System for determining dynamic inspection process

A dynamic examination process using AI-adapted protocols addresses the limitations of predefined medical examination protocols by personalizing the examination to the subject's characteristics, enhancing accuracy and efficiency.

JP2024161335AInactive Publication Date: 2024-11-19SUPERSONIC IMAGINE SA
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2024016542
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-04-25
Filing Date
2024-02-06
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Predefined medical examination protocols are not adaptable to the specific characteristics of individual subjects, leading to inefficiencies and potential inaccuracies in medical data acquisition and analysis.

Method used

A system that determines a dynamic examination process using a predefined algorithm, allowing the process to adapt during execution based on acquired medical data, incorporating artificial intelligence to tailor the examination to the subject's pathological hypothesis.

Benefits of technology

Enhances the accuracy and efficiency of medical data acquisition by personalizing the examination process, reducing unnecessary procedures, and improving diagnostic and prognostic accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024161335000001_ABST
    Figure 2024161335000001_ABST
Patent Text Reader

Abstract

To provide a system for determining a dynamic inspection process.SOLUTION: A processing unit is configured to determine a dynamic inspection process 100 on the basis of a pathology hypothesis by using a predefined algorithm. The dynamic inspection process 100 is associated with acquisition of clinical data. The dynamic inspection process 100 is adaptable during execution on the basis of the acquired clinical data.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Background technology]

[0001] In the field of medicine, the use of predefined protocols can improve the quality of clinical outcomes, e.g. the efficiency of medical examination data of subjects. A system for obtaining medical data, e.g. an ultrasound imaging system, can be equipped with a set of "factory" protocols for predefined applications. The "factory" protocols are programmed by the manufacturer of the system and pre-installed on the system, or predefined by a specific user for a group of users. Such predefined protocols relate, for example, to the examination of human body organs, such as the liver or the breast.

[0002] One advantage of providing such protocols is that even novice users can use the system in an efficient manner by following the relevant predefined protocols. For example, to examine a subject's liver, the user can select the "Liver" protocol and then must perform the examination steps specified by that protocol (e.g., selecting a probe, using predefined parameters of the medical device, holding the ultrasound probe in a specific position on the subject, pressing a "freeze" button or other button on the medical device, etc.).

[0003] However, a drawback of this approach is that the factory protocols are customized by the manufacturer or user for a particular system and / or a particular application or organ scan, but not for the particular subject being examined (i.e., in this example, a patient). Thus, the predefined protocols address a particular application (particularly the examination of a particular organ), but not the idiosyncrasies of the subject currently being scanned. Summary of the Invention

[0004] A system for determining a dynamic inspection process is provided, the system comprising a processing unit configured to determine the dynamic inspection process based on a pathology hypothesis using a predefined algorithm, the dynamic inspection process being further associated (e.g. at run-time) with the acquisition of clinical data, the dynamic inspection process being advantageously adaptable based on the acquired medical data, in particular during its execution.

[0005] By providing such a system, a dynamic inspection process can be realized. In this context, the inspection process may be dynamic by being adaptable during its execution (i.e. dynamically adaptable). Thus, the inspection process is not necessarily completely predetermined but may evolve during its execution. The adaptation of the inspection process, i.e. its dynamic behavior, may thereby be controlled or steered by the acquired medical data.

[0006] For example, the system can be configured to automatically adapt the dynamic inspection process on the fly.

[0007] Before starting the execution of the dynamic inspection process, an initial form of the dynamic inspection process can be defined and / or determined. The initial form can be determined or constructed or selected based on a medical or pathological hypothesis. Thus, the dynamic inspection process can be adapted to the pathological hypothesis of the subject (e.g., a patient). Thus, the basic or initial form of the dynamic inspection process can already be controlled by or adapted to the characteristics of the specific subject. Furthermore, the dynamic inspection process can be adapted during its execution based on newly obtained medical data of the subject, so that the dynamic inspection process can be further customized or adapted to the specific characteristics of the subject (e.g., pathological characteristics discovered based on the obtained medical data). Subsequent stages of the inspection process can be adapted, for example, such that specific medical data is obtained in these additional stages that allow verification, confirmation and / or refinement of discovered (or assumed) pathological features or initial pathological assumptions.

[0008] Thus, the examination process can be advantageously adapted and / or optimized so that the medical data obtained is more accurate, more comprehensive, more reliable, and / or more meaningful for a particular subject. For example, the medical data obtained overall can be more reliable and more accurately evaluated, allowing a user or medical team to better understand, identify, or interpret the subject's pathology, and / or to verify an initial pathological hypothesis, or to fine-tune an initial hypothesis, or to select a pathology within a group of pathologies. This allows for a more efficient examination, saving time and minimizing associated costs.

[0009] A dynamic inspection process may generally define how an examination is performed, including, but not limited to, a set of rules and / or a set of procedures for a medical examination. A dynamic inspection process may define the format and structure of data, such as subject information, diagnostic results, images, and other types of medical records. A dynamic inspection process may outline methods and / or rules for acquiring, transmitting, and / or receiving subject information (i.e., subject data, e.g., medical data). A dynamic inspection process for subject examination may improve the efficiency and accuracy of medical data management.

[0010] Dynamic testing processes for medical testing can improve subject outcomes, particularly when a user of a system performing the dynamic testing process is not skilled enough to manually adapt the testing process to an individual subject's condition. Dynamic testing processes are therefore particularly useful for complex medical conditions and when multiple factors need to be considered and may affect each other.

[0011] Thus, the disclosed system may lead to reduced user dependency (novice vs. expert) on data acquisition and interpretation, with possible automation defining the individual inspection process. The system may further enable: Improved reproducibility of follow-up testing over time and increased standardization and documentation of testing; Providing personalized and optimal testing through precise measurements and appropriate evaluation of pathological biomarkers to improve the accuracy of test results; Improved subject management (a dynamic testing process allows, for example, to avoid unnecessary additional testing, unnecessary interventional procedures, unnecessary follow-up tests, etc., which are typically due to partial and / or inadequate data acquisition leading to misinterpretation of the data) and reduced costs (e.g. reduced manpower for medical professionals and, more generally, reduced healthcare costs). Cost reductions through increased productivity, for example by speeding up the review process.

[0012] An examination may be or may include a medical test. A medical examination may be part of the diagnostic process, and there may be several different types of tests that a medical professional can use to assess a subject's health. A medical examination is a general assessment of the subject's overall health and involves looking, listening, and touching various parts of the body. A medical examination may include blood tests, urine tests, x-rays, magnetic resonance imaging (MRI) tests, computed tomography (CT) tests, and / or ultrasound tests. Blood and urine tests may include laboratory analyses that provide valuable information about the subject's health, such as the presence of an infection or abnormal blood counts. Medical imaging techniques such as x-rays, CT scans, MRI scans, and ultrasound tests are used to create detailed images of the subject's relevant areas, allowing medical professionals to identify potential health problems. These tests help medical professionals make an accurate diagnosis and / or prognosis and develop appropriate treatment plans to help patients manage and recover from their condition.

[0013] A dynamic testing process may be provided for the testing of subjects. The term "subject" may refer to a patient who may be the focus of clinical attention, investigation, or treatment. The term "subject" may include sick or healthy humans undergoing medical care, and animals undergoing treatment by a veterinarian. Subjects who are believed to be healthy may also be tested for prophylactic or preventive purposes, i.e., to avoid the possible development of any disease or condition. Subjects may also be tested during or after medical treatment. In research, the term "subject" is also used to refer to individuals or animals who participate in research as patients or healthy volunteers. The use of the term "subject" may reflect the fact that in medical or research situations, these individuals are the primary focus and observation.

[0014] Determining the dynamic inspection process may be a complex process that involves selecting, generating, constructing, and / or adapting various portions of the dynamic inspection process. Selecting portions of the dynamic inspection process may include reviewing existing medical guidelines and best practices to identify appropriate diagnostic tools, treatment plans, and other clinical recommendations.

