Automatic modification of imaging system settings for medical imaging systems

Abstract

A mechanism for determining one or more imaging system settings for a medical imaging system. Initial imaging system settings are defined using previous imaging system settings for performing a medical imaging procedure on a subject to be imaged. The initial imaging system settings are modified using a machine learning method or algorithm to output the imaging system settings to be used.

Description

Technical Field

[0001] This invention relates to the field of medical imaging systems, and more particularly to the selection of imaging system settings for medical imaging systems. Background Technology

[0002] In the medical field, particularly in diagnostic medicine, medical imaging systems are increasingly used to capture medical imaging data. Medical imaging data is especially useful for assessing the condition of the imaged object, for purposes such as diagnosis, evaluation, and / or monitoring.

[0003] Examples of medical imaging systems include ultrasound imaging systems, magnetic resonance (MR) imaging systems, X-ray imaging systems, computed tomography (CT) imaging systems, positron emission tomography (PET) imaging systems, and so on.

[0004] When performing medical imaging procedures using a medical imaging system, the correct selection of medical system settings (also known as medical system parameters) is important for achieving conversational or clinically acceptable images. Examples of medical system settings for ultrasound imaging systems include maximum scan depth, focal length, dynamic range, scan mode, time gain compensation curve, and digital gain. Other examples of medical system settings, as well as those applicable to other medical imaging systems, are known to those skilled in the art.

[0005] Operators of medical imaging systems typically set system parameters manually. However, this requires training such operators and is often operator-specific, for example, because each operator may have their own preferences regarding the characteristics of the medical imaging data. In some approaches, the medical imaging system may utilize pre-programmed default values ​​for system parameters, which can optionally be adjusted by the operator.

[0006] We have always hoped to improve the consistency, reliability, and comparability of medical imaging data. Summary of the Invention

[0007] This invention is defined by the claims.

[0008] According to an example of an aspect of the invention, a computer-implemented method is provided for defining one or more imaging system settings for use in performing a medical imaging process on an object.

[0009] The computer-implemented method includes: obtaining anatomical information indicating one or more anatomical features of an object to undergo the medical imaging procedure; using the anatomical information to obtain one or more imaging system settings from a database, the one or more imaging system settings being used in a previous medical imaging procedure performed on the indicated one or more anatomical features of the object; and using a machine learning method to tune the obtained one or more imaging system settings, the tuned one or more imaging system settings being used by the medical imaging system in performing the medical imaging procedure on the indicated one or more anatomical features of the object. The database includes multiple sets of different imaging system settings for multiple different objects. In other words, the one or more imaging system settings obtained from the database are imaging parameters used to perform a previous medical imaging procedure, and are therefore object-specific.

[0010] This disclosure proposes a technique for improving the determination of imaging system settings using machine learning methods. Specifically, old or previous settings of a medical imaging system used in previous or historical medical imaging procedures for the same object to be imaged are retrieved from a database or otherwise obtained, and used as a baseline or initial condition for modifying the settings using machine learning methods. Previous settings are settings used for the same object (e.g., a patient), such that the modification utilizes object-specific settings and is for the same anatomical features(s) of said object(s).

[0011] The proposed method has been demonstrated to improve the performance of machine learning-guided modifications or selection of imaging system settings. In particular, for iterative modification techniques, the proposed method has been identified as reducing the number of iterations required to perform machine learning-guided modifications or settings of the imaging system settings, for example, compared to methods that utilize default or randomized imaging system settings as a baseline for machine learning method modifications.

[0012] The proposed method provides a technique for more effectively identifying imaging system settings used to perform medical imaging procedures, utilizing, for example, historical image system settings stored in a database. Therefore, new communication routes are used (e.g., to previously unused portions of the system) to improve the performance of the technique for identifying imaging system settings. This method thereby provides values ​​for parameters used to control a specific technical system (i.e., the medical imaging system).

[0013] The proposed method also restricts the (initial) imaging system settings to those used for imaging the same regions of interest (i.e., the same one or more anatomical features) from previous medical imaging procedures. This method has been shown to further improve the performance of imaging system setting selection and / or reduce the number of iterations required to perform iterative modifications or selections of imaging system settings guided by machine learning methods.

[0014] Anatomical information can identify: one or more anatomical features; and / or a view of an object.

[0015] The steps of tuning one or more imaging system settings may include performing an imaging system setting modification process, which includes: obtaining medical imaging data from a medical imaging system using one or more imaging system settings; processing the medical imaging data using machine learning methods to identify one or more changes to be made to one or more imaging system settings; and modifying one or more imaging system settings using the identified one or more changes.

[0016] The steps of tuning one or more imaging system settings may include iteratively performing an imaging system setting modification process until one or more termination conditions are met.

[0017] One or more termination conditions may include: a quality metric of the medical imaging data satisfies one or more predetermined conditions; the imaging setup modification process is executed at least a predetermined number of times; the measure of the difference between the imaging system settings for the current iteration of the imaging setup modification process and the imaging system settings for the immediately preceding iteration of the imaging setup modification process is less than a predetermined value; and / or a override indicator is provided. Other examples will be apparent to those skilled in the art.

[0018] The steps to obtain anatomical information may include: acquiring initial imaging data of the object from a medical imaging system; and processing the initial imaging data to identify anatomical information. Therefore, anatomical information can be derived from the object's initial imaging data.

[0019] Initial imaging data can be obtained by the medical imaging system using random or pseudo-random imaging system settings. In other examples, initial imaging data can be obtained by the medical imaging system using default imaging system settings and / or user-defined imaging system settings.

