Radiography system having a storage medium for storing a trained model and a method for producing a trained model

Abstract

To provide a radiological imaging system having a storage medium with a trained model suggesting the need for additional imaging.SOLUTION: In one embodiment, a radiological imaging system is provided, the system including a storage medium having a trained model using, as learning data, a first radiological image obtained using a first condition and a second radiological image obtained by additional imaging using a second condition set on the basis of the first radiological image. The learning data of the trained model has, as annotation information, at least a reason for additional imaging. The trained model outputs, when a third radiological image is input, the need for additional imaging and a reason why the additional imaging is necessary as a re-acquisition factor.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a radiographic imaging system, and more particularly to a technique for analyzing radiographic images using a trained model. [Background technology]

[0002] After acquiring radiological images, they may be reviewed and additional imaging may be required. For example, if the presence of a lesion is suspected but the acquired radiological images do not clearly visualize it, but if selecting different imaging parameters or a different reconstruction method, it is likely that an image that can confirm the lesion can be obtained. In such cases, the technologist may make the decision and, after confirming with the physician, perform additional imaging and image reconstruction. Furthermore, the patient may be informed of the need for additional examinations at a later date, which may result in a time and financial burden on the patient.

[0003] As a method for solving such a problem, a technique has been developed in which features are extracted from the first imaging result and imaging settings for the next imaging are set accordingly (for example, Patent Document 1).

[0004] However, there is still a need for a system that is user-friendly and that clearly explains why additional imaging is necessary and makes accurate decisions. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] International Patent Application Publication No. WO2019 / 112050 Summary of the Invention [Problem to be solved by the invention]

[0006] Although AI can automatically suggest the need for additional imaging, increasing the chances that even inexperienced technicians will notice the need for additional imaging, it is desirable to efficiently generate such AI models and improve the validity of the suggestions for the need for additional imaging. [Means for solving the problem]

[0007] The present disclosure provides a radiographic imaging system having a storage medium with a trained model that uses, as training data, a first radiographic image acquired under first conditions and a second radiographic image acquired by additionally capturing the first radiographic image under second conditions set based on the first radiographic image. The training data for the trained model includes, as annotation information, at least a reason for capturing the additional radiographic image. When a third radiographic image is input, the trained model outputs, as recollection factors, whether additional radiographic image capture is necessary and the reason why the additional radiographic image capture is necessary.

[0008] In another aspect of the present disclosure, a method for producing a trained model is provided. The method includes a step of training a learning model using, as training data, a first radiographic image acquired using first conditions and a second radiographic image acquired by additionally capturing the first radiographic image using second conditions set based on the first radiographic image. The training data used for training includes at least a reason for capturing the additional radiographic image as annotation information. The learning model is trained to become a trained model, and when a third radiographic image is input, the learning model outputs, as recollection factors, the necessity of additional capturing and the reason for the necessity of the additional capturing. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram illustrating a schematic configuration of an X-ray CT system according to an embodiment of the present invention. [Figure 2] 2 is a diagram showing the configuration of the main parts of an X-ray tube and an X-ray detection unit. FIG. [Figure 3] FIG. 1 illustrates a learning stage for generating a trained model. [Figure 4]FIG. 1 illustrates the learning stage and inference stage using a trained model. [Figure 5] FIG. 1 is a network diagram showing a trained model. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of the present invention will be described, but the present invention is not limited thereto.

[0011] FIG. 1 is a block diagram showing the configuration of an X-ray CT device 100 according to this embodiment. In this disclosure, a medical X-ray CT device will be described as an example, but the present invention can also be applied to non-destructive testing devices such as dental CT devices and CT devices for baggage inspection. The X-ray CT device 100 includes an operation console 1, an imaging table 10, and a scanning gantry 20. In a preferred embodiment of the present invention, a medical X-ray CT device that collects projection data from a subject, such as a human or a non-human animal, and reconstructs an image will be described as an example. However, the present invention can also be applied to dental CT devices, baggage inspection CT devices, PET devices, SPECT devices, tomosynthesis devices, and the like.

[0012] The operation console 1 has a computer configuration. Specifically, the operation console 1 includes an input device 2 such as a keyboard or a mouse that accepts input from an operator, a central processing unit 3 that executes scan control processing, preprocessing, image generation processing, etc., and a data acquisition buffer 5 that collects X-ray detector data acquired by the scan gantry 20. The operation console 1 also includes a monitor 6 that displays multi-energy images generated by the image generation processing, etc., and a storage device 7 that stores programs, X-ray detector data, X-ray projection data, dual-energy images, etc. The imaging conditions are input from the input device 2 and stored in the storage device 7.

