Radiography) equipment, and the like. The X-ray scanning device utilizes precisely collimated X-rays to irradiate a scanning object, and the intensity of the X-rays transmitted through the scanning object is received by a detector so as to obtain a scanning image of the scanning object. The X-ray scanning equipment has the characteristics of short scanning time, clear image and the like.

In order to ensure that the quality of the scanned image is stable and the scanning dose received by each part of the scanned object is uniform, the thickness of the scanned object needs to be determined first, so that the scanning center and the scanning intensity of the scanning device are adjusted according to the thickness of the scanned object. Usually, an image acquisition device (such as a camera) or a calibration device may be added to the scanning device to measure the thickness of the scanned object in advance, however, the additional device increases the cost and the deployment difficulty, and the measurement accuracy is difficult to guarantee.

Disclosure of Invention

The present disclosure is directed to a method for determining a size of a scanned object, a scanning method, a scanning apparatus, and an electronic device, so as to solve the related problems in the prior art.

In order to achieve the above object, according to a first aspect of embodiments of the present disclosure, there is provided a method of determining a size of a scan object, the method including:

acquiring plain film information obtained by scanning a target scanning object by scanning equipment according to the direction of a first coordinate axis, wherein the plain film information is plain film images or plain film scanning data;

and according to the plain film information, determining first length information of the target scanning object in the first coordinate axis direction through a pre-trained size recognition model.

Optionally, the determining, according to the plain film information, first length information of the target scanning object in the first coordinate axis direction by a pre-trained size recognition model includes:

inputting the plain film information into the size recognition model to obtain the first length information output by the size recognition model, wherein the first length information comprises: and the target area is obtained by dividing according to the projection of the target scanning object on a target plane, and the target plane is vertical to the first coordinate axis.

Optionally, the method further comprises:

determining a second length of each target area in a second coordinate axis direction according to the plain film information, wherein the second coordinate axis is perpendicular to the first coordinate axis;

and for each target area, determining the volume of the target area according to a preset division interval, the first length of the target area in the direction of the first coordinate axis, and the second length of the target area in the direction of the second coordinate axis, wherein the division interval is used for indicating the length of the target area in the direction of a third coordinate axis, and the third coordinate axis is perpendicular to the first coordinate axis and the second coordinate axis respectively.

Optionally, the size recognition model is trained by:

acquiring a sample input set and a sample output set, wherein the sample input set comprises sample plain film information of a plurality of samples scanned in the first coordinate axis direction, and the sample output set comprises first actual length information corresponding to the plurality of samples in the first coordinate axis direction;

and taking the sample input set as the input of the size recognition model, and taking the sample output set as the output of the size recognition model so as to train the size recognition model.

Optionally, the sample output set includes a plurality of sample regions corresponding to the samples, each sample region having a first actual length in the first coordinate axis direction, the sample regions being obtained by dividing according to a projection of the corresponding sample on a target plane, the target plane being perpendicular to the first coordinate axis;

the training the size recognition model by using the sample input set as the input of the size recognition model and the sample output set as the output of the size recognition model comprises:

inputting sample plain film information of any sample into the size recognition model to obtain an initial length of each sample region of the sample in the first coordinate axis direction, wherein the initial length is output by the size recognition model;

training the size recognition model based on the initial length of each of the sample regions of the sample and the first actual length of each of the sample regions of the sample.

Optionally, after the obtaining of the sample input set and the sample output set, the size recognition model is further trained by:

adjusting each piece of sample slide information in the sample input set according to a preset rule, and generating a plurality of pieces of slide expansion information corresponding to the sample slide information so as to expand the sample input set;

expanding the sample output set according to the expanded sample input set;

and taking the sample input set after the expansion as the input of the size recognition model, and taking the sample output set after the expansion as the output of the size recognition model so as to train the size recognition model.

According to a second aspect of the embodiments of the present disclosure, there is provided a scanning method, including:

according to the method for determining the size of the scanning object provided by the first aspect of the embodiment of the present disclosure, first length information of a target scanning object in a first coordinate axis direction is determined;

determining the geometric center of the target scanning object according to the first length information;

according to the geometric center of the target scanning object, the geometric center of the target scanning object is adjusted by controlling the height of a scanning bed of scanning equipment, so that the coordinate of the adjusted geometric center of the target scanning object in the first coordinate axis direction is the same as the coordinate of the scanning center of the scanning equipment in the first coordinate axis direction;

and controlling the scanning equipment to carry out three-dimensional scanning on the target scanning object.

According to a third aspect of the embodiments of the present disclosure, there is provided a scanning method, including:

according to the method for determining the size of the scanning object provided by the first aspect of the embodiment of the present disclosure, the volume of at least one target area on the target scanning object is determined:

determining the scanning intensity corresponding to each target area according to the volume of the target area;

and controlling the scanning equipment to perform three-dimensional scanning on the target area according to the scanning intensity corresponding to each target area.