[0020] Anatomical information can identify one or more anatomical features represented in the initial imaging data. In some examples, anatomical information identifies a view of an object represented by the initial imaging data.

[0021] The steps for processing the initial imaging data may include using a second machine learning method. Therefore, the second machine learning method can be used to process the initial imaging data to identify anatomical data or generate anatomical information, for example, by using one or more classifiers, segmentation algorithms, etc.

[0022] The step of obtaining one or more imaging system settings from a database may include: obtaining purpose information indicating the purpose of medical imaging data to be generated by a medical imaging system performing a medical imaging procedure; and using the purpose information to obtain from the database one or more imaging system settings used by the medical imaging system to perform a previous medical imaging procedure, wherein the previous medical imaging procedure has the same purpose as the medical imaging procedure and is performed on one or more anatomical features represented in the anatomical information.

[0023] This method restricts the (initial) imaging system setup to those used to perform medical imaging procedures with the same purpose as the (current) medical imaging workflow. This method has been shown to further reduce the number of iterations required to perform machine learning-guided modifications or selections of the imaging system setup.

[0024] More specifically, it has been recognized that the same medical imaging system will use different imaging system settings to perform medical imaging procedures with different purposes, such as to produce different types of medical imaging data.

[0025] Purpose information can be defined by user input at the user interface. Other methods include defining purpose information based on medical history data, annotations made by clinicians in the subject's electronic medical records, and / or other data included in the electronic medical records.

[0026] One or more imaging system settings may include one or more settings for capturing medical imaging data. Other suitable examples of imaging system settings include settings for processing (raw) medical imaging data and / or enhancing (raw) medical imaging data.

[0027] Medical imaging systems can be ultrasound imaging systems. Therefore, medical imaging data can include medical ultrasound data. The proposed method is particularly advantageous for use with ultrasound imaging systems because the tuning of the ultrasound imaging system settings significantly affects or influences the quality and values ​​of the resulting medical imaging data. In other forms of medical imaging systems, the effects of different medical imaging settings are less pronounced. However, it should be understood that the proposed method can be applied to any other form of medical imaging system, examples of which are well known in the art.

[0028] Additionally or alternatively, the computer-implemented method includes the step of obtaining purpose information indicating the purpose of medical imaging data to be generated by the medical imaging system performing the medical imaging procedure, and wherein the step of tuning one or more image system settings obtained may further include: using the purpose information to define the number of image system settings to be modified and / or using the machine learning method to modify the amount of image system settings. In some embodiments, for a first purpose of the medical imaging data, the number of image system settings modified is less than the number for a second purpose of the medical imaging data and / or the amount of image system settings modified is less than the amount for the second purpose of the medical imaging data. For example, the first purpose of the medical imaging data is to perform a follow-up on a previous medical imaging procedure.

[0029] A computer program product including a computer program code module is also proposed, which, when run on a computing device having a processing system, causes the processing system to perform all the steps of any of the methods disclosed herein.

[0030] A processing system is also proposed for defining one or more imaging system settings for use in performing medical imaging procedures on objects.

[0031] The processing system is configured to: acquire anatomical information indicating one or more anatomical features of an object to undergo a medical imaging procedure; use the anatomical information to obtain from a database one or more imaging system settings for performing the previous medical imaging procedure on the indicated one or more anatomical features of the object; and use machine learning methods to tune the acquired one or more imaging system settings. The database includes multiple sets of different imaging system settings for multiple different objects.

[0032] The processing system can be modified to perform the functionality of any of the methods disclosed herein, and vice versa.

[0033] An imaging system is also proposed, comprising: a processing system; and a medical imaging system configured to perform a medical imaging process on one or more anatomical features of an object and generate medical imaging data using one or more imaging system settings defined by the processing system.

[0034] The imaging system can also be configured to store in a database one or more imaging settings used by the medical imaging system to perform medical imaging procedures and generate medical imaging data.

[0035] The imaging system may also include a user interface configured to provide a visual representation or display of medical imaging data generated by the medical imaging system.

[0036] A medical imaging system may be an ultrasound imaging system that includes an ultrasound probe configured to capture medical ultrasound data as medical imaging data.

[0037] These and other aspects of the invention will become apparent from the embodiments described below and will be set forth with reference to the embodiments described below. Attached Figure Description

[0038] To better understand the invention and to more clearly illustrate how the invention can be implemented, reference will now be made to the accompanying drawings by way of example only, in which:

[0039] Figure 1 The background workflow is illustrated.

[0040] Figure 2 This is a flowchart of the method proposed in the diagram;

[0041] Figure 3 The diagram illustrates a method for setting up a tuning imaging system;

[0042] Figure 4 The diagram illustrates the steps used to obtain anatomical information;

[0043] Figure 5 The diagram illustrates another step for obtaining the initial imaging system settings;

[0044] Figure 6 The diagram illustrates the processing system; and

[0045] Figure 7 The image system is illustrated. Detailed Implementation

[0046] The invention will be described with reference to the accompanying drawings.

[0047] It should be understood that while the detailed description and specific examples indicate exemplary embodiments of the apparatus, system, and method, they are intended for illustrative purposes only and are not intended to limit the scope of the invention. These and other features, aspects, and advantages of the apparatus, system, and method of the present invention will be better understood from the following description, the appended claims, and the accompanying drawings. It should be understood that the drawings are merely schematic and not drawn to scale. It should also be understood that the same reference numerals are used in all the drawings to indicate the same or similar parts.