[0013] The imaging table 10 is provided with a cradle 12 on which the subject 71 is placed and which is moved in and out of an opening 20a (described later) of the scan gantry 20. The cradle 12 is moved up and down and in a horizontal linear motion by a motor built in the imaging table 10.

[0014] The scanning gantry 20 has an opening 20a through which a subject 71 to be imaged is carried. The scanner gantry 20 also has an X-ray tube 21, an X-ray control unit 22 that controls the X-ray tube voltage, X-ray emission timing, etc. of the X-ray tube 21, and a collimator 23 having an aperture that shapes the X-rays emitted from the X-ray tube 21 into a fan-shaped X-ray beam 81. The scanner gantry 20 also has a collimator control unit 27 that controls the aperture of the collimator 23, an X-ray detector 24 that detects the X-rays emitted from the X-ray tube 21, and a data acquisition system (DAS) 25 that collects X-ray detector data (also called raw data) from the output of the X-ray detector 24. The DAS 25 samples analog data received from detector elements of the X-ray detector 24 and converts the analog data into a digital signal for subsequent processing.

[0015] The scanner gantry 20 further includes a gantry rotation unit 15 that holds the X-ray tube 21, collimator 23, and X-ray detector 24 and rotates around the body axis of the subject 71, and a rotation control unit 26 that controls the gantry rotation unit 15. The scanner gantry 20 also includes a gantry control unit 29 that exchanges control signals between the operation console 1 and the X-ray control unit 22, rotation control unit 26, and imaging table 10. In practice, the scanner gantry 20 includes a beam-forming X-ray filter that spatially controls the dose of the X-ray beam 81 and an X-ray filter that controls the radiation quality of the X-ray beam 81, and the gantry rotation unit 15 holds these filters between the collimator 23 and the opening 20a, but illustration and detailed description thereof are omitted here.

[0016] 2 is a diagram showing the configuration of the main parts of the X-ray tube 21 and the X-ray detection unit 24. Here, the vertical direction is defined as the y-axis direction, the direction of transport of the imaging table 10 (which usually coincides with the thickness direction of the X-ray beam 81 or the body axis direction of the subject 71) as the z-axis direction, and the direction perpendicular to the y-axis and z-axis directions (channel direction) as the x-axis direction.

[0017] These components are supported on a predetermined base of the gantry rotating unit 15 and maintain the positional relationship as shown in the figure. That is, the X-ray tube 21 and the X-ray detector 24 are arranged opposite each other with the opening 20a in between. Then, X-rays emitted from the X-ray tube 21 pass through a slit formed by the collimator 23 (not shown in Fig. 2), thereby forming a fan-shaped X-ray beam 81 having a predetermined thickness (cone angle) and spread (fan angle).

[0018] The X-ray tube 21 has a structure in which a cathode sleeve 21s incorporating a focusing electrode and a cathode filament, and a rotating target electrode 21t are housed in a housing 21h, and generates X-rays that diverge from an X-ray focal point F.

[0019] The X-ray detection unit 24 is a so-called multi-row X-ray detector, which is configured by arranging a plurality of, for example, 1,000 X-ray detection elements 24a in a channel direction CH (the direction in which the X-ray beam 81 spreads), and arranging a plurality of, for example, 64 detection element rows in the z-axis direction (the thickness direction of the X-ray beam 81). Here, the detection element rows are numbered 1, 2, 3, . . . , 64 from the end. This realizes a so-called 64-row multi-slice X-ray CT. However, the 64-row detection element row here is merely an example, and the present invention is not limited to this. The X-ray detector 24 forms an X-ray detection surface 24s, which detects the X-ray beam 81 transmitted through the subject 71, using the plurality of X-ray detection elements 24a. The X-ray detection elements 24a are configured as a so-called solid-state detector, for example, by combining a scintillator and a photodiode.

[0020] The central processing unit 3 has a scan control unit 32, a pre-processing unit 34, and an image generation unit 35. The central processing unit 3 is, for example, a processor such as a CPU (Central Processing Unit). The central processing unit 3 executes the functions of the scan control unit 32, the pre-processing unit 34, and the image generation unit 35 by reading and executing a program stored in the storage device 7. The program is an example of an embodiment of the control program according to the present invention.

[0021] The scan control unit 32 controls the X-ray control unit 22, the rotation control unit 26, the collimator control unit 27, and the imaging table 10 via the gantry control unit 29 so as to perform multi-energy imaging of the subject 71. Specifically, the scan control unit 32 controls the above-mentioned units to rotate the X-ray tube 21 and the X-ray detector 24 around the subject 71 and collect X-ray projection data.