According to a fourth aspect of the embodiments of the present disclosure, there is provided an apparatus for determining a size of a scan object, the apparatus comprising:

the acquisition module is used for acquiring plain film information obtained by scanning a target scanning object by scanning equipment according to the direction of a first coordinate axis, wherein the plain film information is plain film images or plain film scanning data;

and the first determining module is used for determining first length information of the target scanning object in the first coordinate axis direction through a pre-trained size recognition model according to the plain film information.

Optionally, the first determining module is configured to:

Optionally, the apparatus further comprises:

the second determining module is used for determining a second length of each target area in a second coordinate axis direction according to the plain film information, wherein the second coordinate axis is perpendicular to the first coordinate axis; and for each target area, determining the volume of the target area according to a preset division interval, the first length of the target area in the direction of the first coordinate axis, and the second length of the target area in the direction of the second coordinate axis, wherein the division interval is used for indicating the length of the target area in the direction of a third coordinate axis, and the third coordinate axis is perpendicular to the first coordinate axis and the second coordinate axis respectively.

Optionally, the size recognition model is trained by:

acquiring a sample input set and a sample output set, wherein the sample input set comprises sample plain film information of a plurality of samples scanned in a first coordinate axis direction, and the sample output set comprises first actual length information corresponding to the plurality of samples in the first coordinate axis direction;

expanding the sample output set according to the expanded sample input set;

According to a fifth aspect of embodiments of the present disclosure, there is provided a scanning apparatus including:

the first determining module is used for determining first length information of the target scanning object in the first coordinate axis direction through a pre-trained size recognition model according to the plain film information;

the second determining module is used for determining the geometric center of the target scanning object according to the first length information;

the adjusting module is used for adjusting the geometric center of the target scanning object by controlling the height of a scanning bed of the scanning equipment according to the geometric center of the target scanning object, so that the coordinate of the adjusted geometric center of the target scanning object on the first coordinate axis is the same as the coordinate of the scanning center of the scanning equipment on the first coordinate axis;

and the first control module is used for controlling the scanning equipment to carry out three-dimensional scanning on the target scanning object.

According to a sixth aspect of the embodiments of the present disclosure, there is provided a scanning apparatus including:

a third determining module, configured to determine, according to the volume of each target region, a scanning intensity corresponding to the target region;

and the second control module is used for controlling the scanning equipment to carry out three-dimensional scanning on the target area according to the scanning intensity corresponding to each target area.

According to a seventh aspect of embodiments of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the method of the first, second or third aspect of an embodiment of the present disclosure.

According to an eighth aspect of embodiments of the present disclosure, there is provided an electronic apparatus including:

a memory having a computer program stored thereon;

a processor for executing the computer program in the memory to implement the steps of the method of the first, second or third aspect of an embodiment of the present disclosure.

According to the technical scheme, the method comprises the steps that firstly, plain film information obtained by scanning a target scanning object by scanning equipment according to the direction of a first coordinate axis is obtained, wherein the plain film information can be plain film images or plain film scanning data. And then, according to the plain film information, determining first length information of the target scanning object in the direction of the first coordinate axis through a pre-trained size recognition model. The method and the device can accurately identify the length of the scanning object in the scanning direction according to the plain film information of the scanning object on the premise of not increasing the scanning dose and extra measuring equipment.

Additional features and advantages of the disclosure will be set forth in the detailed description which follows.

Drawings

The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:

FIG. 1 is a schematic diagram of a scanning device shown in accordance with an exemplary embodiment;

FIG. 2 is a flow chart illustrating a method of determining a size of a scanned object in accordance with an exemplary embodiment;

FIG. 3 is a projection of a target scan object onto a target plane, shown in accordance with an exemplary embodiment;

FIG. 4 is a flow chart illustrating another method of scan object size determination in accordance with an exemplary embodiment;

FIG. 5 is a flow diagram illustrating a method of training a size recognition model in accordance with an exemplary embodiment;

FIG. 6 is a flow diagram illustrating another method of training a size recognition model in accordance with an illustrative embodiment;

FIG. 7 is a flow diagram illustrating another method of training a size recognition model in accordance with an illustrative embodiment;

FIG. 8 is a flow chart illustrating a scanning method in accordance with an exemplary embodiment;

FIG. 9 is a flow chart illustrating a scanning method according to an exemplary embodiment;

FIG. 10 is a block diagram illustrating an apparatus for determining a size of a scanned object in accordance with an exemplary embodiment;

FIG. 11 is a block diagram illustrating another scanned object size determination apparatus in accordance with an exemplary embodiment;

FIG. 12 is a block diagram illustrating a scanning device in accordance with an exemplary embodiment;

FIG. 13 is a block diagram illustrating a scanning device in accordance with an exemplary embodiment;

FIG. 14 is a block diagram illustrating an electronic device in accordance with an example embodiment.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

In the disclosure, a scanning device performs plain scan on a target scanning object to obtain plain information, and then accurately identifies the length (which may be a thickness or a width, for example, the thickness or the width of a human body) of the scanning object in a scanning direction according to the plain information of the target scanning object. Therefore, the method and the device do not need to add extra measuring equipment on the scanning equipment, reduce the deployment difficulty and save the cost. Simultaneously, this disclosure only needs to carry out plain film scanning once to target scanning object, need not carry out positive position and side position twice scanning, compares in the scanning through positive position and side position to obtain positive position piece and side position piece, thereby confirm the technical scheme of the thickness of scanning object, can not increase the scanning dose that target scanning object received.

