WO2024077712A1

WO2024077712A1 - Method and apparatus for determining exposure parameter, storage medium, and program product

Info

Publication number: WO2024077712A1
Application number: PCT/CN2022/132985
Authority: WO
Inventors: Mengnan JIAO; Xishuai PENG; Yunzhe ZOU
Original assignee: Siemens Shanghai Medical Equipment Ltd.
Priority date: 2022-10-09
Filing date: 2022-11-18
Publication date: 2024-04-18
Also published as: CN117911306A

Abstract

The implementations of the present invention disclose a method and an apparatus for determining an exposure parameter in X-ray imaging, a storage medium, and a program product. The method includes: determining an X-ray imaging protocol and a thickness value of an imaging object; inputting the thickness value into an exposure parameter prediction model corresponding to the X-ray imaging protocol, where the exposure parameter prediction model is trained based on first training data, and the first training data includes exposure parameter historical values in historical exposure operations based on the X-ray imaging protocol and historical thickness values in the historical exposure operations; and receiving, from the exposure parameter prediction model, an exposure parameter predicted value determined based on the thickness value. In the implementations of the present invention, the exposure parameter is automatically determined according to the thickness value of the imaging object, which improves the exposure accuracy and reduces the manual difficulty. In the implementations of the present invention, the exposure parameter can be determined in a general manner independent of the adjustment operation or in a dedicated manner related to the adjustment operation, and the automatic determination of the thickness value is further implemented.

Description

METHOD AND APPARATUS FOR DETERMINING EXPOSURE PARAMETER, STORAGE MEDIUM, AND PROGRAM PRODUCT

TECHNICAL FIELD

The present invention relates to the field of medical imaging technologies, and in particular, to a method and an apparatus for determining an exposure parameter in X-ray imaging, a storage medium, and a program product.

BACKGROUND

X-rays are electromagnetic radiation with wavelengths between ultraviolet and gamma rays. X-rays are penetrable and have different penetrability to substances of different densities. In medicine, X-rays are used to project human organs and bones to form medical images. An X-ray imaging system typically includes an X-ray generator assembly, a bucky-wall-stand (BWS) assembly, an inspection table assembly, a box assembly including a flat panel detector, a control host located remotely, and the like. The X-ray generator uses a high voltage provided by a high voltage generator to emit X-rays that penetrate and illuminate an imaging object, and forms medical image information of the imaging object on the flat panel detector. The flat panel detector sends the medical image information to the control host. The imaging object may stand near the BWS assembly or lie on the inspection table assembly to respectively receive X-ray imaging on parts such as the skull, the chest, the abdomen, and the joints.

In X-ray imaging, X-ray exposure parameters (for example, tube voltage, tube current, and exposure time) have a great impact on X-ray image quality. At present, X-ray exposure parameters are mainly set according to personal experience of technicians, which is difficult to implement.

SUMMARY

The implementations of the present invention provide a method and an apparatus for determining an exposure parameter in X-ray imaging, a storage medium, and a program product.

The technical solutions of the implementations of the present invention include:

A method for determining an exposure parameter in X-ray imaging is provided, including:

determining an X-ray imaging protocol and a thickness value of an imaging object;

inputting the thickness value into an exposure parameter prediction model corresponding to the X-ray imaging protocol, where the exposure parameter prediction model is trained based on first training data, and the first training data includes exposure parameter historical values in historical exposure operations based on the X-ray imaging protocol and historical thickness values in the historical exposure operations; and

receiving, from the exposure parameter prediction model, an exposure parameter predicted value determined based on the thickness value.

As can be seen, in the implementations of the present invention, the exposure parameter is automatically determined according to the thickness value of the imaging object, which improves the exposure accuracy and reduces the manual difficulty. Moreover, the correlation between the thickness value of the imaging object and the exposure parameter is considered in the process of automatically determining the exposure parameter, which further improves the accuracy of the parameter.

In an exemplary implementation, the method further includes:

establishing a first artificial neural network model;

inputting the first training data into the first artificial neural network model; and

training the first artificial neural network model into the exposure parameter prediction model by using the first training data.

Therefore, the artificial neural network is introduced into the exposure parameter prediction process, which further reduces the difficulty of implementation.

In an exemplary implementation, the method further includes:

inputting the exposure parameter predicted value into an exposure parameter adjustment model corresponding to the exposure parameter prediction model, where the exposure parameter adjustment model is obtained based on second training data, and the second training data includes exposure parameter historical predicted values historically outputted by the exposure parameter prediction model, and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values;

receiving, from the exposure parameter adjustment model, an exposure parameter adjusted value determined based on the exposure parameter predicted value; and

performing an exposure operation on the imaging object based on the exposure parameter adjusted value to generate an X-ray image.

As can be seen, in the implementations of the present invention, the exposure parameter predicted value is adjusted through the exposure parameter historical adjusted values, which introduces the adjustment preference habit, implements the determination of the exposure parameter in a manner related to the adjustment operation, and is suitable for the customization situation.

In an exemplary implementation, the method further includes:

establishing a second artificial neural network model;

inputting the second training data into the second artificial neural network model; and

training the second artificial neural network model into the exposure parameter adjustment model by using the second training data.