[0022] In this embodiment, the X-ray tube 21 can irradiate not only monochromatic X-ray beams but also any number of polychromatic X-ray beams equal to or greater than 2. A tube voltage of any value between 50 and 200 kV, such as 80 kV, 85 kV, 100 kV, 120 kV, 130 kV, 140 kV, 150 kV, or 200 kV, is applied to the X-ray tube 21 by switching it for each view to be acquired, and the X-ray tube 21 irradiates an X-ray beam having an energy spectrum corresponding to the applied tube voltage.

[0023] FIG. 3 shows a trained model production system 300. The trained model production system 300 includes an imaging diagnostic device 310, a data store 320, a model trainer 330, a modeler 340, an output processor 350, a feedback unit 360, a selector 370, and an annotation adding unit 380. Each functional block can be executed by one or more processors. In this embodiment, the imaging diagnostic device 310 in FIG. 3 corresponds to the X-ray CT device 100 in FIG. 1, and the data store 320 corresponds to the storage device 7 in FIG. 1. In another embodiment, the imaging diagnostic device 310 in FIG. 3 corresponds to multiple X-ray CT devices including the X-ray CT device 100 in FIG. 1. In a specific example, the data store 320 is located in a server database to which multiple X-ray CT apparatuses are connected via a network, and the data store 320, the model trainer 330, the modeler 340, the output processor 350, the feedback unit 360, the selector 370, and the annotation adding unit 380 are located in a central processing unit 3 that functions as a trained model production terminal. In another specific example, the data store 320, the data store 320, the model trainer 330, the modeler 340, the output processor 350, the feedback unit 360, the selector 370, and the annotation adding unit 380 are located in a server connected to the X-ray CT apparatus 100 via a network. In yet another specific example, the data store 320, the data store 320, the model trainer 330, the modeler 340, the output processor 350, the feedback unit 360, the selector 370, and the annotation adding unit 380 are located in one or more central processing units 3 associated with multiple imaging diagnostic apparatuses connected via a network. The data store 320, model trainer 330, modeler 340, output processor 350, feedback unit 360, selector 370, and annotation unit 380 may be distributed across a network and processed by multiple processing devices. The image data to be analyzed is stored in the data store 320 (e.g., a database, data structure, hard drive, solid-state memory, flash memory, other computer memory, etc.).The data store 320 may also be configured from multiple storage devices distributed across a network.

[0024] The data store 320 stores a set of one or more images (first radiographic images) acquired by the diagnostic imaging device 310 (such as the X-ray CT device 100 in FIG. 1 ), the acquisition parameters (first acquisition parameters) used at the time of acquisition, and the image reconstruction parameters (first image reconstruction parameters) used at the time of acquisition, and a set of one or more images (second radiographic images) re-acquired by the diagnostic imaging device 310, the acquisition parameters (second acquisition parameters) and the image reconstruction parameters (second image reconstruction parameters) used at the time of re-acquisition, with the two sets linked together. In this example, a set based on a single imaging is defined as the first set, and a set based on a single imaging performed thereafter is defined as the second set, and the two sets are linked together. However, multiple sets based on multiple imaging sessions can be defined as the first set, and a single imaging session performed thereafter can be defined as the second set. Furthermore, the first set based on a single imaging session can be defined as the second set, while the second set can be defined as the set based on multiple imaging sessions. Furthermore, both the first set and the second set can be based on multiple imaging sessions. Generally, the first radiographic images include one or more of spatial resolution, contrast resolution, noise, and artifacts, and second radiographic images may be collected for reasons such as making a more appropriate diagnosis. The second radiographic images included in the second set may be of the exact same imaging region as the first radiographic image (e.g., from the chest to the lower abdomen). In another embodiment, the second radiographic images included in the second set may be of a region that is narrower than the first radiographic image (e.g., the abdominal region corresponding to the kidneys). In still another embodiment, the second radiographic images included in the second set may be of a region that is wider than the first radiographic image or a region that is shifted.

[0025] The selector 370 receives multiple target images from the data store 320 and provides a function for selecting one or more specific cross sections. For example, when the X-ray CT apparatus 1 scans the chest and lower abdomen, tens to thousands of slice images may be generated depending on the slice width. The selector 370 provides various functions for selecting one or more first radiographic images to which annotations are to be added from these images. The selector 370 displays an image of the selected organ or region on the monitor 6 in response to an operator's selection of an organ or region. The selected organ or region is automatically or manually enlarged as needed. The selector 370 may also have an image analysis function. The selector 370 applies a trained AI model to the target image to identify the presence of a specific lesion, specific noise, or specific artifact in the image. The selector 370 may also identify the presence of a specific lesion, specific noise, or specific artifact using template matching or the like instead of an AI model.