Before describing the method for determining the size of the scanned object, the scanning method, the scanning device, and the electronic device provided by the present disclosure, an application scenario related to various embodiments of the present disclosure is first described. The application scenario may be that a scanning device (e.g., a CT device, a CR device, a DR device, etc.) is used to scan a target scanning object, which may be, for example, a human, an animal or other objects, or a part of a human or an animal (i.e., a Region of Interest, english: Region of Interest, abbreviated as ROI). As shown in fig. 1, in order to explain a positional relationship between the scanning device and a target scanning object, a three-dimensional coordinate system may be established in a space where the scanning device is located, and includes a first coordinate axis, a second coordinate axis, and a third coordinate axis, where the first coordinate axis is a coordinate axis perpendicular to the ground from top to bottom (or from bottom to top), the second coordinate axis is a coordinate axis parallel to the ground from left to right (or from right to left), and the third coordinate axis is a coordinate axis parallel to the ground in a bed entering or bed exiting direction. In other embodiments, the first coordinate axis is a coordinate axis parallel to the ground from left to right (or from right to left), and the second coordinate axis is a coordinate axis perpendicular to the ground from top to bottom (or from bottom to top). The present disclosure does not specifically limit this.

The scanning frame can be provided with components such as a ray emitter, a collimator and a detector, and can rotate. The radiation emitter is used for emitting radiation, the radiation is emitted to a target scanning object through the beam limiter, and the detector is used for receiving the radiation transmitted through the target scanning object, so that a scanning image (such as a CT image or a plain film image) is generated. It should be noted that the radiation mentioned in the embodiments of the present disclosure may be X-ray, gamma-ray, or the like, and the present disclosure does not specifically limit this. Further, the scanning device may further include an image processing system, an image display module (e.g., a display), a storage module (e.g., a memory such as a hard disk), a recording module (e.g., a workstation), and the like, which are not particularly limited in this disclosure.

Fig. 2 is a flowchart illustrating a method of determining a size of a scanned object, according to an exemplary embodiment, as shown in fig. 2, the method including the steps of:

step 101, acquiring plain film information obtained by scanning a target scanning object by a scanning device according to a direction of a first coordinate axis, wherein the plain film information is plain film images or plain film scanning data. The flatbed scan data may include, among other things, the intensity of radiation received by the detectors of the scanning device.

For example, the scanning device may be controlled to scan the target scanning object in the direction of the first coordinate axis to obtain the corresponding plain film information. It can be understood that the scanning device is controlled to perform plain film scanning on the target scanning object. The plain film information may be a plain film image or plain film scan data. The first axis is directed in the scanning direction, and in the case of the scanning apparatus shown in fig. 1, the radiation emitter on the gantry may be controlled to emit a radiation downwards at a. The detector receives the radiation transmitted through the target scan object and measures the corresponding radiation intensity to obtain the flatwise scan data (i.e., raw data), which may be further processed to obtain a flatwise image. Taking the target scanning object as a person for example, the person can lie on the scanning bed or lie on the scanning bed, if the person lies on the scanning bed, the plain film image is the positive film, and the plain film scanning data is the raw data for generating the positive film. If the person lies on the scanning bed, the plain film image is the side film, and the plain film scanning data is the raw data for generating the side film. It should be noted that, in step 101, only one plain scan, that is, only one straight plain scan or one lateral plain scan, needs to be performed on the target scanning object, and the scanning dose is not increased additionally.

And 102, determining first length information of the target scanning object in the direction of the first coordinate axis through a pre-trained size recognition model according to the plain film information.

For example, after acquiring the plain film information, the plain film information may be identified through a pre-trained size identification model to determine first length information of the target scanning object in the first coordinate axis direction. Similarly, taking the target scanning object as a person for example, the person may lie on the scanning bed or lie on the scanning bed, and if the person lies on the scanning bed, the first length information includes the thickness of the person. If the person lies on his side on the scanning bed, the first length information comprises the width of the person. Further, the target scan object may be divided into a plurality of regions, and accordingly, the first length information may include a first length of each region in the first coordinate axis direction, and an average value of the first lengths of the plurality of regions in the first coordinate axis direction may be used as the first length information of the target scan object in the first coordinate axis direction.

The size recognition model can be obtained by training in advance according to a large number of training samples, and can recognize first length information of the target scanning object in the first coordinate axis direction according to the plain film information. The structure of the size recognition model may be CNN (Convolutional Neural Networks, Chinese), which may be, for example: ResNet (Chinese: Deep residual Network), DenseNet (Chinese: Dense convolution Network), VGGNet (Chinese: image Generator Network), UNet, etc., which are not specifically limited by this disclosure. Specifically, the plain film information may be used as an input of the size recognition model to obtain the first length information output by the size recognition model.