[0048] This invention provides a mechanism for defining one or more imaging system settings for use in performing a medical imaging procedure on an object to be imaged. An initial imaging system setting is defined using the imaging system settings used in a previous medical imaging procedure performed on the same object. A machine learning method or algorithm is used to modify the initial imaging system settings to output the imaging system settings to be used.

[0049] In the context of this disclosure, references to any function that processes imaging system settings include processing one or more values ​​of imaging system parameters or settings.

[0050] In the context of this disclosure, medical imaging data is data generated by a medical imaging system. Medical imaging data may, for example, include one or more 2D or 3D medical images of a region of interest of an object, as is known in the art.

[0051] To improve understanding of the context, Figure 1 A workflow 100 in which embodiments can be implemented is conceptually illustrated.

[0052] A medical imaging procedure is performed on object 190 using medical imaging system 195. Examples of medical imaging systems and corresponding procedures are well known in the art. Example medical imaging systems include ultrasound imaging systems, magnetic resonance (MR) imaging systems, X-ray imaging systems, computed tomography (CT) imaging systems, positron emission tomography (PET) imaging systems, etc.

[0053] To perform a medical imaging procedure, one or more imaging system settings 150 are defined for the medical imaging system 195. The values ​​of these imaging system settings are traditionally defined by the operator of the medical imaging system (e.g., during the medical imaging procedure).

[0054] An alternative method for defining the values ​​of imaging system settings is to utilize, for example, a machine learning method executed by processing system 140. One method for using a machine learning method is to execute an imaging system setting modification process 145 to modify one or more values ​​of the imaging system settings. The imaging system setting modification process is configured to modify the initial imaging system settings.

[0055] The configuration modification process may, for example, use machine learning methods to process 141 medical imaging data 160 (obtained from a medical imaging system 195 having existing imaging system settings 150) to identify one or more changes to the imaging system settings to be made. The imaging system settings are then modified.

[0056] In some examples, the imaging system setting modification process 145 is performed iteratively to modify one or more imaging system settings. Therefore, in this method, the iterative setting modification process 145 is repeated until one or more termination conditions 142 are met. Once one or more conditions are met, the iterative modification process ends 143.

[0057] In other examples, the imaging system setup modification process is a single-step procedure, i.e., it is executed without iteration. Such methods can, for example, utilize machine learning algorithms configured as echo state networks; stochastic vector function links; extreme learning machines; or decision trees.

[0058] It should be understood that, typically, before the setup process 145 is executed, one or more values ​​of the settings 150 for one or more imaging systems are initialized to corresponding initial values. Typically, these initial values ​​are randomized or pseudo-randomized.

[0059] The proposed method suggests alternatively setting initial values ​​for one or more imaging system settings based on previous imaging system settings used to perform previous medical imaging procedures on one or more indicated anatomical features of the object. Therefore, the initial conditions for the iterative setting modification process are defined based on the previous imaging system settings for the object (and more specifically, those settings used to image one or more indicated anatomical features of the object). This has been identified as significantly reducing the number of iterations to obtain improved values ​​for the imaging system settings that yield improved medical imaging data.

[0060] Figure 2 This is a flowchart illustrating a computer-implemented method 200 used to define one or more imaging system settings for a medical imaging system to perform a medical imaging process on an object. The medical imaging system may be, for example, an ultrasound imaging system.

[0061] One or more imaging system settings may include one or more settings for capturing medical imaging data. The exact type and form of the settings may, for example, depend on the type of medical imaging procedure to be performed and / or the type of medical imaging system performing the medical imaging procedure. Other suitable examples of imaging system settings include settings for processing (raw) medical imaging data and / or enhancing (raw) medical imaging data.

[0062] Examples of medical imaging system settings for ultrasound imaging systems include maximum scan depth, focal length, dynamic range, scan mode, time gain compensation curve, and digital gain. Other examples of medical imaging system settings, as well as those applicable to other medical imaging systems, are known to those skilled in the art.

[0063] Method 200 includes step 210 of obtaining anatomical information indicating one or more anatomical features of an object to undergo a medical imaging procedure. Step 210 may include, for example, receiving anatomical information (e.g., a timeline indicating the object to undergo imaging of a specific anatomical feature) from user input and / or electronic medical records or databases. Another example method for performing step 210 using initial imaging data is described later in this document.

[0064] In certain examples, anatomical information can identify one or more anatomical features represented in the initial imaging data. In some examples, anatomical information identifies a view of an object represented by the initial imaging data.

[0065] Method 200 also includes a step or process 220 of obtaining from a database one or more imaging system settings for performing a prior medical imaging procedure on one or more indicated anatomical features of the subject. The prior medical imaging procedure is a medical imaging procedure of the same type as the medical imaging procedure to be performed (e.g., ultrasound or CT scan).

[0066] Therefore, step 220 includes using anatomical information to obtain from the database one or more imaging system settings used by the medical imaging system to perform a previous medical imaging procedure, wherein the previous medical imaging procedure was performed on one or more anatomical features represented in the initial imaging data.

[0067] In this way, one or more (initial) imaging system settings are defined based on imaging system settings used to image the same anatomical region or region as those imaged in previous medical imaging procedures. This has been identified as improving the convergence speed of the imaging system settings, such as improving optimization speed. The method also improves the quality of medical imaging data 160 generated using the (unmodified / original) imaging system settings, thereby improving any medical imaging data obtained for tuning previous imaging system settings (e.g., more likely to converge to or approach the imaging system settings that produce medical imaging data with the desired characteristics).

[0068] In some examples, the database used in process 220 may include a collection of one or more imaging system settings for a single object. For example, the database may be an electronic medical record for a single object. This approach avoids the need to search for or otherwise identify the imaging system settings of a specific object.