[0015] The dynamic inspection process can be determined using a computer-implemented algorithm, for example, an artificial intelligence (AI)-based model or algorithm. In one example, the algorithm used to adapt the dynamic inspection process can be a predefined algorithm or can be linked to a predefined algorithm for determining the dynamic inspection process. For example, the same AI model can be used to first determine the dynamic inspection process and then dynamically adapt it. In a further example, the algorithm used to adapt the dynamic inspection process can be or include a machine learning algorithm, such as, for example, a reinforcement learning (RL) algorithm that uses an intelligent agent to take action to explore an unknown environment (i.e., a specific target). Thus, the algorithm used to adapt the dynamic inspection process can include or involve a Markov decision process (MDP).

[0016] Generating and / or determining parts of the dynamic review process may include using artificial intelligence-based algorithms (e.g., machine learning algorithms) to analyze large medical datasets and identify relevant patterns and trends. Building parts of the dynamic review process may include creating a set of rules and decision trees that guide a user (e.g., a physician) based on a particular subject's characteristics and medical history. For example, a user may be guided to follow certain steps (e.g., types of imaging, types of measurements, etc.) and / or to gather relevant information that leads to an accurate diagnosis and / or prognosis and / or, optionally, to make clinical decisions accordingly. Adapting parts of the dynamic review process may include adjusting the dynamic review process (e.g., in real time and / or during the review process) based on new research findings, subject results, or changes in medical guidelines and best practices. Overall, determining the dynamic review process may include a combination of these different approaches and may be iterative in nature as new data and insights become available (e.g., in real time and / or during the review process).

[0017] The pathological hypothesis may be the pathological hypothesis of the subject. The pathological hypothesis may be a postulated and proposed explanation for the cause and / or nature of a particular disease or disorder. The pathological hypothesis may be established based on a physician, medical expert, doctor, and / or statistics, publications, and / or the physician's empirical knowledge. For example, a particular pathological hypothesis (e.g., liver fibrosis) may be the subject of a particular screening program (e.g., a program evaluating liver fibrosis in a particular population). The physician may use the pathological hypothesis to investigate the underlying mechanism of the disease and, optionally, help identify appropriate treatment and management strategies. The pathological hypothesis may be provided to a predefined algorithm before generating the dynamic examination process. The pathological hypothesis may be set based on the symptoms of the disease in question. The pathological hypothesis may also be based on the underlying biological and physiological knowledge of the affected organ or tissue. In other words, the pathological hypothesis may be external data provided to the system. In another example, the pathological hypothesis may be determined (or predicted) by the system based on the subject's data, as described below. The determined pathological hypothesis may be modified or refined over time and / or in real time while the testing process is running as new data and insights become available (i.e., new medical data is acquired).

[0018] In one embodiment, the pathological hypothesis can be automatically or semi-automatically determined (or predicted) based on the subject's available data, for example, using a computational tool function. For example, determining the pathological hypothesis can be based on the subject's previous medical examination data (i.e., one example of available data). Furthermore, determining the pathological hypothesis can be further based on data related to the subject's symptoms. Also, determining the pathological hypothesis can be based on risk factors related to the subject's health. Furthermore, determining the pathological hypothesis can be further based on the subject's health status. Determining the pathological hypothesis can also be based on the subject's lifestyle.

[0019] A predefined algorithm may be a set of instructions that is established in advance (i.e., prior to execution of the dynamic testing process) and can optionally be used to solve a particular problem or accomplish a particular task. A predefined algorithm can be used to create a dynamic testing process. An algorithm may be designed to analyze and interpret data such as subject characteristics such as age, sex, height and / or weight, the subject's medical history, diagnostic results, and other relevant clinical information.

[0020] The algorithm can use this data to generate a dynamic inspection process and provide guidance on the tests to be performed (i.e., which operations should be followed by the user if the process is not performed fully automatically, e.g., by a robotic system), the data to be acquired (e.g., which mode to use), and / or the tools to be used (e.g., which characteristics or biomarkers to measure, etc.). Examples of pathological hypotheses for which a dynamic inspection process is determined using a predefined algorithm include, but are not limited to, liver disease, musculoskeletal (MSK) disease, sepsis, cancer (e.g., breast cancer), and cardiac disease.

[0021] The term "predefined" may refer to something that may already be established, specified, or preset. A predefined algorithm may be already constructed, established, specified, or set prior to execution of a dynamic inspection process. A predefined algorithm may refer to a specific set of rules, calculations, or instructions that may be preset for use in analyzing medical data and making a diagnosis. A predefined algorithm may be a trained model. A model is developed by training on a data set to learn patterns and relationships in the data. The resulting model can be used to make predictions or classifications for new data.

[0022] Acquisition of medical data involves collecting and storing information about a subject, optionally including the subject's medical history, diagnosis, treatment plan, and other relevant clinical data. Medical data can come from a variety of sources, such as medical devices, electronic medical records, medical imaging, laboratory test results, and / or subject-reported data. Medical data is advantageous for healthcare providers to make informed decisions about the subject's care, optionally including diagnosis and / or prognosis, treatment, and management of medical conditions. Medical data is also used for research purposes, helping to advance understanding of diseases and improve patient outcomes. Acquisition and storage of medical data is conducted in a secure and confidential manner that complies with relevant laws and regulations. Data privacy and security may be important concerns when acquiring medical data, and healthcare providers and institutions may implement appropriate safeguards to protect the subject's confidentiality and prevent data breaches. Acquisition of medical data may be or may include the acquisition of medical data.

[0023] The dynamic inspection process may be associated with acquiring medical data. "Associated" may mean that the dynamic inspection process is linked or connected to the process of acquiring medical data, especially during the execution of the process. In other words, the execution of the process may not only mean acquiring medical data, but also other tasks, such as processing and / or analyzing the medical data, generating reports, etc. (as described in more detail below). The dynamic inspection process may be used to guide the collection and processing of medical data. The dynamic inspection process may also control the process of acquiring medical data (at least semi-automatically, e.g. with limited assistance from a human with a probe). The dynamic inspection process and the collection of medical data may be interdependent, as the dynamic inspection process may collect data according to any set specifications or requirements.

[0024] The dynamic testing process may be adaptable during its execution based on medical data obtained during the subject's testing. As medical data is obtained, the dynamic testing process may be designed to analyze and / or optionally interpret the data in real time, optionally allowing dynamic adjustments to the subject's diagnosis, treatment, and / or care plan. For example, if the subject's results reveal complications or so far unknown symptoms, the dynamic testing process may automatically adapt to provide new recommendations and guidance based on the latest medical evidence and best practices. By leveraging the power of artificial intelligence, the dynamic testing process may provide personalized and responsive medical care, optionally tailored to the needs and pathology of the individual subject. In other words, the dynamic testing process may be adaptive, customized, personalized, or subject-specific. The adaptable dynamic testing process may increase the accuracy of diagnosis and / or prognosis. The adaptable dynamic testing process may further improve the quality and efficiency of medical care by minimizing errors, optionally reducing unnecessary procedures and treatments, and improving subject outcomes. As medical data continues to be collected and analyzed, the dynamic testing process may become even more effective in providing accurate and personalized medical care to subjects. The dynamic testing process may be adaptable on the fly based on other criteria, such as subject-specific factors, external factors, such as the subject's age, medical history, symptoms, response to treatment, etc. External factors may be, for example, changes in medical regulations, new medical research findings, availability of new medical treatments or therapies, etc. Additionally, other criteria may be, for example, personal preferences, cultural and religious considerations, etc.

[0025] The dynamic inspection process may be computer implemented. The dynamic inspection process may include a step, a sequence of steps, and / or a phase or phases. The dynamic inspection process may include operations. In other words, the steps and / or phases may relate to a task or group of tasks, such as acquiring data.