Therefore, the artificial neural network is introduced into the exposure parameter adjustment process, which further reduces the difficulty of implementation.

In an exemplary implementation, the first training data is derived from log data of a plurality of X-ray imaging units; and

the second training data is derived from log data of same X-ray imaging units, or the second training data is derived from log data of same X-ray imaging operators in the same X-ray imaging units.

As can be seen, the first training data is derived from log data of a plurality of X-ray imaging units, and the trained exposure parameter prediction model is a widely applicable general model. The second training data is from the same X-ray imaging units or the same X-ray imaging operators, and the trained exposure parameter adjustment model is applicable to the same X-ray imaging units or the same X-ray imaging operators, which is a customized model.

In an exemplary implementation, the determining a thickness value of an imaging object includes:

determining a distance between an X-ray source and a surface point of the imaging object based on a three-dimensional image including the imaging object;

determining a thickness value of the imaging object at the surface point based on a distance between the X-ray source and an imaging surface, a distance between a contact plate and a detector, and the distance between the X-ray source and the surface point of the imaging object; and

determining the thickness value of the imaging object based on the thickness value of the imaging object at the surface point, where

the thickness value of the imaging object includes a minimum thickness value of the imaging object or an average thickness value of the imaging object.

Therefore, in the implementations of the present invention, the automatic determination of the thickness value of the imaging object is further implemented, which reduces the tedious work of manually measuring the thickness value of the imaging object.

An apparatus for determining an exposure parameter in X-ray imaging is provided, including:

a determining module, configured to determine an X-ray imaging protocol and a thickness value of an imaging object;

a first input module, configured to input the thickness value into an exposure parameter prediction model corresponding to the X-ray imaging protocol, where the exposure parameter prediction model is trained based on first training data, and the first training data includes exposure parameter historical values in historical exposure operations based on the X-ray imaging protocol and historical thickness values in the historical exposure operations; and

a first receiving module, configured to receive, from the exposure parameter prediction model, an exposure parameter predicted value determined based on the thickness value.

In an exemplary implementation, the apparatus further includes:

a first training module, configured to establish a first artificial neural network model; input the first training data into the first artificial neural network model; and train the first artificial neural network model into the exposure parameter prediction model by using the first training data.

In an exemplary implementation, the apparatus further includes:

a second input module, configured to input the exposure parameter predicted value into an exposure parameter adjustment model corresponding to the exposure parameter prediction model, where the exposure parameter adjustment model is obtained based on second training data, and the second training data includes exposure parameter historical predicted values historically outputted by the exposure parameter prediction model, and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values;

a second receiving module, configured to receive, from the exposure parameter adjustment model, an exposure parameter adjusted value determined based on the exposure parameter predicted value; and

an exposure module, configured to perform an exposure operation on the imaging object based on the exposure parameter adjusted value to generate an X-ray image.

In an exemplary implementation, the apparatus further includes:

a second training module, configured to establish a second artificial neural network model; input the second training data into the second artificial neural network model; and train the second artificial neural network model into the exposure parameter adjustment model by using the second training data.

In an exemplary implementation, the determining module is configured to determine a distance between an X-ray source and a surface point of the imaging object based on a three-dimensional image including the imaging object; determine a thickness value of the imaging object at the surface point based on a distance between the X-ray source and an imaging surface, a distance between a contact plate and a detector, and the distance between the X-ray source and the surface point of the imaging object; and determine the thickness value of the imaging object based on the thickness value of the imaging object at the surface point, where the thickness value of the imaging object includes a minimum thickness value of the imaging object or an average thickness value of the imaging object.

An apparatus for determining an exposure parameter in X-ray imaging is provided, including: a processor and a memory, where

the memory stores an application program executable by the processor, to enable the processor to perform the method for determining an exposure parameter in X-ray imaging according to any one of foregoing aspects.

A computer-readable storage medium is provided, storing computer-readable instructions, where the computer-readable instructions, when executed by the processor, implements the method for determining an exposure parameter in X-ray imaging according to any one of the foregoing aspects.

A computer program product is provided, including a computer program, where the computer program, when executed by a processor, implements the method for determining an exposure parameter in X-ray imaging according to any one of the foregoing aspects.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flowchart of a method for determining an exposure parameter in X-ray imaging according to an implementation of the present invention.

FIG. 2 is an exemplary process diagram of determining a thickness value of an imaging object in X-ray imaging according to an implementation of the present invention.

FIG. 3 is an exemplary process diagram of determining an exposure parameter in X-ray imaging according to an implementation of the present invention.

FIG. 4 is an exemplary schematic diagram of a human-computer interaction process of X-ray imaging according to an implementation of the present invention.

FIG. 5 is a structural diagram of an apparatus for determining an exposure parameter in X-ray imaging according to an implementation of the present invention.

FIG. 6 is a structural diagram of an apparatus for determining an exposure parameter in X-ray imaging with a memory-processor architecture according to an implementation of the present invention.