In summary, the present disclosure first obtains the plain film information obtained by the scanning device scanning the target scanning object according to the direction of the first coordinate axis, where the plain film information may be a plain film image or plain film scanning data. And then, according to the plain film information, determining first length information of the target scanning object in the direction of the first coordinate axis through a pre-trained size recognition model. The method and the device can accurately identify the length of the scanning object in the scanning direction according to the plain film information of the scanning object on the premise of not increasing the scanning dose and extra measuring equipment.

In an application scenario, the implementation manner of step 102 may be:

inputting the plain film information into a size recognition model to obtain first length information output by the size recognition model, wherein the first length information comprises: the target scanning method comprises the steps that at least one target area on a target scanning object is of a first length in the direction of a first coordinate axis, the target area is obtained by dividing according to projection of the target scanning object on a target plane, and the target plane is perpendicular to the first coordinate axis.

For example, the target scan object may be divided into at least one target region according to a projection of the target scan object on a target plane, where the target plane is a plane perpendicular to the first coordinate axis, that is, a plane formed by the second coordinate axis and the third coordinate axis. The projection of the target scanning object on the target plane can be understood as the contour of the target scanning object on the target plane. The rule for dividing the target region may be, for example, specifying a division interval on the second coordinate axis or a division interval on the third coordinate axis. Taking the projection of the target scanning object on the target plane as shown in fig. 3 as an example, the range of the projection on the second coordinate axis is 20mm to 300mm, and the range on the third coordinate axis is 0mm to 349mm, and a target area may be divided on the third coordinate axis every 50mm to divide the target scanning object into: target area (1), target areas (2), …, target area (6), target area (7), as indicated by the dashed lines in fig. 3. Wherein the range of the target area (1) on the second coordinate axis is 20mm-300mm, the range on the third coordinate axis is 0-49mm, the range of the target area (2) on the second coordinate axis is 20mm-300mm, the range on the third coordinate axis is 50mm-99mm, and so on.

Correspondingly, the plain film information includes plain film sub-information corresponding to each target area in the plurality of target areas. If the tile information is a tile image, the sub-image corresponding to each target area can be divided from the tile image according to the coordinate range of the target area, and the sub-image is used as the tile sub-information of the target area. If the tile information is the tile scanning data, the scanning data corresponding to each target area may be divided according to the coordinate range of the target area, and the divided scanning data is used as the tile sub-information of the target area. When the first length information is determined through the size recognition model, the plain film information can be input into the size recognition model, and the size recognition model can recognize the first length of each target area in the direction of the first coordinate axis according to the plain film sub-information corresponding to each target area.

Fig. 4 is a flowchart illustrating another method for determining a size of a scan object according to an exemplary embodiment, and as shown in fig. 4, the method may further include:

and 103, determining a second length of each target area in a second coordinate axis direction according to the plain film information, wherein the second coordinate axis is perpendicular to the first coordinate axis.

And 104, determining the volume of each target area according to a preset division interval, the first length of the target area in the direction of a first coordinate axis and the second length of the target area in the direction of a second coordinate axis, wherein the division interval is used for indicating the length of the target area in the direction of a third coordinate axis, and the third coordinate axis is respectively vertical to the first coordinate axis and the second coordinate axis.

For example, the second length of each target region in the direction of the second coordinate axis may be determined according to the plain film information. Taking the target scanning object as a person for example, the person may lie on the scanning bed or lie on the scanning bed, and if the person lies on the scanning bed, the first length of the target region in the first coordinate axis direction is the thickness of the target region, and correspondingly, the second length of the target region in the second coordinate axis direction is the width of the target region. If the person lies on the scanning bed, the first length of the target area in the first coordinate axis direction is the width of the target area, and correspondingly, the second length of the target area in the second coordinate axis direction is the thickness of the target area.

Then, for each target region, a volume of the target region may be determined according to a preset division interval, a first length of the target region in the first coordinate axis direction, and a second length in the second coordinate axis direction. It is to be understood that the target area is divided according to the specified division interval in the direction of the third coordinate axis, and therefore, the length of the target area in the third coordinate axis is known, that is, the division interval indicates the length of the target area in the direction of the third coordinate axis. The volume of the target region may be determined according to a first length of the target region in the first coordinate axis direction, a second length in the second coordinate axis direction, and a division interval in the third coordinate axis direction. For example, if the target region is a cube, the first length, the second length, and the division interval of the target region may be multiplied to obtain the volume of the target region.

FIG. 5 is a flowchart illustrating a method for training a size recognition model according to an exemplary embodiment, where the size recognition model is trained as shown in FIG. 5 by:

step A, a sample input set and a sample output set are obtained, wherein the sample input set comprises sample plain film information of a plurality of samples scanned in the direction of a first coordinate axis, and the sample output set comprises first actual length information corresponding to the plurality of samples in the direction of the first coordinate axis.

And step B, taking the sample input set as the input of the size recognition model, and taking the sample output set as the output of the size recognition model so as to train the size recognition model.