[0026] In a preferred embodiment of the present invention, early-stage lesions that are difficult to detect can also be used as learning targets. For example, if a patient undergoes X-ray CT scans every few months or approximately every year and a progressive lesion is found in the CT images, the selector 370 can display past images of the patient in response to an operator's instruction, giving the operator an opportunity to consider why the early-stage lesion was not detected in the past. The operator can then perform annotations, as described below, as needed, and use the results as training data.

[0027] In a preferred embodiment of the present invention, the selector 370 can automatically extract images from the data store 320 to facilitate annotation by the operator. There are various types of artifacts observed in CT images, such as metal objects, Poisson noise, Gaussian noise, streaks, and scatter. These artifacts can also be classified into types such as system design-induced (e.g., Gaussian noise), X-ray tube-induced (e.g., defocus, X-ray tube vibration), detector-induced (e.g., nonuniformity of detector response), patient-induced (e.g., patient movement during scanning, metal objects), and operator-induced (e.g., insufficient exposure (insufficient dose), inappropriate slice width setting). Poor image quality may also be caused by insufficient spatial resolution or contrast resolution. The operator can use the image analysis function of the selector 370 to extract images containing specific factors of poor image quality, specific artifacts, or specific types of artifacts, and display them on the monitor 6.

[0028] In a preferred embodiment of the present invention, a first image in which a specific artifact is present and a second image in which the specific artifact is suppressed are artificially created. That is, since the above-mentioned artifact can be intentionally generated, the image in which the artifact is present can be used as training data. For example, by imaging a phantom in which metal is embedded using general acquisition parameters and reconstruction parameters, a first image in which a metal artifact is present and a set of first acquisition parameters and first reconstruction parameters (first set) can be obtained.

[0029] Then, a second image and a set of second acquisition parameters and second reconstruction parameters (second set) can be obtained using the same phantom with embedded metal and acquisition parameters and reconstruction parameters that are expected to suppress metal artifacts. Machine learning, as described below, can be performed using these artificially generated first and second sets. Images of other types of artifacts can also be obtained by taking images under conditions in which those artifacts occur, and a second set can be obtained by selecting parameters to suppress the artifacts from the same imaging target. Furthermore, only the artifacts and noise artificially generated using the phantom can be extracted from each of the two sets and combined with images obtained by imaging a human body to obtain training data.

[0030] In a preferred embodiment of the present invention, if the operator selects a large number of first and / or second images that are not suitable as training data, the selector 370 outputs a display and / or a sound prompting the operator to further narrow down the first and / or second images to be selected as training data.

[0031] When the selector 370 identifies the first and second images to be used as training data in response to an operator's request, the operator can use the functions of the annotation adding unit 380. In Fig. 3, the annotation adding unit 380 is shown as a separate functional block from the selector 370, but the two can also be implemented as a single functional block.

[0032] Deep learning techniques have made it possible to detect objects in images and classify them. However, learning requires a large number of images and accompanying information (annotations). It is difficult to uniformly collect and annotate the images required for learning. The annotation addition unit 380 provides both automatic and manual annotation functions, reducing the annotation workload.

[0033] In a preferred embodiment of the present invention, the annotation adding unit 380 displays the first set of images selected by the selector 370 on the left side and the second set of images on the right side on the monitor 6. The operator can display the two types of images in various ways according to their preference, for example, by enlarging one image and reducing the other.

[0034] Manual annotation can be performed in various ways. In a specific embodiment of the present invention, the operator identifies the location of the reacquisition factor by surrounding the desired position with a geometric shape, such as a circle, ellipse, polygon including a rectangle, or a complex curve drawn by edge detection. In another embodiment, the operator identifies the location of the reacquisition factor by selecting the center of the reacquisition factor. Once the location of the reacquisition factor is identified, a pull-down menu prompting the operator to identify the type of reacquisition factor is displayed on the display. The operator selects the type of reacquisition factor from the pull-down menu, thereby completing the annotation of the type of reacquisition factor. The location of the reacquisition factor may be identified on the first image, the second image, or both. If the reacquisition factor is an artifact, it is often easy to identify it on the first image, and if the reacquisition factor is a lesion, it is often easy to identify it on the second image. The annotation adding unit 380 can identify the location on one image corresponding to the location specified on the other image based on a technique such as pattern matching. The operator can correct the automatically identified location on the other image as needed. In addition, streaks and the like may extend from one end of the image to the other, and Poisson noise, Gaussian noise, and the like may appear throughout the entire image. In such cases, the location of the recollection factor may be specified for the entire image, or the location of the recollection factor may not be specified.