Reference numerals are as follows:

DETAILED DESCRIPTION

To make the technical solutions and advantages of the present invention clearer and more comprehensible, the following further describes the present invention in detail with reference to the accompanying drawings and implementations. It should be understood that the specific implementations described herein are merely used to illustratively explain the present invention but are not intended to limit the protection scope of the present invention.

For brief and intuitive description, the following describes the solutions of the present invention by describing several typical implementations. A large quantity of details in the implementations is merely used for helping understand the solutions of the present invention. However, obviously, implementation of the technical solutions of the present invention is not limited to these details. To avoid obscuring the solutions of the present invention, some implementations are not described in detail, but only a framework is provided. In the following text, "include" refers to "include but is not limited to" , and "according to …" refers to "according to at least …, but not being limited only to according to …" . Because of Chinese language habits, the following does not particularly specify the quantity of a component, which means that the component may be one or more, or can be understood as at least one.

In view of the defect of setting X-ray exposure parameters according to the personal experience of technicians in the related art, the implementations of the present invention take into account the correlation between a thickness value of an imaging object and an exposure parameter (normally, a larger thickness value of the imaging object indicates a larger exposure parameter) . The exposure parameter is automatically determined according to the thickness value of the imaging object, which not only improves the exposure accuracy, but also reduces the manual difficulty.

FIG. 1 is a flowchart of a method for determining an exposure parameter in X-ray imaging according to an implementation of the present invention. Preferably, the method shown in FIG. 1 may be performed by a controller. The controller may be implemented as a control host integrated into an X-ray imaging system or as a control unit independent from the control host.

As shown in FIG. 1, the method 100 includes the following steps:

Step 101: Determine an X-ray imaging protocol and a thickness value of an imaging object.

The imaging object is an object on which X-ray imaging needs to be performed. The imaging object may be an organism or an inanimate object, and the implementations of the present invention do not limit the specific characteristics of the imaging object. The X-ray imaging protocol is a specific protocol (for example, an organ protocol (OGP) ) used in the process of performing X-ray imaging on the imaging object. The X-ray imaging protocol may be determined based on a selection operation of the user in a human-computer interaction interface.

In an optional implementation, the thickness value of the imaging object may be measured manually by the staff using a ruler. Preferably, considering many defects of manually measuring the thickness value of the imaging object by using the ruler, the implementations of the present invention further provide a technical solution for automatically determining the thickness value of the imaging object, which can automatically determine the thickness value of the imaging object based on a three-dimensional image of the imaging object, to reduce the tedious work of manually measuring the thickness value.

A camera component may be used to shoot the imaging object to obtain the three-dimensional image of the imaging object, and the three-dimensional image of the imaging object may be alternatively obtained from a storage medium (for example, the cloud or local database) , where the three-dimensional image is obtained by shooting the to-be-detected object by using the camera component. After the camera component acquires the three-dimensional image of the imaging object, the three-dimensional image may be sent to the controller executing the process in FIG. 1 through a wired or wireless interface. Preferably, the wired interface includes at least one of the following: a general serial bus interface, a controller local area network interface, a serial port, and the like; and the wireless interface includes at least one of the following: an infrared interface, a near field communication interface, a Bluetooth interface, a Zigbee interface, a wireless broadband interface, and the like.

In an exemplary implementation, the specific steps of determining the thickness value of the imaging object include: step (1) : determining a distance (Source to Object Surface Distance, SOSD) between an X-ray source and a surface point of the imaging object based on a three-dimensional image including the imaging object; step (2) : determining a thickness value of the imaging object at the surface point based on a distance (Source to Image Distance, SID) between the X-ray source and an imaging surface, a distance (Table to Detector Distance, TDD) between a contact plate and a detector, and the distance between the X-ray source and the surface point of the imaging object; and step (3) : determining the thickness value of the imaging object based on the thickness value of the imaging object at the surface point, where the thickness value of the imaging object includes a minimum thickness value of the imaging object or an average thickness value of the imaging object, and the like.

The contact plate is the plate touched by the to-be-detected object in the X-ray application. The contact plate can isolate the to-be-detected object and the imaging surface. The contact plate generally has the following meanings:

(1) When the X-ray imaging system operates in an inspection table mode, the contact plate is the table plate of the inspection table.

(2) When the X-ray imaging system operates in a bucky-wall-stand mode in a whole spine imaging protocol, the contact plate is a support plate used to assist the to-be-detected object to stand.

(3) When the X-ray imaging system operates in a bucky-wall-stand mode in a non-whole spine imaging protocol (for example, a chest imaging protocol, or a knee imaging protocol) , the contact plate is a panel of a box assembly, where a flat panel detector is inserted into the box assembly.

(4) When the X-ray imaging system operates in a free exposure mode (that is, the to-be-detected object is in direct contact with the flat panel detector) , the contact plate is a panel touched by the flat panel detector and the to-be-detected object.

Both the distance between the X-ray source and the imaging surface and the distance between the contact plate and the detector correspond to the X-ray imaging protocol. When the X-ray imaging protocol is determined, both the distance between the X-ray source and the imaging surface and the distance between the contact plate and the detector are known values.