For example, to train a size recognition model, a sample input set and a corresponding sample output set are first obtained. The sample input set includes a plurality of sample inputs and the sample output set includes a sample output corresponding to each sample input. Specifically, the sample may be a human, an animal, or other objects, or may be a part of a human or an animal. Sample patch information for a plurality of samples may be acquired in advance, and the sample patch information for each sample is input as one sample. The sample slide information is obtained by scanning a corresponding sample by the scanning device in the direction of the first coordinate axis, and may include a sample slide image or sample slide scanning data. The sample slab scan data includes the intensity of radiation received by the detector while scanning the corresponding sample. Further, first actual length information of a plurality of samples in the direction of the first coordinate axis may be obtained in advance, and the first actual length information of each sample is output as one sample. The first actual length information may be understood as a label of the sample slab information of the sample. Specifically, the first actual length information may be obtained by manual actual measurement, or may be obtained by identifying a sample three-dimensional image (e.g., a CT image) of the sample according to a preset image identification algorithm, which is not limited in this disclosure. The sample three-dimensional image is obtained by three-dimensionally scanning a sample by scanning equipment.

The sample input set may then be used as input to a size recognition model, and the size recognition model may be trained based on the output of the size recognition model and the sample output set. For example, parameters of neurons in the size recognition model, such as weights (English: Weight) and offsets (English: Bias) of the neurons, may be corrected by a back propagation algorithm with the goal of reducing the loss function, based on the difference (or mean square error, mean absolute error) between the output of the size recognition model and the sample output set, as the loss function of the size recognition model. And repeating the steps until the loss function meets a preset condition, for example, the loss function is smaller than a preset loss threshold. Specifically, the loss function can be calculated by equation 1:

loss ═ MAE (Label _ C, C) equation 1

Wherein Loss represents a Loss function of the size recognition model, MAE represents an average absolute error, Label _ C represents first actual length information included in the sample output set, and C represents an output of the size recognition model. Further, the weights and offsets of the neurons can be updated using the above-mentioned loss function by equation 2:

wherein, alpha is the learning rate,

represents the weight of the ith node of the l layer and the jth node of the l +1 layer in the size recognition model,

indicating the bias of the ith node of the ith layer.

Fig. 6 is a flowchart illustrating another method for training a size recognition model according to an exemplary embodiment, where each sample output set includes a plurality of sample regions corresponding to samples, each sample region having a first actual length in a direction of a first coordinate axis, the sample regions being divided according to projections of the corresponding samples on a target plane, the target plane being perpendicular to the first coordinate axis, as shown in fig. 6.

Step B may be achieved by:

and step B1, inputting the sample plain film information of any sample into the size recognition model to obtain the initial length of each sample region of the sample in the direction of the first coordinate axis, which is output by the size recognition model.

Step B2, train the size recognition model based on the initial length of each sample region of the sample and the first actual length of each sample region of the sample.

For example, the sample may be divided into a plurality of sample regions according to a projection of the sample on a target plane, wherein the target plane is a plane perpendicular to the first coordinate axis, that is, a plane formed by the second coordinate axis and the third coordinate axis. The projection of the sample on the target plane can be understood as the contour of the sample on the target plane. The rule for dividing the sample region is the same as the rule for dividing the target region, and is not described herein again.

Accordingly, the sample tile information includes sample tile sub-information corresponding to each of the plurality of sample regions. If the sample flat slice information is a sample flat slice image, the sample flat slice image may be divided into sample sub-images corresponding to each sample region according to the coordinate range of the sample region, and the sample sub-images are used as the sample flat slice information of the sample region. If the sample tile information is sample tile scan data, the sample tile scan data may be divided into scan data corresponding to each sample region according to the coordinate range of the sample region, and the scan data may be used as the sample tile sub-information of the sample region. Accordingly, the sample is input into the corresponding sample output, and each of the plurality of sample regions including the corresponding sample has a first actual length in the first coordinate axis direction.

When the size recognition model is trained, the sample flatness information of any sample can be input into the size recognition model, and the size recognition model can determine the initial length of each sample region in the first coordinate axis direction according to the sample flatness sub-information corresponding to each sample region, that is, the size recognition model outputs that each sample region in a plurality of sample regions is at the initial length.

Thereafter, a size recognition model can be trained based on the initial length of each sample region of the sample at the first actual length for each sample region of the sample. For example, the parameters of the neurons in the size recognition model can be modified by a back propagation algorithm with the goal of reducing the loss function according to the initial length of each sample region and the difference (or mean square error, mean absolute error) from the initial length of the sample region as the loss function of the size recognition model. And repeating the steps until the loss function meets a preset condition, for example, the loss function is smaller than a preset loss threshold. Specifically, the loss function can be calculated by equation 3:

where Loss represents a Loss function of the size recognition model, MAE represents an average absolute error, Label _ Cn represents a first actual length of the nth sample region, Cn represents an initial length of the nth sample region output by the size recognition model, and N represents the number of sample regions.

FIG. 7 is a flowchart illustrating another method for training a size recognition model, according to an exemplary embodiment, as shown in FIG. 7, after step A, the size recognition model is further trained by:

and step C, adjusting each sample slide information in the sample input set according to a preset rule, and generating a plurality of slide expansion information corresponding to the sample slide information so as to expand the sample input set.