[0026] For example, a dynamic inspection process may be a set of instructions and guidelines that are programmed into a computer system and used to perform a particular task or process. A dynamic inspection process may be an at least partially automated workflow. Thus, at least some operations of the inspection process may be performed by the system. Other operations may be guided and / or controlled by the system but may require assistance by the user (e.g., by holding an ultrasound probe). In any case, the guidance and / or control advantageously makes it much less necessary for the user to have special skills.

[0027] In one aspect, the dynamic inspection process can define at least one entity that can perform or control (or assist) a task.

[0028] An entity may refer to a particular unit, component, and / or element in a dynamic inspection process. An entity may represent a particular procedure, concept, storage, and / or data point. Examples of data points in a dynamic inspection process may include a subject's demographic information, medical history, vital signs, laboratory test results, diagnostic image data, treatment plans, and / or medication instructions. Further examples of entities may include particular procedures, such as procedures for adjusting and / or fine-tuning diagnostic image data, procedures for utilizing at least one of diagnostic data, measuring diagnostic parameters, and evaluating diagnostic descriptors.

[0029] An entity may include a set of rules and / or a set of moves for performing a task. Thus, an entity may be configured to perform a task. However, an entity may also be configured to control or assist in a task or the execution of a task. For example, a task is performed at least in part by a user (or an external device or system) under the control or assistance of the entity. An entity may also configure information that may be relevant for performing a dynamic testing process. Entities in a dynamic testing process may be organized into a standardized structure, such as a database or electronic medical record system, which may allow for easy searching and analysis of the information they contain. This structure may also allow for the use of machine learning algorithms and other computational tools to analyze large medical data sets to identify relevant patterns and trends that can guide clinical decision making. Entities in a dynamic testing process may be advantageous, for example, because they allow health care providers to collect, store, and / or analyze subject data in a standardized manner, which may lead to more accurate diagnoses, more effective treatment plans, improved subject outcomes, more efficient and reproducible testing, and higher productivity.

[0030] The entity may be, for example, an intelligent component, e.g., an analytical and / or processing component. The component may be or include a software module. Thus, the component may be software-based. The component may be, for example, composed of or may be an artificial intelligence (AI)-based algorithm. Thus, the component may be a particular intelligent type of computational and / or processing tool, as described above, and thus may be referred to as an "intelligent component." Thus, an intelligent component may differ from other conventional tools by incorporating algorithms (e.g., AI-based algorithms) that equip the component with a certain level of intelligence or understandability of the particular task handled by the component.

[0031] The component may be configured, for example, to analyze the data and may therefore be referred to, for example, as an analytical component. For example, the component may be configured to derive subject information based on the acquired medical data. This information is an insight into the subject's condition, for example to evaluate the subject's pathology. The component may process various medical parameters related to the examination, such as, for example, biomarkers, measurement parameters, or other relevant data. The component may analyze this data to generate an output, optionally related to the input biomarkers, in the form of another biomarker or a pathological scoring, grading, and / or evaluation.

[0032] A component may also be configured to control a testing process and may therefore be referred to as, for example, a processing component. For example, a component may automate the measurement of a given biomarker. For example, the measurement may be used as input for another component (e.g., focused on data analysis).

[0033] The selection of components in the dynamic inspection process and / or the characteristics of the components used may be adaptive during the execution of the dynamic inspection process, e.g., in real time, based on acquired medical data, i.e., a component may adjust its processing technique or criteria, for example, depending on specific needs related to the current subject.

[0034] In one embodiment, the at least one entity may be adaptable during execution of the dynamic inspection process based on the acquired medical data.

[0035] For example, if a medical device is used to measure a subject's blood pressure during a medical exam, an entity in the dynamic exam process may analyze the resulting medical data and determine that the subject's blood pressure is outside of an expected range for their age and health condition. In response, the entity in the dynamic exam process may adapt by adding a step or phase to the process to address hypertension, such as Doppler measurements or organ elasticity.

[0036] An adaptable entity in a dynamic inspection process may refer to an entity that can adapt its behavior or characteristics in response to changing circumstances or inputs. This may include adding, removing, adjusting, or replaying certain behaviors or steps or phases of the entity. It may also include adapting some parameters of the device used to perform some steps or phases of the inspection. It may also include adding or removing entities. The adaptation can be based on previous inspection results, based on data acquired in "real-time" or pseudo-real-time, or other relevant data. Training data may include knowledge about adaptability created by experts in past tests. Training data can be used to inform the behavior of such adaptable entities in a dynamic inspection process. By analyzing past cases and results, the system can learn how to adjust its approach to get better results in the future. This may include changing the selection of the type of data to acquire, changing the selection of probes or modes, adjusting the measurement tools, or other relevant factors.

[0037] In one embodiment, the tasks and / or entities may relate to at least one of a biomarker (e.g., obtaining a value of a particular biomarker for a subject), a calculation tool, a measurement task and / or a measurement value, an acquisition mode, a probe type, a machine parameter, a machine setting, a scan parameter, an acquisition process for obtaining medical data, a processing tool, a reporting tool, and a selection of medical data.

[0038] The calculated biomarker values ​​for the subject may be determined based on the medical data. A calculation tool may be used, for example, to process the acquired data or to assist a user to acquire the medical data and / or scan the subject. The measurements may be based on or with the medical data. The measurements may be performed automatically and / or manually. Furthermore, the measurements may be performed on an image displayed on a screen. An acquisition mode may be a set of parameters and algorithms, optionally allowing different types of data (e.g., having different physical meanings) to be acquired, calculated and / or displayed.

[0039] The acquisition mode is e.g. B-mode or SWE (ShearWave TM The medical data may be acquired using a probe type, such as a probe for imaging, a medical elastography, or a medical imaging elastography. The probe type may be used to acquire the medical data. Machine parameters and / or settings (e.g., controlled by a user) may be used to control a medical device for acquiring the medical data. The acquisition process for acquiring the medical data may be, for example, according to a particular acquisition mode and / or with a particular probe. A processing tool may be used to process the acquired medical data. The selection of the medical data may be, for example, a selection of an image or may include a selection of an image. Furthermore, the tasks and / or entities may relate to diagnostic support, prognostic support, decision support, and / or second opinions.

[0040] Machine parameters and / or settings of the ultrasound diagnostic device (e.g., controlled by a user) can include settings such as mode, gain, depth, frequency, focus, dynamic range, persistence, compression, harmonic imaging, spatial compounding, speckle reduction, edge enhancement, color Doppler settings, pulsed wave Doppler settings, and power Doppler settings, among others.

[0041] Additionally, the tasks and / or entities may relate to the selection of a particular probe, clinical guidelines, best practice recommendations, and / or medical research findings.

[0042] In one aspect, the dynamic inspection process can define at least one predefined capture mode of the medical device, the capture mode being associated with a group of entities.

[0043] In other words, at least one entity or group of entities may be associated with a predefined acquisition mode of the medical device, which may be or may include a predefined inspection mode.

[0044] In one embodiment, adapting the dynamic inspection process may comprise re-running the first entity or the first group of entities using at least one adapted machine parameter based on the medical data acquired by the first entity or the first group of entities. Additionally, adapting the dynamic inspection process may include selecting and / or adapting a second entity or the second group of entities based on the medical data acquired by the first entity or the first group of entities. Additionally, the first and second groups of entities may be associated with different acquisition modes.

[0045] In one embodiment, adapting the second entity may include at least removing the second entity from the dynamic inspection process. Furthermore, adapting the second entity may include at least adding the second entity to the dynamic inspection process. Adapting the second entity may also include at least adapting at least one parameter of the second entity, such as a machine parameter or another characteristic of the second entity. For example, such another characteristic may characterize the context of the second entity, such as a mode of use or an algorithm associated with the second entity.