[0035] As described above, the annotation adding unit 380 provides the operator with the ability to manually add annotations. However, in another preferred embodiment of the present invention, the annotation adding unit 380 can also provide the ability to automatically add annotations. The annotation adding unit 380 applies image recognition to identify recollection factors contained in an image and adds the corresponding type of recollection factor as an annotation. In a specific embodiment, the annotation adding unit 380 presents the type and position of the recollection factor to the operator so that they can be modified. The operator can accept the annotations automatically added by the annotation adding unit 380, reject them and perform all manual annotations, or fine-tune the automatically added annotations and use them.

[0036] The trainer 330 and the modeler 340 receive the first and second sets via the selector 370 and the annotation unit 380. The trainer 330 trains and tests deep learning networks and / or other machine learning networks (e.g., convolutional neural networks (CNNs), generative adversarial networks (GANs), recurrent neural networks (RNNs), random forests, etc.) that are deployed as models in the modeler 340. The trained network models (trained models) are used by the modeler 340 to analyze images. Examples of trained models that may be used include those conventionally used by those skilled in the art, such as YOLO for object detection and ResNet for classification.

[0037] The modeler 340 processes the input new CT image (third radiographic image) using a deployed artificial intelligence model (a trained model, e.g., a neural network, a random forest, etc.) to identify the type and location of the reacquisition factor contained in the CT image and the acquisition and / or reconstruction parameters to be used during reacquisition. The acquisition parameters include the tube voltage, tube current magnitude, slice width, contrast agent protocol, etc. used for dual-energy imaging. The reconstruction parameters include the selection of a reconstruction algorithm, such as analytical image reconstruction, filtered backprojection, iterative reconstruction, iterative reconstruction, model-based iterative reconstruction, or deep learning image reconstruction, the on / off selection of various artifact removal algorithms, and the coefficients used in the reconstruction algorithm. If the reacquisition factor is a lesion, the monitor 6 displays the input radiographic image, the location on the radiographic image of any lesions present or suspected to be present in the subject, the type of lesion, and the reason for the proposed reacquisition. The reason for the proposal can be displayed using a color code to attract the operator's attention. The operator and / or doctor of the X-ray CT device 1 determines the validity of the reacquisition factors and their explanations generated by the modeler 340, and if valid, performs imaging using the X-ray CT device 1 using the acquisition parameters generated by the modeler 340. The operator can display on the monitor 6 a simulation image that is expected to be output when acquired using the acquisition parameters generated by the modeler 340 and reconstructed using the reconstruction parameters generated by the modeler 340. In this case, the monitor 6 displays numerical values ​​corresponding to feature quantities of image quality indexes (including spatial resolution, contrast resolution, noise, artifacts, etc.) of the simulation image. The operator can modify the numerical values ​​of these feature quantities via the input device 2, and a simulation image corresponding to the modified numerical values ​​is displayed on the monitor 6. The acquisition parameters and / or reconstruction parameters generated by the modeler 340 are modified accordingly.

[0038] The trained model can output a contrast agent protocol corresponding to the reacquisition factor, and the contrast agent protocol may include one or more of the type of contrast agent, the density of the contrast agent, the dose of the contrast agent, the contrast agent injection speed, and the waiting time from the injection of the contrast agent to the imaging. The projection data acquired by the imaging is reconstructed using the reconstruction parameters generated by the modeler 340. The operator and / or doctor of the X-ray CT device 1 determines the appropriateness of the reacquisition factor generated by the modeler 340 and its explanation, and, if necessary, modifies the acquisition parameters generated by the modeler 340 and uses them to perform imaging using the X-ray CT device 1. The projection data acquired by the imaging is reconstructed using the reconstruction parameters generated by the modeler 340 or the image reconstruction parameters modified by the operator. If the operator and / or doctor determines the appropriateness of the reacquisition factor generated by the modeler 340 and its explanation, and finds that the factor is not appropriate, additional imaging is not performed.

[0039] In a preferred embodiment of the present invention, the acquisition and / or reconstruction parameters generated by modeler 340 are provided to output processor 350 to generate output for display, storage, processing by another modality system, communication with another device (e.g., a tablet or smartphone), etc. Appropriate advice can also be obtained by transmitting necessary information to a remote physician or medical institution. If reacquisition of images is performed with an X-ray CT scanner of the same modality or another modality capable of more advanced processing, the acquisition and / or reconstruction parameters generated by modeler 340 are transmitted to such X-ray CT scanner or other modality.