And D, expanding the sample output set according to the expanded sample input set.

Correspondingly, the implementation manner of step B may be:

and taking the extended sample input set as the input of the size recognition model, and taking the extended sample output set as the output of the size recognition model so as to train the size recognition model.

For example, before the size recognition model is trained, the sample input set and the sample output set may be expanded according to a preset rule. If the sample slide information is sample slide scanning data, for each sample slide scanning data, the sample slide scanning data can be scrambled, disordered and the like according to a preset rule to obtain a plurality of expansion data corresponding to the sample slide scanning data, and the plurality of expansion data are used as slide expansion information and are put into a sample input set, so that the purpose of expanding the sample input set is achieved. If the sample tile information is a sample tile image, then for each sample tile image, the pixels in the sample tile image may be adjusted according to a preset rule (e.g., translation, rotation, scaling, etc.) to obtain a plurality of extended images corresponding to the sample tile image, and the plurality of extended images are used as tile extended information and placed into the sample input set, so as to achieve the purpose of extending the sample input set.

Correspondingly, the first actual lengths of the plurality of slice extension information corresponding to the sample slice information are the first actual lengths of the sample slice information, so that in the sample output set, the corresponding sample output can be added to each slice extension information, thereby achieving the purpose of extending the sample output set. Finally, the expanded sample input set can be used as the input of the size recognition model, and the expanded sample output set can be used as the output of the size recognition model, so as to train the size recognition model.

Fig. 8 is a flow chart illustrating a scanning method according to an exemplary embodiment, as shown in fig. 8, the method including:

first, according to the determination method of the size of the scanning object provided by the embodiment of the present disclosure, first length information of the target scanning object in the direction of the first coordinate axis is determined. Scanning is then achieved by the following steps.

Step 201, determining the geometric center of the target scanning object according to the first length information.

Step 202, according to the geometric center of the target scanning object, the geometric center of the target scanning object is adjusted by controlling the height of the scanning bed of the scanning device, so that the coordinate of the adjusted geometric center of the target scanning object in the first coordinate axis direction is the same as the coordinate of the scanning center of the scanning device in the first coordinate axis direction.

And step 203, controlling the scanning device to perform three-dimensional scanning on the target scanning object.

For example, the geometric center of the target scan object may be determined based on the first length information. Specifically, the coordinate of the geometric center of the target scanning object in the direction of the first coordinate axis may be determined according to the first length information and the current height of the scanning bed. For example, a first length of the target scanning object in the first coordinate axis direction may be determined according to the first length information, and then a midpoint of the first length of the target scanning object in the first coordinate axis direction is used as a geometric center, that is, a distance from the geometric center to the ground is equal to + a current height of the scanning bed (the first length of the target scanning object)/2. Taking the first length of the target scanning object on the first coordinate axis as 20mm and the current height of the scanning bed as 50mm as an example, the distance from the geometric center to the ground is 50+ 20/2-60 mm. If the ground has coordinates of 120mm on the first axis, i.e. the scanning bed has coordinates of 70mm on the first axis, then the geometric center has coordinates of 60mm on the first axis.

Then, the height of the scanning bed of the scanning device can be controlled according to the geometric center of the target scanning object, so that the purpose of adjusting the geometric center of the target scanning object is achieved, and the coordinate of the adjusted geometric center of the target scanning object in the first coordinate axis direction is the same as the coordinate of the scanning center of the scanning device in the first coordinate axis direction. In particular, the scan center may be understood as the center of a circle of an aperture of the gantry, and therefore the coordinate of the scan center in the direction of the first coordinate axis is known, and the geometric center of a target scan object located on the scanning bed may be adjusted by raising or lowering the height of the scanning bed. Similarly, for example, the first length of the target scanning object in the direction of the first coordinate axis is 20mm, the current height of the scanning bed is 50mm, and the coordinate of the geometric center determined in step 201 on the first coordinate axis is 60 mm. If the coordinate of the scan center in the direction of the first coordinate axis is 50mm, the height of the scanning bed may be raised by 10mm, and if the coordinate of the scan center in the direction of the first coordinate axis is 70mm, the height of the scanning bed may be lowered by 10 mm. Finally, the scanning frame can be controlled to rotate, and the target scanning object is scanned in three dimensions to obtain a three-dimensional scanning image (such as a CT image). The coordinate of the geometric center of the adjusted target scanning object in the first coordinate axis direction is the same as the coordinate of the scanning center of the scanning equipment in the first coordinate axis direction, so that the scanning dosage received by the target scanning object can be ensured to be uniform, and the quality of a three-dimensional scanning image is ensured.