[0046] In one embodiment, adapting the second group of entities may include adding the second group of entities to the dynamic inspection process. Further, adapting the second group of entities may include at least selecting the second group of entities for a dynamic inspection process to be performed. Also, adapting the second group of entities may include at least removing the second group from the dynamic inspection process. Further, adapting the second group of entities may include at least deselecting the second group from being performed. Further, adapting the second group of entities may include at least fine-tuning the second group of entities.

[0047] Fine-tuning may refer to the process of making small adjustments or modifications to an existing model or system to improve its validity and / or performance. In the context of a dynamic testing process, fine-tuning may refer to making changes to a second group of entities to better accommodate a particular test or pathological hypothesis. These changes may include adding or removing entities, adjusting parameters of existing entities, changing communication channels between entities, etc.

[0048] In one embodiment, at least one entity is executable using a medical device.

[0049] Examples of medical devices include magnetic resonance imaging machines (MRI), computed tomography machines (CT), X-ray machines, ultrasound machines, dialysis machines, blood pressure monitors, thermometers, blood glucose meters, infusion pumps, ventilators, electrocardiogram (ECG) machines, and pulse oximeters.

[0050] In one aspect, the dynamic inspection process can define a group of interconnected entities, where at least two entities of the group can be dependent on each other.

[0051] In a group of interconnected entities, the entities may be interdependent. For example, the behavior or output of one entity may affect or be affected by another entity in the group. For example, in a dynamic testing process for a medical checkup, data obtained from one test may influence the decision-making process for a subsequent test. Similarly, in an AI algorithm for medical diagnosis, the output of one stage of the algorithm may be used as the input for the next stage. The degree of interdependence between the entities in the group may vary, and some entities may be more influential than others in terms of their impact on the overall performance of the group.

[0052] For example, in a medical device, selecting the appropriate acquisition mode may be important to obtain meaningful and accurate information of a subject, while the contrast setting of the displayed image may not be as important. Understanding the interdependencies between entities in a group may be important to design and optimize the performance of the entire group. Examples of interdependencies between entities may be or include correlations, sequential dependencies, hierarchical dependencies, and / or feedback loops.

[0053] The dependency between two entities may be determined based on confounding factors between the entities. Also, in some cases, the interdependence between two entities may be exploited to improve the results of one of these two entities or to improve the results of a third entity. As a first example, if the first entity is an imaging biomarker and the other entity is another imaging biomarker, and the value of the first biomarker affects the value of the second biomarker, this interdependence may be exploited to improve the results of the third entity, which aims to evaluate a particular disease taking into account the values ​​of the two biomarkers. As a second example, if the first entity corresponds to data from a given imaging mode and the second entity corresponds to data from another imaging mode, the data acquired in the second mode may be used to correct the value of the data related to the first imaging mode. As a third example, if a first entity corresponds to a biomarker from a given imaging mode and a second entity corresponds to another biomarker from a second imaging mode (which may be the same mode as the first imaging mode or a different imaging mode), it may be possible to use the value(s) of the biomarker from the second imaging mode to correct the value(s) of the biomarker from the first imaging mode.

[0054] In one embodiment, at least two entities of a group of interconnected entities can communicate with each other via a digital communication channel optionally established according to a dynamic inspection process.

[0055] A digital communication channel is a medium through which entities exchange information with each other. There are many types of digital communication channels, such as wired and wireless channels, serial and parallel channels, synchronous and asynchronous channels, etc. A digital communication channel may be or include the Internet, a wireless network, or a specialized medical data network. The choice of communication channel depends on factors such as the speed and reliability of the channel, the distance between the entities, and the amount of data that needs to be transmitted. A digital communication channel may also be a virtual channel. For example, if two entities are both software programs, they may also communicate through a communication channel in the form of a program interface (e.g., an API). Thus, a communication channel may include, for example, a data interface or an application programming interface.

[0056] For example, in a medical device that includes a sensor, a processing unit, and a display, the sensor can be connected to the processing unit using a wired communication channel, such as a USB cable, and the processing unit can then transmit data to the display using a wireless communication channel, such as Bluetooth.

[0057] In one aspect, the dynamic inspection process can organize at least a portion of a group of interconnected entities.

[0058] The dynamic inspection process may organize at least a portion of the group of interconnected entities according to an architecture selected from the group including a sequential architecture, a parallel architecture, a hierarchical architecture, and a distributed architecture. The dynamic inspection process may be organized as a network or as a web of interconnected entities.

[0059] A sequential architecture may involve organizing entities in a linear manner, where one entity follows another in a predetermined order. A sequential architecture may be used for processes that have a clear and predictable sequence of stages, such as a medical procedure or treatment process. A parallel architecture involves organizing entities in a way that multiple processes occur simultaneously. A parallel architecture may be used for complex medical procedures or tests that require multiple data points to be processed simultaneously. A hierarchical architecture organizes entities in a tree-like structure, with smaller subgroups of entities grouped under larger categories. A hierarchical architecture may be used to organize medical data in a way that makes it easy to navigate and find specific information. A distributed architecture may involve organizing entities in a way that allows them to be stored and accessed across multiple locations or devices. A distributed architecture may be used for medical data that needs to be accessed from multiple health care providers or facilities.

[0060] The dynamic inspection process may organize a first portion of the group of interconnected entities according to a first architecture and a second portion of the group of interconnected entities according to a second architecture. The selection of an architecture for organizing the entities within the dynamic inspection process may depend on the particular needs and requirements of the clinical scenario.

[0061] In one embodiment, the architecture may be adaptable during execution of a dynamic inspection process based on acquired medical data, including adaptation of interconnections and / or communication channels.

[0062] For example, if new information may become available while the dynamic inspection process is running, the architecture may adapt by dynamically changing communication channels between different entities so that the updated information is appropriately transferred, processed, and / or analyzed. This dynamic adaptation of the architecture may help ensure that the dynamic inspection process remains efficient and effective in achieving its intended goals, even in the face of changing or uncertain medical data.

[0063] The subject's health risk factors are, for example, genetic factors, the subject's age. The subject's health condition may include or be defined by, for example, diabetes. The subject's lifestyle may include or be defined by, for example, junk food fan, smoker, sedentary, etc.

[0064] In one embodiment, the acquired medical data may include data acquired in real time from a subject. Further, the acquired medical data may include at least medical examination data. Also, the acquired medical data may include at least data acquired by at least one medical device, optionally including an ultrasound system.

[0065] In one aspect, the dynamic examination process is adaptable in real time during execution of the digital protocol and / or during an ongoing medical examination.

[0066] The dynamic testing process may be adaptive in real-time or pseudo-real-time based on data it receives while it is running. The dynamic testing process may be designed to monitor the subject's condition and adjust the treatment plan in real-time based on changes in the subject's health. For example, if the subject's vital signs suddenly change, the dynamic testing process can modify the treatment plan to address the new condition. Real-time adaptability may be achieved through the use of fast algorithms, such as pre-trained and customized ML algorithms or other forms of AI algorithms. Such algorithms can analyze data from sensors, medical devices, and other sources to detect patterns and optionally make predictions about the subject's health in real-time. The system can then use this information to dynamically adapt the testing process, adjust which medical data is still to be acquired, and optionally change the diagnostic / prognostic tools used accordingly. Treatment plans, medication dosage adjustments, and other actions can then be changed, respectively, to optimize the subject's treatment.

[0067] The real-time adaptation can be performed such that a user of the system is not or is only distracted by any adaptations. For example, the user may simply be informed (e.g., via a user interface such as a display) of the subsequent step or stage (or a limited number of subsequent steps or stages) to be performed, and optionally the estimated time to complete the full inspection process. Modifications regarding other steps or stages can thus be made in the background without distracting the user.