[0069] In other examples, the database used in process 220 may include, for example, multiple different sets of imaging system settings for multiple different objects. Specifically, each set of imaging system settings may include an object identifier that identifies an object undergoing a medical imaging procedure. Process 220 may include identifying a set(s) of imaging system settings(s) associated with a particular object, for example, by identifying a set(s) of object identifiers(s) associated with an object to undergo a medical imaging procedure. Thus, process 220 may include obtaining object identifiers (e.g., object name, number, etc.) and using the object identifiers to identify (one or more) sets of imaging system settings(s).

[0070] The anatomical information can then be used to process one or more sets of imaging system settings to identify one or more sets of imaging system settings used to image the same anatomical features that have undergone the medical imaging process.

[0071] It is possible that more than one or more imaging system settings sets will be associated with the same object and one or more of the same anatomical features. In these cases, one or more of the most recent imaging system settings sets can be selected in process 210—for example, those used with the most recently performed medical imaging procedure. Thus, each imaging system settings set can be associated with a timestamp, etc.

[0072] As previously identified, the previous medical imaging procedure was the same type of medical imaging procedure as the one to be performed (e.g., ultrasound or CT scan). If the database includes imaging system settings for different types of medical imaging procedures, additional filters or restrictions may be available for selecting one or more sets of medical imaging system settings to match the type of medical imaging system / procedure to be performed on the object.

[0073] Other methods for narrowing down which set of imaging system settings(s) to obtain in step 210 are described later in this document, such as utilizing target information.

[0074] Method 200 further includes step 230 of tuning one or more obtained imaging system settings using machine learning methods. Therefore, step 230 can effectively include performing an imaging system setting modification process. Step 230 can be performed, for example, using the previously described iterative imaging system setting modification process or by using a single-step (i.e., non-iterative) imaging system setting modification process. Other methods will be apparent to those skilled in the art.

[0075] Figure 3 An example of an imaging system settings modification process 230 is provided.

[0076] The exemplary imaging system setup modification process 230 includes step 321 of obtaining medical imaging data from a medical imaging system using one or more imaging system settings. The exemplary imaging system setup modification process 230 also includes step 322 of processing the medical imaging data using machine learning methods to identify one or more changes to be made to the one or more imaging system settings. The exemplary imaging system setup modification process 230 further includes step 323 of modifying the one or more imaging system settings using the identified one or more changes.

[0077] In some examples, process 230 is repeated iteratively (as indicated by the dashed lines) until one or more termination conditions are met. Therefore, the imaging system setup modification process 220 may also include a step 325 to determine whether one or more termination conditions are met. In response to an affirmative determination in step 325, process 230 terminates in step 330. Otherwise, process 230 continues.

[0078] Examples of suitable termination conditions will be obvious to those skilled in the art.

[0079] An example termination condition is that the medical imaging data meets one or more predetermined conditions (e.g., desired quality, desired sharpness, desired contrast level, desired signal-to-noise ratio, etc.). In particular, the termination condition can be a measure or metric of the medical imaging data that meets one or more predetermined conditions. Methods for determining or quantifying the quality and / or properties of medical imaging data are well known in the art. Any such measure or value can be labeled as a "quality metric".

[0080] Campanella, Gabriele, et al., “Towards machine-learned quality control: A benchmark for sharpness quantification in digital pathology.” (Computerized Medical Imaging and Graphics 65 (2018): 142-151), provides an example technique for quantifying the sharpness of medical images. Zhang, Zhicheng, et al., “Can signal-to-noise ratio perform as a baseline indicator for medical image quality assessment.” (IEEE Access 6 (2018): 11534-11543), describes a method for evaluating the signal-to-noise ratio of medical images. Tripathi, Abhishek Kumar, Sudipta Mukhopadhyay, and Ashis Kumar Dhara, “Performance metrics for image contrast.” (2011 International Conference on Image Information Processing. IEEE, 2011), discloses a method for quantifying contrast in imaging data. Other methods for determining any suitable metrics will be readily apparent to those skilled in the art.

[0081] Another example termination condition is that the imaging system settings have been changed by less than a predetermined amount of time to reach a predetermined number of iterations. Yet another example termination condition is that a predetermined number of iterations of the settings modification process has been performed. Yet another example termination condition is the receipt of an overshoot signal or a trigger. Various other examples will readily become apparent to those skilled in the art.

[0082] Therefore, one or more termination conditions may include one or more of the following: the imaging setup modification process is executed at least a predetermined number of times; the measure of the difference between the imaging system settings for the current iteration of the imaging setup modification process and the imaging system settings for the immediately preceding iteration of the imaging setup modification process is less than a predetermined value; and / or an overshoot indicator is provided.

[0083] The proposed method utilizes machine learning to identify one or more changes to the imaging system setup. A machine learning method is any self-trained algorithm that processes input data to produce or predict output data. Here, the input data includes medical imaging data and / or imaging system settings, and the output data includes changes to the imaging system setup.

[0084] Suitable machine learning methods used in this invention will be apparent to those skilled in the art. Examples of suitable machine learning methods include decision tree algorithms and artificial neural networks. Other machine learning methods, such as logistic regression, support vector machines, or Naive Bayes models, are suitable alternatives.

[0085] Artificial neural networks (or simply neural networks) are inspired by the human brain. A neural network consists of layers, each containing multiple neurons. Each neuron performs a mathematical operation. Specifically, each neuron can include different weighted combinations of a single type of transformation (e.g., the same type of transformation, sigmoid, etc., but with different weights). In processing input data, the mathematical operation of each neuron is performed on the input data to produce a numerical output, and the output of each layer in the neural network is sequentially fed into the next layer. The final layer provides the output.