FIG. 2 is an exemplary process diagram of determining a thickness value of an imaging object in X-ray imaging according to an implementation of the present invention. In FIG. 2, an exemplary process of determining the thickness value of the imaging object is described using whole spine imaging as an example.

A to-be-detected object 75 stands at a support plate 76. For example, the left or right shoulder thereof is a specific surface point C of the whole spine imaging. A distance between an X-ray source (usually located in an X-ray tube) in an X-ray generator assembly 77 and an imaging surface of a flat panel detector in a box assembly 78 is SID; a distance between the support plate 76 and the flat panel detector in the box assembly 78 is TDD; and a distance between a spinal datum line 79 and the support plate 76 is TOD, that is, the thickness value of the imaging object.

A distance between the X-ray source and the surface point C is SOSDc, where SOSDc=R*Ip+T; R is a rotation matrix between a camera component coordinate system of the three-dimensional image formed by shooting the to-be-detected object 75 and an X-ray tube coordinate system; Ip is a transformation matrix between the camera component coordinate system and the X-ray tube coordinate system; and T is a depth value of the point C (T may be determined based on the three-dimensional image) .

A thickness value TODc at the surface point C may be determined based on SOSDc, where TODc= SID-TDD-SOSDc.

Similarly, a thickness value at each surface point can be calculated in the whole spine imaging. In an implementation, a minimum value is selected from thickness values at all surface points and used as the thickness value of the imaging object in step 101. In an implementation, an average thickness value is calculated based on the thickness values at all surface points and used as the thickness value of the imaging object in step 101.

The typical example of the thickness value of the imaging object is exemplarily described above. A person skilled in the art may realize that the description is merely exemplary and is not used to limit the protection scope of the implementations of the present invention.

Step 102: Input the thickness value into an exposure parameter prediction model corresponding to the X-ray imaging protocol, where the exposure parameter prediction model is trained based on first training data, and the first training data includes exposure parameter historical values in historical exposure operations based on the X-ray imaging protocol and historical thickness values in the historical exposure operations.

It can be seen that the correlation between the thickness value of the imaging object and the exposure parameter is included in the exposure parameter prediction model trained by using the exposure parameter historical values in the historical exposure operations and the historical thickness values in the historical exposure operations. Therefore, the exposure parameter prediction model can improve the accuracy of the exposure parameter. The exposure parameter may include at least one of: tube voltage, tube current, exposure time, exposure dose, product of tube current and exposure time, and exposure density.

In an exemplary implementation, the method 100 further includes: establishing a first artificial neural network model; inputting the first training data into the first artificial neural network model; and training the first artificial neural network model into the exposure parameter prediction model by using the first training data. Therefore, the artificial neural network is introduced into the exposure parameter prediction process, which further reduces the difficulty of implementation.

Step 103: Receive, from the exposure parameter prediction model, an exposure parameter predicted value determined based on the thickness value.

The exposure parameter value predicted by the exposure parameter prediction model based on the inputted thickness value is the exposure parameter predicted value. An exposure operation may be performed on the imaging object based on the exposure parameter predicted value to generate an X-ray image.

Therefore, in the implementations of the present invention, the exposure parameter predicted value is outputted through the exposure parameter prediction model, and the exposure parameter predicted value can be directly used to perform the exposure operation on the imaging object, so that the exposure parameter can be determined in a general manner independent of the adjustment operation, improving the exposure efficiency.

Considering that technicians usually adjust the automatically determined exposure parameter predicted value based on personal habits, the exposure parameter predicted value may be further automatically adjusted based on the preferences of the technicians.

In an exemplary implementation, the method 100 further includes: inputting the exposure parameter predicted value into an exposure parameter adjustment model corresponding to the exposure parameter prediction model, where the exposure parameter adjustment model is obtained based on second training data, and the second training data includes exposure parameter historical predicted values historically outputted by the exposure parameter prediction model, and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values; receiving, from the exposure parameter adjustment model, an exposure parameter adjusted value determined based on the exposure parameter predicted value; and performing an exposure operation on the imaging object based on the exposure parameter adjusted value to generate an X-ray image. The meaning of the exposure parameter historical predicted values is: predicted values that are historically outputted by the exposure parameter prediction model for the exposure parameter. The meaning of the exposure parameter historical adjusted values is: actual values of the exposure parameter obtained after the exposure parameter historical predicted values are historically adjusted (usually adjusted manually by the technicians) . The meaning of the exposure parameter adjusted value is: an actual value of the exposure parameter obtained after the exposure parameter predicted value inputted into exposure parameter adjustment model is adjusted. Therefore, the exposure parameter adjusted value outputted by the exposure parameter adjustment model is used as the actual value of the adjusted exposure parameter, which can be directly applied to the exposure operation.

As can be seen, the exposure parameter adjustment model trained by using the exposure parameter historical predicted values outputted by the exposure parameter prediction model and the exposure parameter historical adjusted values (usually adjusted manually by the technicians) corresponding to the exposure parameter historical predicted values includes the correlation between the exposure parameter historical predicted values and the exposure parameter historical adjusted values, thereby manifesting the adjustment preferences of the technicians. Therefore, the accuracy of the exposure parameter can be further improved.