[0068] In one embodiment, the dynamic testing process can define at least one entity configured to estimate a biomarker for a subject based on acquired medical data for the subject. Further, the dynamic testing process can define at least one entity configured to evaluate a pathological hypothesis based on the one or more estimated biomarkers.

[0069] In one embodiment, the pathology hypothesis may constitute a hypothesis of a pathologies family. Furthermore, the dynamic testing process may define at least one entity configured to evaluate a pathology hypothesis or a particular member of a pathologies family as a function of the obtained medical data.

[0070] In one aspect, the predefined algorithm may be or may include an artificial intelligence (AI) model.

[0071] In a further aspect, the at least one entity may be or may include an artificial intelligence (AI) model.

[0072] Examples of predefined algorithms include AI models such as decision trees, random forests, logistic regression, gradient boosting, and / or support vector machines. Predefined algorithms can be used to perform classification and prediction tasks such as image recognition. For example, a decision tree algorithm can be used to classify images of objects based on their features. Other examples of predefined algorithms can include neural networks, k-nearest neighbors, and / or clustering algorithms, which can be used for tasks such as anomaly detection and data segmentation. AI can refer to the development of computer systems and algorithms that perform tasks that typically require human intelligence, such as visual recognition, speech recognition, decision-making, and language translation. AI systems and algorithms are trained with large datasets and advanced algorithms to identify patterns and trends, make predictions, and optimize outcomes.

[0073] The present disclosure also relates to a computer-implemented method of determining a dynamic inspection process for a medical inspection, the method comprising: determining the dynamic inspection process based on a pathological hypothesis using a predefined algorithm, the method further comprising: the dynamic inspection process being associated with acquisition of medical data, the dynamic inspection process being adapted during its execution based on the acquired medical data.

[0074] The present disclosure also relates to a data processing system comprising means for performing the method according to any embodiment of the present disclosure.

[0075] The present disclosure also relates to a computer program comprising instructions which, when said program is executed by a computer, cause said computer to carry out the method according to any embodiment of the present disclosure.

[0076] Where the method operations may involve any aspect (e.g. physical) beyond mere data processing (e.g. ultrasonic signal processing), the computer program may further include computer readable instructions which, when executed by a data processing system, cause any external element of the system (e.g. ultrasonic transducer device) to perform these operations.

[0077] The present disclosure also relates to a computer readable medium comprising instructions that, when executed by a computer, cause the computer to perform a method according to any embodiment of the present disclosure.

[0078] Dynamic inspection processes may have many advantages in medical imaging. For example, dynamic inspection processes can improve the accuracy of ultrasound examinations by providing access to personalized care and utilizing optimal, adapted, and targeted tools (e.g., tools identified based on pathological hypothesis assumptions) that are automatically called by the dynamic inspection process. Dynamic inspection processes can also improve subject management by providing targeted examinations and reducing healthcare costs. Dynamic inspection processes can improve productivity by optimizing workflow and expediting examinations. Additionally, dynamic inspection processes can improve standardization and reproducibility of examinations by providing consistent modes and measurements. Dynamic inspection processes can improve documentation by automatically incorporating pathology-related reference values ​​and publications (or targeted publication sources) into reports. Dynamic inspection processes can also reduce operator dependency and increase examination reproducibility by optionally bridging the gap between experts and non-experts (e.g., technicians, or non-expert physicians) through automation and systemization. Finally, dynamic testing processes may increase the reproducibility of a specific process-defined sequence over time in follow-up tests (e.g., physical examinations or ultrasound examinations).

[0079] A processing unit may be a component of an electronic device that is responsible for performing computational tasks. There may be different types of processing units, each designed for a specific purpose. A processing unit may be or include a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a field programmable gate array (FPGA), and / or an application specific integrated circuit (ASIC). A CPU is the main processing unit of a computer and executes most of the instructions and calculations required by the computer. A GPU is a specialized processing unit designed to handle the complex calculations required for rendering graphics and video. A DSP is a specialized processing unit that handles signal processing tasks, and an FPGA is a reconfigurable processing unit that is programmed to perform various computational tasks. An ASIC is a customized processing unit designed to perform a specific set of tasks, optionally very efficiently and effectively for its intended purpose. These processing units are found in a wide range of electronic devices, such as medical electronic devices. Medical electronic devices may include computers, smartphones, and other digital devices, and optionally enable these devices to perform various computational tasks efficiently and accurately. Methods according to the present disclosure may also be performed remotely, on cloud-based and / or virtual servers.

[0080] Except where inconsistent, it is contemplated that combinations of the elements described above with those herein are possible.

[0081] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are provided for purposes of explanation and are not restrictive of the disclosure, as claimed.

[0082] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to support and explain the principles thereof. [Brief description of the drawings]

[0083] [Figure 1] FIG. 1 illustrates a schematic diagram of a dynamic examination process 100 for medical examination of a subject according to an embodiment of the present disclosure.

[0084] [Figure 2A] FIG. 2A illustrates another schematic example of a dynamic inspection process 100 according to an embodiment of the present disclosure.

[0085] [Figure 2B] FIG. 2B illustrates an architectural adaptation of the dynamic inspection process 100 of FIG. 2A.

[0086] [Figure 3A] FIG. 3A shows a first example of a dynamic examination process 100 for the assessment of liver disease, which includes several stages (S1-S8).

[0087] [Figure 3B] In FIG. 3B, the dynamic inspection process 100 selects entities relevant to the first four stages (S1, S2, S3, S4) of the evaluation process of FIG. 3A.

[0088] [Figure 3C] FIG. 3C illustrates a dynamic inspection process 100 in which entities relevant to further steps S5 and S6 of the evaluation process of FIG. 3A are selected.

[0089] [Figure 3D] FIG. 3D illustrates a dynamic inspection process 100 in which relevant entities are selected at a further stage S7 of the evaluation process of FIG. 3A.

[0090] [Figure 3E] FIG. 3E illustrates a dynamic inspection process 100 in which relevant entities are selected at a further stage S8 of the evaluation process of FIG. 3A.

[0091] [Figure 3F] FIG. 3F illustrates a second example of a dynamic screening process 100' for evaluating liver disease that includes several stages.

[0092] [Figure 4A] FIG. 4A illustrates a schematic flow chart for determining a dynamic inspection process 100 using an AI model according to an embodiment of the present disclosure.

[0093] [Figure 4B] FIG. 4B is a schematic diagram of a clinical evaluation process according to an embodiment of the present disclosure.

[0094] [Figure 5A] FIG. 5A is a schematic diagram of an ultrasound system 10 according to an embodiment of the present disclosure.

[0095] [Figure 5B] FIG. 5B is a schematic diagram of an exemplary dynamic inspection process 100 of the ultrasound system 10 of FIG. 5A. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0096] Description of the drawings Reference will now be made in detail to the embodiments of the present disclosure, which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.

[0097] In general, the present disclosure relates to a system and method for determining a dynamic examination process, as described in more detail in the context of the examples below. The system for determining a dynamic examination process can include the use of techniques to develop a systematic and / or highly customized approach to obtain reliable, comprehensive, and meaningful medical data of a subject. The system is further designed to streamline, for example, the medical data acquisition process, medical decision-making and treatment process, reduce the possibility of errors, and improve the efficiency of medical care delivery.

[0098] With the assistance of the dynamic inspection process according to the present disclosure, it is possible to process large amounts of medical data, which will result in increased accuracy of diagnosis in the medical field and improved recommendations for treatment. Furthermore, computational tools allow automation of routine tasks and decision-making processes, allowing medical professionals to focus on more complex and important tasks, improving efficiency and reducing costs. Personalized care may be possible by analyzing patient data and suggesting tailor-made treatments based on the patient's medical history and current condition. The dynamic inspection process may provide medical professionals with standardized and consistent treatment guidelines, reducing errors, improving patient safety, and leading to improved patient outcomes. Overall, the combination of dynamic inspection processes and computational tools has the potential to revolutionize the medical field, providing standardized and efficient treatments, improving patient outcomes, and reducing costs while maintaining patient safety.