[0086] Methods for training machine learning methods are well-known. Typically, such methods involve obtaining a training dataset that includes training input data entries and corresponding training output data entries. The training output data entries can be defined, for example, by a suitable qualified and / or experienced professional. An initialized machine learning method is applied to each input data entry to generate a predicted output data entry. The error between the predicted output data entry and its corresponding training output data entry is used to modify the machine learning method. This process can be repeated until the error converges and the predicted output data entry is sufficiently similar to the training output data entry (e.g., ±1%). This is often referred to as supervised learning.

[0087] For example, in machine learning methods formed by neural networks, the mathematical operations (weights) of each neuron can be modified until the error converges. Known methods for modifying neural networks include gradient descent, backpropagation, and others.

[0088] The training input data entries correspond to example instances of medical imaging data and / or imaging system settings. The training output data entries correspond to the changes to be made to (one or more) imaging system settings.

[0089] Machine learning methods can be specific to one or more anatomical features to be imaged in a medical imaging procedure and / or the purpose of the medical imaging procedure. This ensures that the medical imaging data output using the medical imaging system settings defined (using machine learning methods) is suitable for the specific medical imaging procedure. Table 1

[0090] Table 1 illustrates the effectiveness of the proposed method when employing an iterative modification process. Specifically, Table 1 quantifies the number of iterations required for the image system settings to converge to the same / similar loss value from one or more different initial values ​​of the image system settings. As shown in Table 1, the number of iterations is significantly reduced (e.g., a 70% improvement) if previous image system settings are identified and used.

[0091] Figure 4 The illustration shows a method for performing step 210 to obtain anatomical information.

[0092] In the illustrated example, step 210 includes a sub-step 410 of obtaining initial imaging data of the object from the medical imaging system. For example, the initial imaging data may have already been generated using a random or pseudo-random imaging system setup.

[0093] In the illustrated example, step 210 includes a sub-step 420 that processes the initial imaging data to identify anatomical information indicating one or more anatomical features of the object represented in the initial imaging data. Sub-step 420 may, for example, include processing the initial imaging data using a second machine learning method (i.e., one that is appropriately trained to generate anatomical information about the object).

[0094] Methods for processing imaging data to identify anatomical information are known in the art and include one or more segmentation and / or classification techniques. For example, anatomical classification techniques can be used to process imaging data to predict the presence or absence of one or more anatomical elements represented in the imaging data and / or to identify anatomical views of the imaging data.

[0095] Various methods are known for processing imaging data to identify anatomical features or structures, and can be used in some examples to generate anatomical information. Example methods may utilize one or more machine learning methods, such as neural networks.

[0096] Khan, Sameer, and Suet-Peng Yong, “A deep learning architecture for classifying medical images of anatomy object.” (2017 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC). IEEE, 2017), discloses a method for identifying the presence / absence of anatomical structures in imaging data. Roth, Holger R. et al., “Anatomy-specific classification of medical images using deepconvolutional nets.” (2015 IEEE 12th international symposium on biomedicalimaging (ISBI). IEEE, 2015), discloses another technique. Several other methods are known.

[0097] US Patent 11,468,567 discloses a method for identifying anatomical views provided by imaging data. European Patent Application Publication No. EP 4136617 A1 discloses another method for identifying views provided by imaging data.

[0098] Figure 5 The diagram illustrates alternative methods for performing step 220.

[0099] In this method, step 220 further includes a sub-step 510 of obtaining purpose information, which indicates the purpose of the medical imaging data to be generated by the medical imaging system performing the medical imaging procedure. The purpose information may, for example, contain code or text data.

[0100] The target information can be defined by user input at the user interface. Therefore, sub-step 510 may include obtaining the target information from the user interface.

[0101] Another approach to defining purpose information is to process the medical data and / or medical records of the object (to be imaged), for example, to identify the purpose for a scheduled medical imaging procedure. For instance, there may be a known or defined correlation or mapping between historical information about a particular object and the purpose for performing the medical imaging procedure.

[0102] By way of example, if cancer is suspected (e.g., as indicated in the notes of a medical record), the purpose of a medical imaging procedure could be to identify any cancerous growth. As another example, if a subject has a high fever and a rapid heart rate, the purpose of a medical imaging procedure could be to identify the presence of an infection. As yet another example, if a stroke score (e.g., a score on the NIH Stroke Scale) was recently recorded and is below a predetermined score, the purpose of a medical imaging procedure could be to identify the presence or absence of a stroke. Many other potential mappings between medical data and the purposes of medical imaging procedures will be apparent to technicians.

[0103] Step 220 further includes a sub-step 520 of obtaining from the database one or more imaging system settings used by the medical imaging system to perform a previous medical imaging procedure using anatomical information and purpose information, wherein the previous medical imaging procedure was performed on one or more anatomical features represented in the initial imaging data and has the same purpose as the medical imaging procedure.

[0104] Therefore, the initial medical imaging system settings are based on those from previous medical imaging procedures used for the same purpose as the upcoming medical imaging procedure. This significantly reduces the number of iterations required to converge the values ​​of one or more imaging system settings.

[0105] Of course, suitable imaging system settings (one or more) may not be available in step 210. This could happen, for example, if this is the first time a medical imaging procedure has been performed on the subject, or if a medical imaging procedure has not yet been performed on the same anatomical region of the subject, or if imaging system settings (one or more) have not been recorded for any such imaging procedure. As another example (if purpose information is used), a medical imaging procedure may not have been performed on the subject for the same purpose.