Therefore, in the implementations of the present invention, the exposure parameter adjusted value is outputted by the exposure parameter prediction model and exposure parameter adjustment model, and the exposure parameter adjusted value can be directly used for performing the exposure operation on the imaging object and further reflects the adjustment preferences of the technicians, so that the exposure parameter can be determined in a dedicated manner related to the adjustment operation, improving the accuracy of the exposure parameter.

In an exemplary implementation, the method 100 further includes: establishing a second artificial neural network model; inputting the second training data into the second artificial neural network model; and training the second artificial neural network model into the exposure parameter adjustment model by using the second training data. Therefore, the artificial neural network is introduced into the exposure parameter adjustment process, which further reduces the difficulty of implementation.

The first training data and the second training data are described in detail below. An X-ray imaging unit may be a specific medical institution, such as a hospital or a physical examination institution. An X-ray imaging operator may be an individual, such as a technician, who operates the X-ray imaging system in the X-ray imaging unit.

In a first case, the first training data is derived from log data of a plurality of X-ray imaging units; and the second training data is derived from log data of same X-ray imaging units.

In the first case, the exposure parameter prediction model trained based on the first training data is a general model widely applicable to all X-ray imaging units. The exposure parameter adjustment model trained based on the second training data is a customized model applicable to the same X-ray imaging units.

For example, the first training data is derived from log data of hospitals A, B, and C. The first training data further includes: (1) exposure parameter historical values in historical exposure operations of the hospital A and historical thickness values of historical exposure operations of the hospital A; (2) exposure parameter historical values in historical exposure operations of the hospital B and historical thickness values of historical exposure operations of the hospital B; and (3) exposure parameter historical values in historical exposure operations of the hospital C and historical thickness values of historical exposure operations of the hospital C. The exposure parameter prediction model trained based on the first training data is a general model widely applicable to various hospitals (for example, the hospitals A to C that provide the first training data, and a hospital D that does not provide the first training data) .

After the exposure parameter prediction model is delivered to each hospital, each hospital may use the exposure parameter prediction model to determine an exposure parameter predicted value in a respective X-ray imaging operation.

In addition, second training data of the hospital A may be used to adjust the exposure parameter prediction model delivered to the hospital A. The second training data of the hospital A further includes: exposure parameter historical predicted values historically outputted by the hospital A using the exposure parameter prediction model, and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values (after observing the exposure parameter historical predicted values, technicians of the hospital A manually adjust the exposure parameter historical predicted values to exposure parameter historical adjusted values) . The exposure parameter adjustment model trained based on the second training data of the hospital A is a customized model applicable to the hospital A. The output of the exposure parameter prediction model delivered to the hospital A is connected to the input by the customized model applicable to the hospital A, so that the exposure parameter prediction model delivered to the hospital A and the customized model of the hospital A can be connected to form an adjusted exposure parameter prediction model applicable to the hospital A.

Similarly, second training data of the hospital C may be used to adjust the exposure parameter prediction model delivered to the hospital C. The second training data of the hospital C includes: exposure parameter historical predicted values historically outputted by the hospital C using the exposure parameter prediction model, and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values (after observing the exposure parameter historical predicted values, technicians of the hospital C manually adjust the exposure parameter historical predicted values to exposure parameter historical adjusted values) . The exposure parameter adjustment model trained based on the second training data of the hospital C is a customized model applicable to the hospital C. The output of the exposure parameter prediction model delivered to the hospital C is connected to the input by the customized model applicable to the hospital C, so that the exposure parameter prediction model delivered to the hospital C and the customized model of the hospital C can be connected to form an adjusted exposure parameter prediction model applicable to the hospital C.

In a second case, the first training data is derived from log data of a plurality of X-ray imaging units; and the second training data is derived from log data of same X-ray imaging operators in the same X-ray imaging units.

In the second case, the exposure parameter prediction model trained based on the first training data is a general model widely applicable to all X-ray imaging units. The exposure parameter adjustment model trained based on the second training data is a customized model applicable to the same X-ray imaging operators in the same X-ray imaging units.

In addition, second training data of a technician Zhang San in the hospital A may be used to adjust the exposure parameter prediction model delivered to the hospital A. The second training data of the technician Zhang San includes: exposure parameter historical predicted values historically outputted by the technician Zhang San using the exposure parameter prediction model, and exposure parameter historical adjusted values obtained after the technician Zhang San adjusts the exposure parameter historical predicted values (after observing the exposure parameter historical predicted values, the technician Zhang San manually adjusts the exposure parameter historical predicted values to exposure parameter historical adjusted values) . The exposure parameter adjustment model trained based on the second training data of the technician Zhang San is a customized model applicable to the technician Zhang San. The output of the exposure parameter prediction model delivered to the hospital A is connected to the input by the customized model applicable to the technician Zhang San, so that the exposure parameter prediction model delivered to the hospital A and the customized model of the technician Zhang San can be connected to form an adjusted exposure parameter prediction model applicable to the technician Zhang San in the hospital A.