[0040] Feedback from the operator can be provided to the trainer 330 via the feedback unit 360, and the trainer 330 can use the feedback to improve the artificial intelligence model. For example, if a recollection proposal is rejected (canceled) and recollection is not performed, a display prompting the operator to enter a reason for the rejection is displayed on the monitor 6. The operator can select a reason, such as incorrect identification of the recollection factors or using a modality different from that suggested by the system. The feedback can be used for model training and / or testing, and can be used to periodically trigger the trainer 330 to regenerate a model that is deployed to the modeler 340. The trainer 330 adjusts the weights of the nodes in the trained model used in the recollection proposal depending on the reason for the rejection. Conversely, if recollection is performed according to the recollection proposal without any corrections, the trainer 330 strengthens the weights of the nodes in the trained model used in the recollection proposal.

[0041] In a preferred embodiment of the present invention, the trained model 460 is transmitted to a central management device that manages multiple X-ray CT scanners and / or other imaging devices. The trained model 460 is automatically or manually retrained, or both, in the X-ray CT scanner 1 and / or other imaging devices. During retraining, the trained model can be retrained using a third radiographic image acquired using the third condition and a fourth radiographic image acquired by additionally capturing the third radiographic image using fourth conditions set based on the third radiographic image as training data. Retraining can be performed for each attending physician, each hospital, or each affiliated hospital group with multiple hospitals. While performing retraining for each attending physician can provide image quality that meets the attending physician's preferences, the amount of training data may be insufficient. Conversely, performing retraining for each affiliated hospital group with multiple hospitals increases the likelihood of obtaining a sufficient amount of training data, but may decrease the likelihood of simultaneously satisfying the different preferences of individual radiologists.

[0042] FIG. 4 illustrates exemplary implementations of trainer 330 and modeler 340. As shown in the example of FIG. 4, trainer 330 includes an input processor 410, a training network 420, an output validator 430, and a feedback processor 440. In this example, modeler 340 includes a preprocessor 450, a trained model 460, which is a deployed model, and a postprocessor 470. In the example of FIG. 4, inputs such as images, collection and / or reconstruction parameters, and annotations received directly from data store 320 or via selector 370 are provided to input processor 410, which selects a training network 420 from multiple training networks 420 corresponding to the input and prepares the data to be input to training network 420. For example, the data can be filtered, supplemented, and / or modified in ways such as sanitizing to make the images, collection and / or reconstruction parameters input to network 420 more suitable for training. The input processor 410 compares the first image with the second image and weights the learning more heavily if there is significant improvement, or weights the learning less heavily if there is not significant improvement. In certain embodiments, the trainer 330 does not have an input processor 410.

[0043] In the illustrated example, the training network 420 analyzes input data from the input processor 410 and generates outputs that are validated by the output validator 430. Thus, the training network 420 receives input image data and develops connections to dynamically form a learning algorithm that identifies acquisition and / or reconstruction parameters depending on the type and location of reacquisition factors contained in the image. The output validator 430 can, for example, verify the accuracy of the output from the training network 420. If the acquisition or reconstruction parameters are outside of an available range, for example, the output is deemed incorrect. If the output of the network 420 is inaccurate, the network 420 can be modified (e.g., by adjusting network weights, changing node connections, etc.) to update the network 420 and generate the correct output. Once the output of the training network 420 is validated by the output validator 430, the training network 420 can be used to generate and deploy a trained model 460 for the modeler 340.

[0044] Periodically, feedback can be provided to a feedback processor 440, which processes the feedback and evaluates whether to trigger an update or regeneration of the network 420. For example, if the output of the deployed trained model 460 remains accurate, there may be no reason to update. However, for example, if the output of the trained model 460 becomes inaccurate, this can trigger a regeneration or other update of the network 420 and deploy an updated trained model 460 (e.g., based on additional data, new constraints, updated configuration, etc.).

[0045] The deployed trained model 460 is used by the modeler 340 to process input image data and identify any recollection factors. Input data, such as from the data store 320, is prepared by the preprocessor 450, which then supplies the input data to a trained model 460 (e.g., a deep learning network model, a machine learning model, another network model, etc.) selected by the preprocessor 450 in response to the input image data. The preprocessor 450 can thin the input image data (e.g., remove even-numbered slices), adjust the contrast, brightness, levels, artifacts / noise, etc., before providing the image data to the trained model 460. The trained model 460 processes the image data from the preprocessor 450 and identifies any recollection factors. Output from the trained model 460 is post-processed by a postprocessor 470, which can clean up, organize, and / or otherwise modify the output data to form a composite 2D image. In certain examples, the post-processor 470 may verify and / or otherwise perform quality checks on the output acquisition and / or reconstruction parameters before storing, displaying, transmitting to another system, etc., the composite image.