[0099] FIG. 1 illustrates a schematic diagram of a dynamic examination process 100 for a medical examination of a subject according to an embodiment of the present disclosure. The dynamic examination process 100 defines entities 200 capable of performing tasks. The entities 200 are interconnected and can communicate with each other via digital communication channels 300 established according to the dynamic examination process 100. The dynamic examination process 100 is designed to define how the entities 200 interact during the examination process, and the communication channels 300 facilitate data transfer between the entities 200. The dynamic examination process 100 enables the execution of a medical examination of a subject. The entities 200 can dynamically adapt during the execution of the dynamic examination process 100 based on the medical data obtained, allowing real-time adjustment and optimization of the examination process. The digital communication channels 300 facilitate communication between the entities 200, allowing exchange of information and coordination of tasks.

[0100] The acquisition mode 400 responsible for acquiring medical data from a subject is associated with a group of entities 201, 202, and 203. The entities 201, 202, and 203 are capable of performing specific tasks related to the acquisition of medical data and are connected and communicate with each other via a digital communication channel 300 established by the dynamic inspection process 100. The specific tasks may include guidance and measurement tools, as well as tools for evaluating tissue properties such as stiffness and elasticity and viscoelasticity.

[0101] At the captured time, the entities 201, 202, 203 associated with the capture mode 400 are activated. The entities 201, 202, and 203 associated with the capture mode 400 are interdependent.

[0102] 2A shows another schematic example of a dynamic inspection process 100 according to an embodiment of the present disclosure. The dynamic inspection process 100 includes a group of entities 401. During the execution of the dynamic inspection process 100, new information (i.e., medical data of the subject) is obtained by the group of entities 401, so that the architecture needs to be adapted by dynamically changing the entities and the communication channels 300 between the different entities.

[0103] FIG. 2B illustrates the dynamic inspection process 100 of FIG. 2A to which an architecture has been adapted.

[0104] The dynamic inspection process 100 includes a group of entities 401 that remain unchanged from Fig. 2A in Fig. 2B. Based on the obtained medical data, a new group of entities 402 is added to the dynamic inspection process 100. Entities 200 that do not belong to either group 401 or 402 are removed, resulting in a new architecture for the dynamic inspection process 100. Dynamic adaptation of the architecture may help ensure that the dynamic inspection process 100 remains efficient and effective in achieving its intended goals, even in the face of changing or uncertain medical data.

[0105] In other words, the dynamic inspection process 100 is adapted based on the medical data acquired by the first group of entities 401. The second group of entities 402. Furthermore, the first and second groups of entities 401, 402 may be associated with different acquisition modes.

[0106] FIG. 3A shows a first example of a dynamic examination process 100 for the assessment of liver disease, which includes several stages (S1-S8).

[0107] In the first stage S1, automated measurements may be used to assess liver size and volume. In the second stage S2, a B-mode ratio tool may be used. For example, the B-mode ratio tool may determine the relative brightness between two different media or between two different regions within a media, e.g., between the liver and the kidney. On the B-mode image, the ratio may be calculated, e.g., between the echogenicity of the liver parenchyma and the kidney cortex. The third stage S3 includes application of a liver surface regularity tool, which may be an example of an intelligent analysis and / or processing component, as described above. In the fourth stage S4, a liver sound speed and attenuation tool for steatosis assessment is used. The four stages S1, S2, S3, and S4 may be four independent stages that depend on (or are based on) B-mode data.

[0108] In a fifth step S5, a tool for liver stiffness assessment is applied, such as a box / Q-box tool, which may be a further example of an intelligent analysis and / or processing component, as described above. A sixth step S6 includes a liver viscosity tool for activity assessment. Steps S5 and S6 may be implemented using, for example, a ShearWave TMThe seventh stage, S7, consists of the use of a component for "Diffuse Liver Disease" and allows for a multi-parametric analysis, optionally taking into account stiffness values, viscoelastic values, and analysis of confounding factors. Analysis in this stage can include anatomy, fibrosis, steatosis, activity staging, risk, and subject management recommendations. The component in stage S7 can receive as input the output of the components in stages S3 and S5.

[0109] Finally, the evaluation can end in an eighth stage S8 including a report, generated based on the output of the component for "Diffuse Liver Disease" in stage S7, optionally including a specific reference to the pathology (e.g. any data making it possible to assess the validity of the initial pathological hypothesis).

[0110] As shown in Figures 3B to 3E below, the dynamic inspection process can be adapted during its execution by selecting an appropriate entity at each instance of progression and optionally providing this entity with information such as already obtained medical data. This entity is then activated, for example, for performing or controlling a task. Thus, in this example, the architecture of the dynamic inspection process is not adapted (e.g., by adding or removing entities or communication channels), but by the respective selection of predefined entities. Nevertheless, the selected entities can be configured based on already obtained medical data and / or pathological hypotheses.

[0111] In FIG. 3B, the dynamic inspection process 100, through the use of entity 200, selects entities relevant to the first four stages (S1, S2, S3, S4) of the evaluation process of FIG. 3A.

[0112] FIG. 3C illustrates a dynamic inspection process 100 in which entities relevant to further steps S5 and S6 of the evaluation process of FIG. 3A are selected.

[0113] Figure 3D shows a dynamic screening process 100, where relevant entities are selected in a further step S7 of the assessment process of Figure 3A. In step S7, the "Diffuse Liver Disease" component is applied.

[0114] Figure 3E shows a dynamic inspection process 100, where the relevant entities are selected for a further step S8 of the evaluation process of Figure 3A. Step S8 consists of a report that may be generated based on the output of the component for "diffuse liver disease" of step S7, optionally including a specific reference to pathology (e.g. any data making it possible to assess the validity of an initial pathological hypothesis).

[0115] FIG. 3F illustrates a second example of a dynamic screening process 100' for evaluating liver disease that includes several stages.

[0116] Conventionally, ultrasound imaging systems can be equipped with "factory" protocols, for example for abdominal examinations. Since the steps of such protocols are predefined, these steps or the protocol in general are not adapted to the particular subject to be examined. On the one hand, therefore, there is a risk that the predefined protocol contains examination steps that are unnecessary for a particular subject. Thus, a predefined protocol implies a waste of time for the acquisition of medical data and the analysis of examination results. On the other hand, there is a risk that the predefined protocol lacks steps that are necessary to obtain relevant information about a particular subject. Thus, the final examination results may be inaccurate or even inaccurate.

[0117] In view of these shortcomings, the present disclosure provides a dynamic examination process that can be automatically adapted to a particular subject during the acquisition of medical data. For example, as described in the context of FIG. 3F, an ongoing ultrasound examination process can be automatically adapted in real time during its execution as a function of the obtained information of the subject. The steps described below are not necessarily sequential. Rather, the steps can be performed simultaneously. Furthermore, if the steps are performed sequentially, the order can be different from the example of FIG. 3F. Furthermore, some of the illustrated steps can be skipped if it is determined that they are not necessary in view of the results of the steps already performed. At the same time, other steps not shown in this embodiment can also be added if it is determined that they are necessary in view of the results of the steps already performed.

[0118] In step O, a pathology hypothesis may be provided, for example as a manual input or based on a machine-based determination as described above. The pathology hypothesis may be associated with a pathology family. For example, the pathology family may be hypothesized to be diffuse liver disease.

[0119] At step A1, a B-mode pattern may be evaluated based on the acquired B-mode data of the liver. Such a pattern may include, for example, homogeneity and / or brightness pattern information associated with the liver.

[0120] In step A2, morphological measurements can be made based on the acquired B-mode data of the liver, for example, the size of the liver or a region of interest in the liver can be measured.