Similarly, second training data of the technician Li Si in the hospital C may be used to adjust the exposure parameter prediction model delivered to the hospital C. The second training data of the technician Li Si includes: exposure parameter historical predicted values historically outputted by the technician Li Si using the exposure parameter prediction model, and exposure parameter historical adjusted values obtained after the technician Li Si adjusts the exposure parameter historical predicted values (after observing the exposure parameter historical predicted values, the technician Li Si manually adjusts the exposure parameter historical predicted values to exposure parameter historical adjusted values) . The exposure parameter adjustment model trained based on the second training data of the technician Li Si is a customized model applicable to the technician Li Si. The output of the exposure parameter prediction model delivered to the hospital C is connected to the input by the customized model applicable to the technician Li Si, so that the exposure parameter prediction model delivered to the hospital C and the customized model of the technician Li Si can be connected to form an adjusted exposure parameter prediction model applicable to the technician Li Si.

A technician 10 of each hospital performs imaging protocol selection processing 11 through a human-computer interaction interface on a control host to select an imaging protocol. The control host displays an initial exposure parameter value (usually pre-set) corresponding to the imaging protocol. The technician 10 performs exposure parameter adjustment processing 12 to manually adjust the initial exposure parameter value, and uses the manually adjusted exposure parameter to perform an exposure operation. The control host determines a thickness value of an imaging object in the exposure operation. The control host generates log data 13 including the exposure parameter and the thickness value in the exposure operation. The control host sends the log data 13 to a database 15 in a cloud 14.

The cloud 14 extracts, from the database 15, exposure parameter values in exposure operations of various hospitals that are historical data and thickness values in the exposure operations as first training data. First model training processing 16 is performed on a first artificial neural network model based on the first training data, to train a general exposure parameter prediction model 19 widely applicable to all hospitals.

The cloud 14 delivers the general exposure parameter prediction model 19 to each hospital, and each hospital can use the exposure parameter prediction model 19 to automatically predict an exposure parameter. The control host of each hospital further sends log data in the process of using the exposure parameter prediction model 19 to the database 15.

The cloud 14 further extracts, from the database 15, exposure parameter historical predicted values predicted by a specific hospital using the exposure parameter prediction model 19 and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values as the second training data. Second model training processing 17 is performed on a second artificial neural network model based on the second training data to obtain an exposure parameter adjustment model for the specific hospital. The output of the exposure parameter prediction model 19 delivered to the specific hospital is connected to the input by the exposure parameter adjustment model of the specific hospital, to form an adjusted exposure parameter prediction model 18 applicable to the specific hospital. The exposure parameter prediction model 18 is a dedicated model of the specific hospital. Similarly, a dedicated model can be generated for each hospital.

The technician at each hospital may then select a model from the general exposure parameter prediction model 19 or the exposure parameter prediction model 18 that is applicable to the hospital to perform the exposure operation. In the specific exposure operation, the technician 10 needs to perform imaging protocol selection processing 21, and performs exposure processing 22 based on an exposure parameter determined by a selected model.

FIG. 4 is an exemplary schematic diagram of a human-computer interaction process of X-ray imaging according to an implementation of the present invention. Based on the human-computer interaction process in FIG. 4, the training data can be conveniently acquired. In FIG. 4, human-computer interaction 30 includes a to-be-detected object information display area 31, an imaging protocol display area 35, an exposure button 38, and an exposure parameter display area 40. A to-be-detected object identifier 32, a to-be-detected object gender 33, and a to-be-detected object age 34 are displayed in the to-be-detected object information display area 31.

In the imaging protocol display area 35, a left side neck imaging protocol 36 and a right side neck imaging protocol 37 are exemplarily displayed for users to select specific imaging protocols.

In the exposure parameter display area 40, specific values of tube voltage 41, tube current 42, exposure time 43, exposure dose 44, and exposure density 45 are displayed. Moreover, the technician 10 can adjust the specific values of the tube voltage 41, the tube current 42, the exposure time 43, the exposure dose 44, and the exposure density 45.

After the technician completes the adjustment, the exposure button 38 is triggered, so that exposure can be performed according to the adjusted exposure parameter.

FIG. 5 is a structural diagram of an apparatus for determining an exposure parameter in X-ray imaging according to an implementation of the present invention. As shown in FIG. 5, an apparatus 500 for determining an exposure parameter in X-ray imaging includes:

a determining module 501, configured to determine an X-ray imaging protocol and a thickness value of an imaging object;

a first input module 502, configured to input the thickness value into an exposure parameter prediction model corresponding to the X-ray imaging protocol, where the exposure parameter prediction model is trained based on first training data, and the first training data includes exposure parameter historical values in historical exposure operations based on the X-ray imaging protocol and historical thickness values in the historical exposure operations; and

a first receiving module 503, configured to receive, from the exposure parameter prediction model, an exposure parameter predicted value determined based on the thickness value.

In an exemplary implementation, the apparatus further includes:

a first training module 504, configured to establish a first artificial neural network model; input the first training data into the first artificial neural network model; and train the first artificial neural network model into the exposure parameter prediction model by using the first training data.