[0046] 5 shows a typical deep learning neural network 500. Typical neural network 500 includes layers 520, 540, 560, and 580. Layers 520 and 540 are connected by neural connection 530. Layers 540 and 560 are connected by neural connection 550. Layers 560 and 580 are connected by neural connection 570. Data flows forward from input layer 520 via inputs 512, 514, and 516 to output layer 580 and output 590.

[0047] Layer 520, in the example of Figure 5, is an input layer that includes multiple nodes 522, 524, and 526. Layers 540 and 560 are hidden layers that, in the example of Figure 5, include nodes 542, 544, 546, 548, 562, 564, 566, and 568. Neural network 500 may include more or fewer hidden layers 540 and 560 than are shown. Layer 580 is an output layer that, in the example of Figure 5, includes node 582 having output 590. Each input 512-516 corresponds to a node 522-526 in input layer 520, and each node 522-526 in input layer 520 has a connection 530 to a respective node 542-548 in hidden layer 540. Each node 542-548 in hidden layer 540 has a connection 550 to a respective node 562-568 in hidden layer 560. Each node 562-568 in hidden layer 560 has a connection 570 to output layer 580. Output layer 580 has an output 590 that provides the output from the exemplary neural network 500.

[0048] Of the connections 530, 550, and 570, certain exemplary connections 532, 552, and 572 may be given additional weighting, while other exemplary connections 534, 554, and 574 may be given less weighting in the neural network 500. Input nodes 522-526 are activated, for example, by receiving input data via inputs 512-516. Nodes 542-548 and 562-568 in the hidden layers 540 and 560 are activated by the forward flow of data through the network 500 via connections 530 and 550, respectively. Node 582 in the output layer 580 is activated after data processed in the hidden layers 540 and 560 is sent via connection 570. When output node 582 in the output layer 580 is activated, node 582 outputs an appropriate value based on the processing accomplished in the hidden layers 540 and 560 of the neural network 500. In this way, a model can be created using the image, the recollection factors contained in the image, and their position information for learning object detection, shape detection (segmentation), and classification. Also, predetermined collection parameters and reconstruction parameters can be adjusted based on the image, the recollection factors contained in the image, and their position information.

[0049] The above description focuses on the most preferred embodiment of the present invention, but as will be apparent to those skilled in the art, the present invention can be implemented by making various changes and modifications to the embodiment within the technical scope of the present invention.

[0050] Furthermore, a program for causing a computer to function as each of the means for controlling and processing the X-ray CT apparatus is also an example of an embodiment of the invention. [Explanation of symbols]

[0051] 2 Input devices 3. Central Processing Unit 5 Data Collection Buffer 6 monitors 7 Storage device 10 Photographic Table 12 Cradle 15 Gantry rotating part 20 Scanning Gantry 20a opening 21 X-ray tube 21h Housing 21s Cathode Sleeve 21t target electrode 22 X-ray control unit 23 Collimator 24 X-ray detector 24a X-ray detector 24s X-ray detection surface 25 DAS 26 Rotation control section 27 Collimator control section 29 Gantry control unit 32 Scan control section 34 Pretreatment section 35 Image generation unit 71 Subject / Imaging Subject 81 X-ray beam 100 X-ray CT device 300 Trained Model Production System 310 Diagnostic Imaging Devices 320 Data Store 330 Model Trainer 340 Modeler 350 Output Processor 360 Feedback Unit 370 Selector 380 Annotation Addition Section 410 Input Processor 420 Training Network 430 Output Validator 440 Feedback Processor 450 Preprocessor 460 trained models 470 Post Processor

Claims

1. A radiographic imaging system having a storage medium having a trained model using, as training data, a first radiographic image obtained using a first condition and a second radiographic image obtained by additionally capturing the first radiographic image using a second condition set based on the first radiographic image, the learning data includes at least a reason for taking additional photographs as annotation information; When a third radiographic image is input, the trained model outputs the necessity of additional imaging and the reason why the additional imaging is necessary as re-acquisition factors; the learning data includes the first condition, a radiographic imaging system, wherein the first condition includes a first imaging parameter used when acquiring the first radiographic image and / or a first reconstruction parameter used when reconstructing the first radiographic image.

2. A radiographic imaging system as described in claim 1, wherein the first condition includes a first imaging parameter used when collecting the first radiographic image and a first reconstruction parameter used when reconstructing the first radiographic image.

3. the learning data includes the second condition, 3. The radiographic image capturing system according to claim 2, wherein the second conditions include second imaging parameters used when acquiring the second radiographic image and / or second reconstruction parameters used when reconstructing the second radiographic image.

4. the trained model is configured to identify an image quality index corresponding to the reacquisition factor; The radiation image capturing system according to claim 3 , wherein the image quality index is at least two of spatial resolution, contrast resolution, noise, and artifacts.