[0121] In step A3, the regularity of the liver capsule can be quantified based on the obtained B-mode data of the liver. This step may depend on the results of steps A1 and A2. In other words, step A3 may be performed only if it is determined that the step is necessary in consideration of the results of steps A1 and A2. For example, it may be determined that it is necessary to perform step A3 if a decrease in liver homogeneity is determined in step A1 and / or an increase in liver size is measured in step A2.

[0122] Stages A1 to A3 may be part of B-mode stage A.

[0123] These steps A1 to A3, and steps B to G described below, may be performed at least partially automatically by the ultrasound diagnostic system and / or the system may guide a user when performing the steps, for example to position the ultrasound probe at a particular location on the subject.

[0124] In step B1, vascular flow is assessed and measured based on the color Doppler data and / or pulsed wave Doppler data. This step may depend on the results of steps A1, A2, and A3. In one example, the assessment and measurement process of step B1 may be fine-tuned based on these results of steps A1, A2, and A3, for example by adjusting the liver region of interest, selecting a particular ultrasound probe, selecting settings, or selecting a measurement type (e.g., average velocity vs. resistivity index). For example, the region of interest for obtaining color Doppler data and / or pulsed wave Doppler data may be adjusted based on the measurements made in step A2.

[0125] If the results of steps A and B indicate the absence of diffuse liver disease, the dynamic testing process 100' may be stopped. In other words, further steps C-F may be omitted and, optionally, a report may be generated, with reference to step G.

[0126] For example, depending on the results of steps A and B (e.g. B-mode data and color Doppler data and / or pulsed wave Doppler data), it is determined whether an attenuation coefficient measurement is additionally required to improve the accuracy of the ultrasound examination and / or the accuracy of the examination results. According to a further embodiment, the attenuation coefficient can be measured for one or more regions of interest of the liver identified in steps A and B.

[0127] In step D, a sound speed coefficient can be estimated based on the result of step C. For example, depending on the result of step D, it can be determined whether an additional measurement of the sound speed coefficient is required to improve the accuracy of the ultrasound examination and / or the examination results. According to a further embodiment, the estimated attenuation coefficient can be taken into account in the sound speed estimation process. Furthermore, the sound speed coefficient can be estimated for the same region of interest as the attenuation coefficient.

[0128] At stage E2, the obtained liver ShearWave TM Based on the elastic dispersion data, a viscosity quantification of the liver can be performed. This step may depend on the results of steps A to D as well as step E1. For example, a viscosity model can be used that takes into account the quantified elasticity of the liver.

[0129] Stages E1 and E2 are ShearWave TM It may be part of the elastic stage E.

[0130] In stage F, stages A1, A2 and E1 can be applied to the subject's spleen. This stage can depend on the results of stages A-E. For example, spleen evaluation can detect portal hypertension, which is associated with a high stage of fibrosis. Spleen evaluation can include at least one of the following: evaluating portal hypertension, defining the need for further examination of organs other than the liver and spleen, and correcting liver elasticity depending on the presence and severity of splenic sclerosis. By applying stages A1, A2 and E1 to the spleen, the same tools and characteristics can be used for spleen evaluation as for liver evaluation.

[0131] In step G, a report is generated based on the results of steps A to F. For this purpose, a multiparametric analysis process can be performed, optionally corrected for confounding factors related to the data obtained in the different steps. The report can include indicators of staging, for example steatosis, inflammation, liver congestion. Furthermore, the report can include indicators of liver stratification level. For example, a classification in categories can be determined automatically, for example a classification regarding liver decompensation risk. Furthermore, the report can include indicators related to health risks, for example esophageal varices, and optionally recommendations regarding patient management, for example performing an echoendoscopy.

[0132] FIG. 4A shows a schematic flow chart of determining a dynamic inspection process 100 using an AI model according to an embodiment of the present disclosure. A predefined algorithm can be developed as an artificial intelligence (AI) model trained with data from hundreds or thousands of ultrasound (U / S) examinations, including information on known pathologies from reports, images, sequence of steps, and other relevant factors. This training allows the model to learn patterns and correlations in the data and use that knowledge to make accurate predictions and recommendations for future examinations. The AI ​​model can use the pathological hypothesis to determine a customized dynamic inspection process 100 for an ultrasound examination. The dynamic inspection process 100 can include one or more components. The dynamic inspection process 100 defines one entity 204 configured to evaluate a pathological hypothesis or a particular member of a pathology family as a function of the obtained medical data.

[0133] The real-time data acquired by the entities of the dynamic inspection process 100 (i.e., acquired medical data of the subject) can be used in a feedback loop. To that end, the real-time data can be used as an input for an AI model. Thus, the AI ​​model can adapt the dynamic inspection process based on the real-time data.

[0134] FIG. 4B is a schematic diagram of a clinical evaluation process according to an embodiment of the present disclosure. The clinical evaluation process, which may include evaluating a subject's medical history, symptoms, and physical examination, may be used to identify a suspected pathology family. The clinical evaluation process may be based on various types of a priori information, such as symptoms the subject is experiencing, the situation in which the examination is being performed, the subject's medical history, risk factors, the subject's overall health, lifestyle factors, and results of blood or other diagnostic tests. This information may be used to determine a dynamic examination process 100 for a medical examination using an ultrasound examination system and an AI model trained on data collected from hundreds or thousands of past ultrasound examinations.

[0135] The dynamic inspection process 100 may include a sequence of related stages that are automatically defined based on the family of suspected pathology, which may be achieved using intelligent predefined algorithms. The dynamic inspection process 100 may also include the selection of the type of data to acquire, such as an image, a cine loop, or a selected 3D (i.e., three-dimensional) volume of the scanned medium of the subject (e.g., in the case of 3D imaging), as well as the selection of the appropriate probe(s) to use. The volume may be the sum of 2D (i.e., two-dimensional) slices as used in conventional 2D ultrasound imaging. Other aspects that may be determined by the dynamic inspection process include the mode to be used, the measurement tools employed, and the intelligent components to be utilized for various purposes, such as diagnosis, prognosis, and automated measurements. Incorporating all these elements into the dynamic inspection process 100 may standardize ultrasound inspection and improve its accuracy, efficiency, and productivity.

[0136] 5A is a schematic diagram of an ultrasound system 10 according to an embodiment of the present disclosure. The ultrasound system 10 may be an example of a system for determining a dynamic inspection process 100 for an inspection.

[0137] The ultrasound imaging system 10 includes: - probe 20, a processing unit 30 for processing an image based on the signals received by the probe, a control panel 40a connected to the processing unit, the control panel 40a including at least a button 41 and a touchpad 42; and a display 50 for visualizing the images.

[0138] The probe 20 is connected to a processing unit 30, for example via a cable 21 or via a wireless connection, and is capable of emitting ultrasonic waves W into the medium M and / or receiving ultrasonic waves W from the medium M, said received ultrasonic waves resulting from or resulting from the reflection of the emitted ultrasonic waves on scattering particles present in the medium M. The probes emitting and receiving waves may be the same or different.

[0139] The display screen 50 may be a screen for visualizing an image processed by the processing unit 30. The display 50 may also visualize other information such as the scale used in the image, value maps, configuration information for processing, or any information such as help information for the touchpad 42 or contextual gesture help.

[0140] The display screen is articulatable on the support arm 51 for better positioning for the user. The display screen is typically a large size (at least 20 inches) high definition screen for better image visualization for the user.

[0141] The control panel 40a is, for example, a portion of the system casing 31 that includes a panel casing having a substantially flat surface angled toward a user for one-handed operation by the user. The control panel 40a can include a control panel display screen 49 for visualizing some configuration information or any information dedicated to the user.