In an exemplary implementation, the apparatus further includes: a second input module 505, configured to input the exposure parameter predicted value into an exposure parameter adjustment model corresponding to the exposure parameter prediction model, where the exposure parameter adjustment model is obtained based on second training data, and the second training data includes exposure parameter historical predicted values historically outputted by the exposure parameter prediction model, and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values; a second receiving module 506, configured to receive, from the exposure parameter adjustment model, an exposure parameter adjusted value determined based on the exposure parameter predicted value; and an exposure module 507, configured to perform an exposure operation on the imaging object based on the exposure parameter adjusted value to generate an X-ray image.

In an exemplary implementation, the apparatus further includes: a second training module 508, configured to establish a second artificial neural network model; input the second training data into the second artificial neural network model; and train the second artificial neural network model into the exposure parameter adjustment model by using the second training data.

In an exemplary implementation, the determining module 501 is configured to determine a distance between an X-ray source and a surface point of the imaging object based on a three-dimensional image including the imaging object; determine a thickness value of the imaging object at the surface point based on a distance between the X-ray source and an imaging surface, a distance between a contact plate and a detector, and the distance between the X-ray source and the surface point of the imaging object; and determine the thickness value of the imaging object based on the thickness value of the imaging object at the surface point, where the thickness value of the imaging object includes a minimum thickness value of the imaging object or an average thickness value of the imaging object.

As shown in FIG. 6, the apparatus 600 for determining an exposure parameter in X-ray imaging includes a processor 601, a memory 602, and a computer program stored on memory 602 and capable of being run on the processor 601, where the computer program, when executed by the processor 601, implements any of the foregoing method for determining an exposure parameter in X-ray imaging. The memory 602 may be further implemented into a variety of storage media such as an electrically erasable programmable read-only memory (EEPROM) , a flash memory, and a programmable program read-only memory (PROM) . The processor 601 may be implemented to include one or more central processing units or one or more field programmable gate arrays, where the field programmable gate array integrates one or more central processing unit cores. Specifically, the central processing unit or the central processing unit core may be implemented as CPU, MCU, DSP, or the like.

It should be noted that, not all steps and modules in the foregoing procedures and structural diagrams are necessary, and some steps or modules may be omitted according to actual needs. An execution sequence of the steps is not fixed, and may be adjusted according to needs. Division of the modules is merely functional division for ease of description. During actual implementation, one module may be implemented separately by a plurality of modules, and functions of the plurality of modules may alternatively be implemented by the same module. The modules may be located in the same device or in different devices.

Hardware modules in the implementations may be implemented in a mechanic manner or an electronic manner. For example, a hardware module may include specially designed permanent circuits or logic devices (for example, an application-specific processor such as an FPGA or an ASIC) to complete specific operations. The hardware module may also include temporarily configured programmable logic devices or circuits (for example, including a general processor or another programmable processor) to perform specific operations. The hardware module is implemented by specifically using the mechanical manner, using the application-specific permanent circuits, or using the temporarily configured circuits (for example, configured by software) , which can be decided according to consideration of costs and time.

The present invention further provides a machine readable storage medium, storing instructions used for causing a machine to execute the method described in this specification. Specifically, a system or an apparatus that is equipped with a storage medium may be provided. The storage medium stores software program code that implements functions of any implementation in the foregoing embodiments, and a computer (a CPU or an MPU) of the system or the apparatus is enabled to read and execute the program code stored in the storage medium. In addition, a program code based instruction may also be used to enable an operating system or the like running in the computer to complete some or all actual operations. The program code read from the storage medium may also be written into a memory that is disposed in an expansion board inserted in the computer, or may be written into a memory that is disposed in an expansion unit connected to the computer, and then a CPU or the like that is installed on the expansion board or expansion unit may be enabled to execute some or all actual operations based on the instructions of the program code, to implement the functions of any one of the foregoing implementations. Implementations of the storage medium for providing the program code may include a floppy disk, a hard disk, a magneto-optical disk, an optical memory (for example, a CD-ROM, a CD-R, a CD-RW, a DVD-ROM, a DVD-RAM, a DVD-RW, and a DVD+RW) , a magnetic tape, a non-volatile storage card, and a ROM. Optionally, the program code may be downloaded from a server computer or a cloud by using a communication network.

The foregoing descriptions are merely preferred implementations of the present invention, and are not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