[0106] In such cases, imaging system settings can be simply selected randomly or pseudo-randomly. In another example, imaging system settings can be set based on default settings used in medical imaging procedures. In yet another example, imaging system settings can be set based on recommended settings for the subject's demographic information. Those skilled in the art will readily understand a wide variety of other procedures and processes.

[0107] In some embodiments, if the imaging system settings are available for objects that share the same anatomical information but not the same purpose information (where applicable), then in step 220, the available system settings for sharing the same anatomical information are obtained.

[0108] In some embodiments, if the imaging system settings are available for objects that share the same objective information (where applicable) but not the same anatomical information, then in step 220, the available system settings that share the same objective information are obtained instead.

[0109] In some embodiments, if the imaging system settings are available to the object but do not share the same anatomical information and / or purpose information (in the case of use), then the available system settings are obtained in step 220.

[0110] In some examples, if no imaging system settings are available for the object, the imaging system settings can be simply selected randomly or pseudo-randomly or based on the default settings used for the medical imaging process in step 220.

[0111] In the previously described embodiments, the step of obtaining the image system settings utilizes only anatomical and target information (if used). However, in some other examples, the step of tuning the image system settings also utilizes anatomical and / or target information.

[0112] In particular, the purpose and / or tuning of a machine learning approach may differ in response to information contained in the anatomical and / or target information. Specifically, the tuning and / or selection of the (available) tuning settings may change in response to the anatomical and / or target information.

[0113] As an example, if the objective information indicates that the purpose of the medical imaging procedure is to capture disease progression, then the imaging system settings that affect or alter the amount of speckle in the medical imaging data should remain unchanged compared to the imaging system settings of previous medical imaging procedures. Changes in speckle represent underlying changes in the microstructure of the target anatomy caused by disease progression. Technicians will be able to easily identify those settings (if modified) that will lead to changes in the speckle in the medical imaging data.

[0114] As another example, if anatomical and / or purpose information indicates that the medical imaging process is a fetal medical imaging process (e.g., the anatomy is fetus or the purpose is obstetrics), there may be no limit to the number or amount of modifications to the imaging system settings, as the lower fetus is expected to change in different triplets, making image quality a higher priority.

[0115] In the context of this invention, the amount of modification is a quantification of the change to each image system setting performed in step 230, for example, which can be expressed as a percentage.

[0116] Depending on the specific use case scenario of the medical imaging process, other examples will be obvious to technicians.

[0117] Therefore, return to the reference. Figure 2Method 200 can be adapted, wherein step 230 of tuning (one or more) image system settings includes further using anatomical information and / or purpose information to perform tuning. Specifically, the anatomical information and / or purpose information may define image system settings that can be tuned or modified in step 230, and / or be provided as input to machine learning methods for performing tuning and / or modification of the image system settings (e.g., to control the extent of tuning performed).

[0118] In other words, the steps of tuning one or more imaging system settings may further include: using anatomical information and / or purpose information to define the number of imaging system settings to be modified and / or using machine learning methods to modify the amount of imaging system settings.

[0119] Specifically, the tuning performed in step 230 may be designed to balance consistency across previous and current medical imaging procedures with consistency between the image quality of the current medical imaging procedure (depending on the precise use case scenario defined by anatomical and / or purpose information).

[0120] In other words, step 230 may include weighting the tuning of the imaging system settings in response to anatomical and / or purpose information. The weighting may represent the desired outcome for a specific use case scenario defined by the anatomical and / or purpose information. Of course, the exact weighting will vary depending on the specific embodiment, as will be apparent to those skilled in the art.

[0121] More specifically, if the purpose of a medical imaging procedure is to perform a follow-up to an earlier medical imaging procedure (i.e., to identify a specific etiology or undesirable anatomical feature (e.g., a tumor or growth) and / or disease progression), the number of image system settings modified and / or the amount of image system settings modified may be less than in cases where the purpose of the medical imaging procedure is not to perform a follow-up to an earlier medical imaging procedure (e.g., for obstetrics or any other procedure that does not investigate etiology or undesirable anatomical features and / or disease progression).

[0122] Those skilled in the art will be able to readily develop processing systems for performing any of the methods described herein. Therefore, each step of the flowchart can represent a different action performed by the processing system and can be executed by the corresponding module of the processing system.

[0123] Figure 6 An example of a suitable processing system 600 is illustrated.

[0124] The various operations discussed above can utilize the capabilities of computer 600. For example, one or more parts of the processing system for modifying imaging system settings (of a medical imaging system) can be incorporated into any element, module, application, and / or component discussed herein. In this regard, it should be understood that system functional blocks can run on a single computer or can be distributed across several computers and locations (e.g., via an Internet connection).

[0125] Computer 600 includes, but is not limited to, PCs, workstations, laptops, PDAs, handheld devices, servers, storage devices, etc. Typically, in terms of hardware architecture, computer 600 may include one or more processors 601, memory 602, and one or more I / O devices 607 communicatively coupled via a local interface (not shown). The local interface may be, for example, but not limited to, one or more buses or other wired or wireless connections, as known in the art. The local interface may have additional elements, such as controllers, buffers (caches), drivers, repeaters, and receivers, to enable communication. Furthermore, the local interface may include address, control, and / or data connections to enable appropriate communication between the aforementioned components.

[0126] Processor 601 is a hardware device for running software that can be stored in memory 602. Processor 601 can be virtually any custom or commercially available processor, central processing unit (CPU), digital signal processor (DSP), or auxiliary processor among several processors associated with computer 600, and processor 601 can be a semiconductor-based microprocessor (in the form of a microchip) or microprocessor.