Before three-dimensional scanning is performed on the target scanning object, a region of interest (i.e., ROI) on the target scanning object may be determined, and the region of interest is determined. Specifically, the doctor may manually mark the target region on the plain image, or may use a pre-trained segmentation model to segment the plain image to obtain the region of interest output by the segmentation model, or may use a preset algorithm such as edge segmentation, threshold segmentation, region segmentation, or cluster segmentation to extract the region of interest. After the region of interest is determined, a three-dimensional scan of the region of interest may be performed. Thus, in one implementation, step 201 may be preceded by determining a region of interest of the target scan object from the surview image. Accordingly, step 201 may be to determine the geometric center of the region of interest according to the first length information, and step 202 may be to adjust the geometric center of the region of interest by controlling the height of the scanning bed of the scanning device according to the geometric center of the region of interest, so that the coordinates of the adjusted geometric center of the region of interest in the first coordinate axis direction are the same as the coordinates of the scanning center of the scanning device in the first coordinate axis direction. Step 203 may control the scanning device to scan the region of interest in three dimensions. That is, the first length information determined in the embodiment of the present disclosure may be used for three-dimensional scanning of the target scanning object, and may also be used for three-dimensional scanning of a region of interest of the target object, which is not specifically limited by the present disclosure.

Fig. 9 is a flow chart illustrating a scanning method according to an exemplary embodiment, as shown in fig. 9, the method including:

first, according to the method for determining the size of the scanning object provided by the embodiment of the disclosure, the volume of at least one target area of the target scanning object is determined. Scanning is then achieved by the following steps.

Step 301, determining the scanning intensity corresponding to each target region according to the volume of the target region.

Step 302, controlling the scanning device to perform three-dimensional scanning on each target area according to the corresponding scanning intensity of the target area.

For example, after determining the first length information and determining the volume of each target region, the scan intensity corresponding to the target region may be further determined according to the volume of the target region. The scanning intensity corresponding to the target region can be determined according to the preset corresponding relationship between the volume and the scanning intensity, wherein the volume is positively correlated with the scanning intensity, that is, the larger the volume is, the larger the scanning intensity is. Specifically, the corresponding relationship may be a relationship function fitted in advance according to a large amount of sample data, may also be a relationship table counted in advance according to a large amount of sample data, and may also be a corresponding relationship model trained in advance according to a large amount of sample data, which is not specifically limited by this disclosure.

Finally, the scanning device can be controlled to perform three-dimensional scanning on each target area according to the scanning intensity corresponding to the target area. That is, for each target area, when the scanning device is controlled to perform three-dimensional scanning on the target area, the intensity of the radiation emitted by the radiation emitter is the scanning intensity corresponding to the target area. Therefore, the scanning dose received by the target scanning object can be ensured to be uniform, and the quality of a three-dimensional scanning image is ensured.

Also, in one implementation, step 301 may be preceded by determining a region of interest of the target scan object from the surview image. Correspondingly, the second length of each target region in the region of interest on the second coordinate axis may be determined according to the tile information, and the volume of each target region in the region of interest may be determined according to the first length of each target region in the region of interest on the first coordinate axis direction in combination with the preset division interval. Accordingly, step 301 may determine a corresponding scan intensity according to the volume, and step 302 may control the scanning device to perform a three-dimensional scan on each target region in the region of interest according to the scan intensity corresponding to the target region. That is, the size information determined in the embodiment of the present disclosure may be used for three-dimensional scanning of the target scanning object, and may also be used for three-dimensional scanning of a region of interest of the target object, which is not specifically limited by the present disclosure.

Fig. 10 is a block diagram illustrating an apparatus for determining a size of a scanned object according to an exemplary embodiment, and as shown in fig. 10, the apparatus 400 may include:

the obtaining module 401 is configured to obtain flatting information obtained by scanning a target scanning object by a scanning device in a direction of a first coordinate axis, where the flatting information is a flatting image or flatting scanning data.

The first determining module 402 is configured to determine, according to the tile information, first length information of the target scan object in the first coordinate axis direction through a pre-trained size recognition model.

In an application scenario, the first determining module 402 may be configured to:

Fig. 11 is a block diagram illustrating another apparatus for determining a size of a scan object according to an exemplary embodiment, and as shown in fig. 11, the apparatus 400 may further include:

the second determining module 403 is configured to determine, according to the plain film information, a second length of each target area in a second coordinate axis direction, where the second coordinate axis is perpendicular to the first coordinate axis. And for each target area, determining the volume of the target area according to a preset division interval, the first length of the target area in the direction of a first coordinate axis and the second length of the target area in the direction of a second coordinate axis, wherein the division interval is used for indicating the length of the target area in the direction of a third coordinate axis, and the third coordinate axis is respectively vertical to the first coordinate axis and the second coordinate axis.

In one implementation, the size recognition model is trained by:

In another implementation, the sample output set includes a plurality of sample regions corresponding to the samples, each sample region having a first actual length in the direction of the first coordinate axis, the sample regions being obtained by dividing a projection of the corresponding sample onto a target plane, the target plane being perpendicular to the first coordinate axis.

Step B may include:

In another implementation, after step a, the size recognition model is further trained by:

Correspondingly, the implementation manner of step B may be:

With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Fig. 12 is a block diagram illustrating a scanning apparatus according to an exemplary embodiment, and as shown in fig. 12, the apparatus 500 includes:

the obtaining module 501 is configured to obtain flatting information obtained by scanning a target scanning object by a scanning device in a direction of a first coordinate axis, where the flatting information is a flatting image or flatting scanning data.