A method (100) for determining an exposure parameter in X-ray imaging, comprising:

determining an X-ray imaging protocol and a thickness value of an imaging object (101) ;

inputting the thickness value into an exposure parameter prediction model corresponding to the X-ray imaging protocol, wherein the exposure parameter prediction model is trained based on first training data, and the first training data comprises exposure parameter historical values in historical exposure operations based on the X-ray imaging protocol and historical thickness values in the historical exposure operations (102) ; and

receiving, from the exposure parameter prediction model, an exposure parameter predicted value determined based on the thickness value (103) .
The method (100) for determining an exposure parameter in X-ray imaging according to claim 1, further comprising:

establishing a first artificial neural network model;

inputting the first training data into the first artificial neural network model; and

training the first artificial neural network model into the exposure parameter prediction model by using the first training data.
The method (100) for determining an exposure parameter in X-ray imaging according to claim 1, further comprising:

inputting the exposure parameter predicted value into an exposure parameter adjustment model corresponding to the exposure parameter prediction model, wherein the exposure parameter adjustment model is obtained based on second training data, and the second training data comprises exposure parameter historical predicted values historically outputted by the exposure parameter prediction model, and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values;

receiving, from the exposure parameter adjustment model, an exposure parameter adjusted value determined based on the exposure parameter predicted value; and

performing an exposure operation on the imaging object based on the exposure parameter adjusted value to generate an X-ray image.
The method (100) for determining an exposure parameter in X-ray imaging according to claim 3, further comprising:

establishing a second artificial neural network model;

inputting the second training data into the second artificial neural network model; and

training the second artificial neural network model into the exposure parameter adjustment model by using the second training data.
The method (100) for determining an exposure parameter in X-ray imaging according to claim 3 or 4, wherein

the first training data is derived from log data of a plurality of X-ray imaging units; and

the second training data is derived from log data of same X-ray imaging units, or the second training data is derived from log data of same X-ray imaging operators in the same X-ray imaging units.
The method (100) for determining an exposure parameter in X-ray imaging according to claim 1, wherein the determining a thickness value of an imaging object (101) comprises:

determining a distance between an X-ray source and a surface point of the imaging object based on a three-dimensional image comprising the imaging object;

determining a thickness value of the imaging object at the surface point based on a distance between the X-ray source and an imaging surface, a distance between a contact plate and a detector, and the distance between the X-ray source and the surface point of the imaging object; and

determining the thickness value of the imaging object based on the thickness value of the imaging object at the surface point, wherein

the thickness value of the imaging object comprises a minimum thickness value of the imaging object or an average thickness value of the imaging object.
An apparatus (500) for determining an exposure parameter in X-ray imaging, comprising:

a determining module (501) , configured to determine an X-ray imaging protocol and a thickness value of an imaging object;

a first input module (502) , configured to input the thickness value into an exposure parameter prediction model corresponding to the X-ray imaging protocol, wherein the exposure parameter prediction model is trained based on first training data, and the first training data comprises exposure parameter historical values in historical exposure operations based on the X-ray imaging protocol and historical thickness values in the historical exposure operations; and

a first receiving module (503) , configured to receive, from the exposure parameter prediction model, an exposure parameter predicted value determined based on the thickness value.
The apparatus (500) for determining an exposure parameter in X-ray imaging according to claim 7, further comprising:

a first training module (504) , configured to establish a first artificial neural network model; input the first training data into the first artificial neural network model; and train the first artificial neural network model into the exposure parameter prediction model by using the first training data.
The apparatus (500) for determining an exposure parameter in X-ray imaging according to claim 7, further comprising:

a second input module (505) , configured to input the exposure parameter predicted value into an exposure parameter adjustment model corresponding to the exposure parameter prediction model, wherein the exposure parameter adjustment model is obtained based on second training data, and the second training data comprises exposure parameter historical predicted values historically outputted by the exposure parameter prediction model, and exposure parameter historical adjusted values corresponding to the exposure parameter historical predicted values;

a second receiving module (506) , configured to receive, from the exposure parameter adjustment model, an exposure parameter adjusted value determined based on the exposure parameter predicted value; and

an exposure module (507) , configured to perform an exposure operation on the imaging object based on the exposure parameter adjusted value to generate an X-ray image.
The apparatus (500) for determining an exposure parameter in X-ray imaging according to claim 9, further comprising:

a second training module (508) , configured to establish a second artificial neural network model; input the second training data into the second artificial neural network model; and train the second artificial neural network model into the exposure parameter adjustment model by using the second training data.
The apparatus (500) for determining an exposure parameter in X-ray imaging according to claim 9 or 10, wherein

the determining module (501) is configured to determine a distance between an X-ray source and a surface point of the imaging object based on a three-dimensional image comprising the imaging object; determine a thickness value of the imaging object at the surface point based on a distance between the X-ray source and an imaging surface, a distance between a contact plate and a detector, and the distance between the X-ray source and the surface point of the imaging object; and determine the thickness value of the imaging object based on the thickness value of the imaging object at the surface point, wherein the thickness value of the imaging object comprises a minimum thickness value of the imaging object or an average thickness value of the imaging object.
An apparatus (600) for determining an exposure parameter in X-ray imaging, comprising: a processor (601) and a memory (602) , wherein

the memory (602) stores an application program executable by the processor (601) , to enable the processor to perform the method (100) for determining an exposure parameter in X-ray imaging according to any one of claims 1 to 6.
A computer-readable storage medium, storing computer-readable instructions, wherein the computer-readable instructions, when executed by the processor, implements the method (100) for determining an exposure parameter in X-ray imaging according to any one of claims 1 to 6.
A computer program product, comprising a computer program, wherein the computer program, when executed by a processor, implements the method (100) for determining an exposure parameter in X-ray imaging according to any one of claims 1 to 6.