[0127] Memory 602 may include any one or a combination of the following: volatile memory elements (e.g., random access memory (RAM), such as dynamic random access memory (DRAM), static random access memory (SRAM), etc.) and non-volatile memory elements (e.g., ROM, erasable programmable read-only memory (EPROM), electronically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic tape, optical disc read-only memory (CD-ROM), magnetic disk, floppy disk, cassette tape, magnetic tape cartridge, etc.). Furthermore, memory 602 may contain electronic, magnetic, optical, and / or other types of storage media. Note that memory 602 may have a distributed architecture, wherein various components are located remotely from each other but are accessible by processor 601.

[0128] The software in memory 602 may include one or more individual programs, each comprising an ordered list of executable instructions for implementing logical functions. According to an exemplary embodiment, the software in memory 602 includes a suitable operating system (O / S) 605, a compiler 604, source code 603, and one or more applications 606. As illustrated, application 606 includes numerous functional components for implementing features and operations of the exemplary embodiment. According to an exemplary embodiment, application 606 of computer 600 may represent various applications, computing units, logic, functional units, processes, operations, virtual entities, and / or modules, but application 606 is not intended to be limiting.

[0129] Operating system 605 controls the execution of other computer programs and provides scheduling, input / output control, file and data management, memory management, communication control, and related services. The inventors anticipate that application 606, used to implement exemplary embodiments, can be applied to all commercially available operating systems.

[0130] Application 606 can be a source program, an executable program (object code), a script, or any other entity including a set of instructions to be executed. In the case of a source program, the program is typically translated by a compiler (such as compiler 604), assembler, parser, etc., and may or may not be included in memory 602 to operate appropriately in conjunction with O / S 605. Furthermore, application 606 can be written in an object-oriented programming language with classes of data and methods, or a procedural programming language with routines, subroutines, and / or functions, such as, but not limited to, C, C++, C#, Pascal, BASIC, API calls, HTML, XHTML, XML, ASP scripts, JavaScript, FORTRAN, COBOL, Perl, Java, ADA, .NET, etc.

[0131] I / O device 607 may include input devices, such as, but not limited to, a mouse, keyboard, scanner, microphone, camera, etc. Furthermore, I / O device 607 may also include output devices, such as, but not limited to, a printer, monitor, etc. Finally, I / O device 607 may also include devices that transmit both input and output, such as, but not limited to, a NIC or modulator / demodulator (for accessing remote devices, other files, devices, systems, or networks), radio frequency (RF) or other transceivers, telephone interfaces, bridges, routers, etc. I / O device 607 also includes components for communication over various networks, such as the Internet or intranets.

[0132] If the computer 600 is a PC, workstation, intelligent device, etc., the software in memory 602 may also include a Basic Input / Output System (BIOS) (omitted for simplicity). The BIOS is a collection of basic software routines that initialize and test the hardware at startup, boot O / S 605, and support data transfer between hardware devices. The BIOS is stored in some type of read-only memory (such as ROM, PROM, EPROM, EEPROM, etc.) so that it can run when the computer 600 is activated.

[0133] When the computer 600 is operating, the processor 601 is configured to: run software stored in the memory 602, transfer data to and from the memory 602, and generally control the operation of the computer 600 according to the software. Applications 606 and O / S 605 are read, perhaps buffered within the processor 601, and then executed.

[0134] When application 606 is implemented in software, it should be noted that application 606 can be stored on virtually any computer-readable medium for use by or in connection with any computer-related system or method. In the context of this document, a computer-readable medium can be an electronic, magnetic, optical, or other physical device or module that may contain or store computer programs for use by or in connection with a computer-related system or method.

[0135] Application 606 can be implemented in any computer-readable medium for use by or in conjunction with an instruction execution system, apparatus, or device, such as a computer-based system, a processor-containing system, or other system that can fetch and execute instructions from and from an instruction execution system, apparatus, or device. In the context of this document, "computer-readable medium" can be any module that can store, deliver, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. Computer-readable media can be, for example, but not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, devices, or propagation media.

[0136] Figure 7 An imaging system 700 according to an embodiment is illustrated.

[0137] Imaging system 700 includes processing system 600, which can be implemented as previously described. Imaging system 700 also includes medical imaging system 195, which is configured to perform medical imaging procedures and generate medical imaging data 160 using one or more imaging system settings 150 defined by the processing system.

[0138] The medical imaging system 195 may be, for example, an ultrasound imaging system that includes an ultrasound transducer for capturing ultrasound imaging data or medical ultrasound data (as medical imaging data).

[0139] In some examples, the imaging system is also configured to store in a database one or more imaging system settings used by the medical imaging system to perform medical imaging procedures and generate medical imaging data.

[0140] Therefore, the imaging system 700 may also include a database 750.

[0141] The imaging system 700 may also include a user interface 760 configured to display a visual representation of any generated medical imaging data 160. Methods for controlling the user interface 760 to provide a visual representation of the generated medical imaging data 160 are known.

[0142] It should be understood that the disclosed methods are preferably computer-implemented methods. This also introduces the concept of a computer program comprising code modules for implementing any described method when the program is run on a processing system such as a computer. Therefore, different portions, lines, or blocks of code of the computer program according to embodiments can be run by a processing system or computer to perform any of the methods described herein.

[0143] A non-transient storage medium for storing or carrying computer programs or computer code is also proposed, which, when run by a processing system, causes the processing system to perform any of the methods described herein.