5. The radiation image capturing system according to claim 1 , wherein the first radiation image is an image acquired by any one of a CT device, a PET device, a SPECT device, and a tomosynthesis device.

6. the first radiological image is an image acquired from a subject that is a human or a non-human animal; 6. The radiographic imaging system according to claim 5, wherein the reacquisition factors include noise included in the first radiographic image, artifacts included in the first radiographic image, spatial resolution, contrast resolution, and / or information on a lesion present or suspected to be present in the subject.

7. A radiographic imaging system having a storage medium with a trained model that uses as training data a first radiographic image obtained using a first condition and a second radiographic image obtained by additionally capturing the first radiographic image using a second condition set based on the first radiographic image, the learning data includes at least a reason for taking additional photographs as annotation information; When a third radiographic image is input, the trained model outputs the necessity of additional imaging and the reason why the additional imaging is necessary as re-acquisition factors; the trained model is further configured to output a contrast agent protocol corresponding to the reacquisition factor; A radiographic imaging system, wherein the contrast agent protocol includes one or more of the type of contrast agent, the density of the contrast agent, the dose of the contrast agent, the injection speed of the contrast agent, and the waiting time from the injection of the contrast agent until imaging.

8. 4. The radiographic image capturing system according to claim 3, further comprising a user interface including: a display device that displays the third radiographic image, a re-acquisition factor included in the third radiographic image and imaging parameters to be used during re-acquisition, which are identified by the trained model; and an input device that inputs an instruction to perform imaging using the imaging parameters to be used during re-acquisition.

9. 9. The radiographic imaging system according to claim 8, wherein the display device is configured to display a position on the third radiographic image of a lesion that is present or suspected to be present in the subject, a type of the lesion, and a reason for proposing re-acquisition.

10. 9. The radiation image capturing system according to claim 8, wherein the display device is configured to display a simulation image that is expected to be output when acquired using the imaging parameters used during the re-acquisition and reconstructed using the reconstruction parameters used during the re-acquisition.

11. the display device is configured to display a numerical value corresponding to a feature amount of an image quality index of the simulation image; an operator can modify the numerical value via the input device; the display device is further configured to display a simulation image corresponding to the modified numerical value; The radiographic image capturing system according to claim 10 , wherein a reconstruction parameter generating unit of the radiographic image capturing system is configured to output new reconstruction parameters based on the corrected numerical values.

12. The radiation image capturing system according to claim 8 , wherein the input device is configured to allow input of an instruction to cancel imaging using the imaging parameters to be used during the re-collection.

13. the input device is capable of accepting manual modification of imaging parameters to be used during the reacquisition; The radiation image capturing system according to claim 8 , wherein the reconstruction parameter generating unit of the radiation image capturing system is configured to output new reconstruction parameters based on the recollection factors and the corrected imaging parameters.

14. a transmitting unit that transmits the imaging parameters to be used at the time of the re-acquisition and the reconstruction parameters to be used at the time of the re-acquisition to a second imaging device; 9. The radiation image capturing system according to claim 8, wherein the second imaging device is configured to perform imaging based on the imaging parameters to be used during the re-acquisition in response to receiving the imaging parameters to be used during the re-acquisition and the reconstruction parameters to be used during the re-acquisition, and raw data collected by the imaging based on the imaging parameters to be used during the re-acquisition is reconstructed based on the reconstruction parameters to be used during the re-acquisition.

15. 15. The radiographic imaging system according to claim 14, wherein the second imaging device is different from the first imaging device that captured the third radiographic image, and the modality of the first imaging device is the same as or different from the modality of the second imaging device.

16. The system according to claim 3, wherein the imaging parameters used during the re-acquisition are different from the imaging parameters of the third radiographic image in at least one of an acquisition range, an acquisition pitch, the number of energy levels of radiation used for acquisition, and the energy intensity of radiation used for acquisition.

17. 1. A method for producing a trained model, comprising: training a learning model using, as learning data, a first radiographic image obtained under a first condition and a second radiographic image obtained by additionally capturing the first radiographic image under a second condition set based on the first radiographic image; the learning data includes at least a reason for taking additional photographs as annotation information; The learning model is trained to become a learned model, and when a third radiographic image is input, The necessity of additional imaging and the reason why additional imaging is necessary are output as recollection factors; the learning data includes the first condition, A method, wherein the first conditions include first imaging parameters used when acquiring the first radiographic image and / or first reconstruction parameters used when reconstructing the first radiographic image.

18. 18. The method of claim 17, further comprising: retraining the learning model using, as learning data, a third radiographic image acquired using a third condition and a fourth radiographic image acquired by additionally capturing the third radiographic image using a fourth condition set based on the third radiographic image.

Images