The first determining module 502 is configured to determine, according to the plain film information, first length information of the target scan object in the first coordinate axis direction through a pre-trained size recognition model.

And a second determining module 503, configured to determine a geometric center of the target scanning object according to the first length information.

An adjusting module 504, configured to adjust the geometric center of the target scanning object by controlling the height of the scanning bed of the scanning device according to the geometric center of the target scanning object, so that the coordinate of the adjusted geometric center of the target scanning object on the first coordinate axis is the same as the coordinate of the scanning center of the scanning device on the first coordinate axis.

The first control module 505 is configured to control the scanning device to perform three-dimensional scanning on a target scanning object.

Fig. 13 is a block diagram illustrating a scanning apparatus according to an exemplary embodiment, and as shown in fig. 13, the apparatus 600 includes:

the obtaining module 601 is configured to obtain flatting information obtained by scanning a target scanning object by a scanning device in a direction of a first coordinate axis, where the flatting information is a flatting image or flatting scanning data.

The first determining module 602 is configured to determine, according to the plain film information, first length information of the target scan object in the first coordinate axis direction through a pre-trained size recognition model.

A third determining module 603, configured to determine, according to the volume of each target region, a scanning intensity corresponding to the target region.

The second control module 604 is configured to control the scanning device to perform three-dimensional scanning on each target region according to the scanning intensity corresponding to the target region.

Fig. 14 is a block diagram illustrating an electronic device 700 according to an example embodiment. As shown in fig. 14, the electronic device 700 may include: a processor 701 and a memory 702. The electronic device 700 may also include one or more of a multimedia component 703, an input/output (I/O) interface 704, and a communication component 705.

The processor 701 is configured to control the overall operation of the electronic device 700, so as to complete all or part of the steps in the above-mentioned method for determining the size of the scanned object, or the scanning method. The memory 702 is used to store various types of data to support operation at the electronic device 700, such as instructions for any application or method operating on the electronic device 700 and application-related data, such as contact data, transmitted and received messages, pictures, audio, video, and the like. The Memory 702 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk, or optical disk. The multimedia components 703 may include screen and audio components. Wherein the screen may be, for example, a touch screen and the audio component is used for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may further be stored in the memory 702 or transmitted through the communication component 705. The audio assembly also includes at least one speaker for outputting audio signals. The I/O interface 704 provides an interface between the processor 701 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless Communication, such as Wi-Fi, bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more of them, which is not limited herein. The corresponding communication component 705 may thus include: Wi-Fi module, Bluetooth module, NFC module, etc.

In an exemplary embodiment, the electronic Device 700 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above-described method for determining the size of the scan object, or the scanning method.

In another exemplary embodiment, there is also provided a computer readable storage medium comprising program instructions which, when executed by a processor, implement the above-mentioned method of determining a size of a scan object, or scanning method. For example, the computer readable storage medium may be the memory 702 described above including program instructions executable by the processor 701 of the electronic device 700 to perform the method for determining the size of a scanned object, or the scanning method described above.

In another exemplary embodiment, a computer program product is also provided, which comprises a computer program executable by a programmable apparatus, the computer program having code portions for performing the above-mentioned method for determining a size of a scan object, or the scan method, when executed by the programmable apparatus.

The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.

It should be noted that, in the foregoing embodiments, various features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various combinations that are possible in the present disclosure are not described again.

In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.

Claims

1. A method for determining the size of a scanned object, the method comprising:

2. The method of claim 1, wherein determining, according to the tile information, first length information of the target scan object in the first coordinate axis direction through a pre-trained size recognition model comprises:

3. The method of claim 2, further comprising:

4. The method according to any one of claims 1-3, wherein the size recognition model is trained by:

5. The method of claim 4, wherein the sample output set comprises a plurality of sample regions corresponding to the samples, each of the sample regions having a first actual length in the direction of the first coordinate axis, the sample regions being partitioned according to a projection of the corresponding sample onto a target plane, the target plane being perpendicular to the first coordinate axis;

6. The method of claim 4, wherein after the obtaining of the input set of samples and the output set of samples, the size recognition model is further trained by:

expanding the sample output set according to the expanded sample input set;

7. A scanning method, comprising:

the method for determining the size of the scanned object according to any one of claims 1-6, determining first length information of the target scanned object in the direction of the first coordinate axis;

controlling the scanning equipment to carry out three-dimensional scanning on the target scanning object; or, comprising:

the method of claim 3, determining the volume of at least one target region on the target scan object:

8. An apparatus for determining the size of a scanned object, the apparatus comprising:

9. A scanning device, comprising:

the first control module is used for controlling the scanning equipment to carry out three-dimensional scanning on the target scanning object; or the like, or, alternatively,

the method comprises the following steps: the acquisition module is used for acquiring plain film information obtained by scanning a target scanning object by scanning equipment according to the direction of a first coordinate axis, wherein the plain film information is plain film images or plain film scanning data;

10. An electronic device, comprising:

a memory having a computer program stored thereon;

a processor for executing the computer program in the memory to carry out the steps of the method of any one of claims 1 to 6.