[0144] In some alternative implementations, the functions indicated in one or more block diagrams or flowcharts may not occur in the order shown in the figures. For example, two blocks shown consecutively may operate substantially concurrently, or the blocks may sometimes operate in reverse order, depending on the functions involved.

[0145] By studying the accompanying drawings, disclosure, and claims, those skilled in the art can understand and implement variations of the disclosed embodiments when practicing the claimed invention. Although specific measures are recited in different dependent claims, this does not imply that combinations of these measures cannot be advantageously used.

[0146] In the claims, the word "comprising" does not exclude other elements or steps, and the words "a" or "an" do not exclude a plurality. If the term "suitable" is used in the claims or description, it should be noted that the term "suitable" is intended to be equivalent to the term "configured as." If the term "arranged" is used in the claims or description, it should be noted that the term "arranged" is intended to be equivalent to the term "system," and vice versa.

[0147] A single processor or other unit may perform the functions of several items recounted in the claims. If a computer program has been discussed above, the computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium provided together with or as part of other hardware, but the computer program may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems.

[0148] Any reference numerals in the claims should not be construed as limiting the scope.

Claims

1. A computer-implemented method (200) for defining one or more imaging system settings (150) for use in performing a medical imaging process on an object (190), said computer-implemented method comprising: Obtain (210) anatomical information indicating one or more anatomical features of the object to undergo the medical imaging process; Using the anatomical information, one or more imaging system settings are obtained from a database (220) that were used in a previous medical imaging procedure performed on one or more anatomical features indicated on the object, the database comprising a collection of multiple different imaging system settings for multiple different objects; One or more image system settings obtained by using machine learning methods (230) are tuned for use in the medical imaging system to perform the medical imaging process on one or more anatomical features indicated on the object.

2. The computer-implemented method (200) according to claim 1, wherein, The steps of tuning one or more image system settings include performing an imaging system setting modification process, which includes: (321) Obtain medical imaging data from the medical imaging system using the settings of the one or more imaging systems; The machine learning method described above is used to process (322) the medical imaging data to identify one or more changes to be made to the settings of the one or more imaging systems; and Use one or more of the identified changes to modify (323) the one or more imaging system settings.

3. The computer-implemented method according to claim 2, wherein, The steps of tuning one or more image system settings include iteratively performing the imaging system setting modification process until one or more termination conditions are met.

4. The computer-implemented method according to claim 3, wherein, The one or more termination conditions include: The quality measure of the medical imaging data meets one or more predetermined conditions. The imaging settings modification process is executed no less than a predetermined number of times; The measure of the difference between the imaging system settings in the current iteration of the imaging settings modification process and the imaging system settings in the immediately preceding iteration of the imaging settings modification process is less than a predetermined value; and / or Provides an overshoot indicator.

5. The computer-implemented method according to any one of claims 1 to 4, wherein, The steps to obtain anatomical information include: Obtain initial imaging data of the object from the medical imaging system; and The initial imaging data is processed to identify the anatomical information.

6. The computer-implemented method according to any one of claims 1 to 5, wherein, The steps for obtaining one or more imaging system settings from a database include: Obtaining purpose information, which indicates the purpose of the medical imaging data to be generated by the medical imaging system performing the medical imaging procedure; and Using the stated purpose information and the anatomical information, one or more imaging system settings are obtained from the database for the medical imaging system to perform the previous medical imaging procedure. The preceding medical imaging procedure has the same purpose as the medical imaging procedure and is performed on one or more anatomical features represented in the anatomical information.

7. The computer-implemented method according to any one of claims 1 to 5, further comprising the step of obtaining purpose information, the purpose information indicating the purpose of medical imaging data to be generated by the medical imaging system performing the medical imaging process, and wherein, The step of tuning one or more image system settings obtained further includes: using the target information to define the number of image system settings to be modified and / or using the machine learning method to modify the amount of image system settings.

8. The computer-implemented method of claim 7, wherein the number of modified image system settings for a first purpose of medical imaging data is lower than the number for a second purpose of medical imaging data and / or the amount of modification of the image system settings is lower than the amount for the second purpose of medical imaging data.

9. The computer-implemented method of claim 8, wherein the first objective of the medical imaging data is to perform a follow-up examination of the previous medical imaging procedure.

10. The computer-implemented method according to any one of claims 1 to 9, wherein, The medical imaging system is an ultrasound imaging system, and the medical imaging data is medical ultrasound data.

11. A computer program product comprising a computer program code module, said computer program code module, when run on a computing device having a processing system, causing said processing system to perform all the steps of the method according to any one of claims 1 to 10.

12. A processing system for defining one or more imaging system settings for use in performing a medical imaging process on an object, said processing system being configured to: Obtain anatomical information indicating one or more anatomical features of the object to undergo the medical imaging process; The anatomical information is used to obtain one or more imaging system settings from a database, the one or more imaging system settings being used to perform a previous medical imaging procedure on one or more indicated anatomical features of the object, the database comprising multiple sets of different imaging system settings for multiple different objects; as well as Machine learning methods are used to tune one or more image system settings.

13. An imaging system, comprising: The processing system according to claim 12; as well as The medical imaging system is configured to perform the medical imaging process on the one or more anatomical features of the object and generate medical imaging data using the one or more imaging system settings defined by the processing system.

14. The imaging system according to claim 13, wherein, The medical imaging system is an ultrasound imaging system that includes an ultrasound probe configured to capture medical ultrasound data as the medical imaging data.

15. The imaging system of claim 13 or 14, further configured to store in the database the one or more imaging settings used by the medical imaging system to perform the medical imaging process and generate the medical imaging data.

Images