[0142] The processing unit 30 is configured to determine the dynamic examination process 100 based on the pathological hypothesis using a predefined algorithm. That is, an operator or user of the system 10 (which may be, for example, a medical doctor or merely technically skilled) first derives a pathological hypothesis. For example, if the operator or user is a medical doctor, said pathological hypothesis may be derived or determined based on empirical knowledge, based on observations of the examined object. It is also possible (particularly in the exemplary case where the operator or user is merely technically skilled) to automatically or at least semi-automatically derive or determine the pathological hypothesis using a computational tool. The computational tool may be or include a software-based solution or (AI) algorithm. The computational tool may automatically process and / or analyze information available in the object record. The computational tool may be at least partially external to the system 10, i.e. may be at least partially executed on an external computing device. The operator then provides the pathological hypothesis as an input to the ultrasound imaging system 10 using the control panel 40a and / or the display screen 50. The processing unit 30 can then use a predefined algorithm to automatically determine a dynamic inspection process 100 using the pathological hypothesis provided by the operator. The dynamic inspection process 100 can be a complete sequence of relevant steps or stages for a family of suspected pathologies. The operator can also use the control panel 40a and display screen 50 to monitor and adjust the dynamic inspection process 100 as needed during the inspection.

[0143] 5B is a schematic diagram of an exemplary dynamic inspection process 100 for the ultrasound system 10 of FIG. 5A. The dynamic inspection process 100 may include a set of entities 200 that perform specific tasks, including a probe 20 that generates ultrasound signals and receives the resulting echoes, a processing unit 30 that may convert the signals into images, a control panel 40a that allows an operator to manually readjust system settings and image parameters (which may have been automatically set or adjusted by a system according to the present disclosure), and a display 50 that presents the resulting images. The entities associated with the control panel 40a are connected to entities associated with buttons 41, touch pad 42, and control panel display screen 49.

[0144] The dynamic examination process 100 can specify how these entities should be interconnected and how they should communicate with each other via the digital communication channel 300. An operator can follow the dynamic examination process 100 using the buttons 41 and touch pad 42 on the operation panel to adjust system settings and select the appropriate examination mode(s), resulting in an accurate and efficient medical examination.

[0145] The entities can be adapted during the execution of the dynamic inspection process 100 based on the obtained medical data. For example, if a doctor uses a first type of probe 20, the processing unit 30 of the dynamic inspection process 100 can analyze the obtained medical data and determine whether the probe 20 is suitable for this inspection. In response, the processing unit 30 of the dynamic inspection process 100 can provide information for adapting the probe 20 by using a second type of probe 20. The probe 20 can be adapted accordingly. Furthermore, machine parameters such as gain, depth, frequency, etc. may be adapted automatically by the processing unit 30.

[0146] Another embodiment of the dynamic examination process is the evaluation of age-related musculoskeletal (MSK) disorders. The dynamic examination process can include automatic stages of B-mode and SWE mode. In B-mode, an intelligent guidance component for MSK can be used to recognize typical patterns and acquire the appropriate B-mode plane. The echo intensity tool is used for MSK to calculate and position the region of interest (ROI) for musculoskeletal disorders evaluation. SWE (ShearWave TM In the ShearWave Elastography mode, the Guidance component can be used to determine the alignment of the MSK fibers. The Guidance component can further be used to obtain optimal SWE data, for example, taking into account the determined alignment of the MSK fibers. The Intelligent Box / Q-box component is a TM Can be used to evaluate speed.

[0147] The dynamic screening process may include an "age-related MSK disorders" component that performs a multi-parametric analysis including myosteatosis as a biomarker for sarcopenia. This component can diagnose sarcopenia and provide a prognosis. The report generated by the dynamic screening process may include specific references to pathology and recommendations for subject management. The dynamic screening process allows for automated assessment of age-related MSK disorders and provides an accurate diagnosis and prognosis.

[0148] Another embodiment of the dynamic examination process is the assessment of NACT (neo-adjuvant chemotherapy) responsiveness of breast cancer, which includes several stages. The first stage can include query / retrieve to obtain ultrasound data from the subject's records and / or previous examinations. The second stage uses B-mode with automatic annotation and body markers, and measurement of lesion size. The third stage uses Box / Q-box for automatic breast hardness assessment. TM Component SWE mode and the so-called TriVu, which combines B-mode and SWE mode with Ultrasensitive Doppler TMIn the fourth stage, the 3D PROBE can be used in both B-mode and SWE modes. In the fifth stage, a component dedicated to "NACT response evaluation" can be introduced, including multi-parametric analysis to evaluate chemotherapy response. The final stage is to generate a report based on the analysis.

[0149] Throughout this specification, including the claims, the term "comprising" should be understood to be synonymous with "comprising at least one" unless otherwise specified. In addition, ranges defined in this specification, including the claims, should be understood to be inclusive of their final values ​​unless otherwise specified. The specific values ​​of the elements described should be understood to be within acceptable manufacturing or industrial tolerances known to those skilled in the art, and the use of the terms "substantially" and / or "approximately" and / or "generally" should be understood to mean falling within such tolerances.

[0150] The terms "record" and "receive" may be used interchangeably throughout this disclosure unless otherwise indicated.

[0151] Although the present disclosure has been described herein with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present disclosure.

[0152] It is intended that the specification and examples be considered as exemplary only, with the true scope of the disclosure being indicated by the following claims.

[0153] Reference herein to a patent document or other matter identified as prior art is not to be construed as an admission that the document or other matter was publicly known at the priority date of any of the claims or that the information contained in the document or other matter was part of the common general knowledge.

[0154] The following provides examples of aspects of the present disclosure. (Aspect 1) A system (10) for determining a dynamic inspection process, comprising: a processing unit (30) configured to determine a dynamic examination process (100) based on a pathological hypothesis using a predefined algorithm, The dynamic inspection process (100) is associated with acquiring medical data; the dynamic inspection process (100) being adaptable on the fly based on the acquired medical data; system. (Aspect 2) The dynamic inspection process (100) defines at least one entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) that can perform or control a task; The system of embodiment 1. (Aspect 3) said at least one entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) being adaptable during the execution of said dynamic inspection process (100) based on said acquired medical data; The system according to embodiment 2. (Aspect 4) The tasks and / or entities (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) Biomarkers Calculation tools, Analyzing and / or processing the components; Measurement tasks and / or measurements, Acquisition mode, Probe type, Machine parameters, scan parameters, the acquisition process for obtaining the medical data; Processing tools, Reporting tools, and Selecting medical data associated with at least one of The system according to any one of aspects 2 and 3. (Aspect 5) The dynamic inspection process (100) defines at least one predefined acquisition mode (400) of the medical device, the acquisition mode (400) being associated with a group of entities (401, 402); 5. The system of any one of aspects 2 to 4. (Aspect 6) Adapting the dynamic inspection process (100) includes: re-executing the first entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) or the first group of entities (401, 402) using at least one adapted machine parameter based on the medical data obtained by the first entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) or the first group of entities (401, 402); and / or selecting and / or adapting a second entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) or a second group of entities (401, 402) based on medical data acquired by the first entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) or a first group of entities (401, 402); 6. The system of any one of aspects 2 to 5. (Aspect 7) Adapting the second entity includes: removing the second entity from the dynamic inspection process (100); adding the second entity to the dynamic inspection process (100); and Adapting at least one parameter of the second entity. At least one of 7. The system of any one of aspects 2 to 6. (Aspect 8) The at least one entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) is executable using a medical device; 8. The system of any one of aspects 2 to 7. (Aspect 9) The dynamic inspection process (100) defines a group (401, 402) of interconnected entities, where at least two entities (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) of the group are dependent on each other; The system of any one of aspects 2 to 8. (Aspect 10) at least two entities (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) of said group of interconnected entities are capable of communicating with each other via a digital communication channel (300), optionally established according to said dynamic inspection process (100), The system according to embodiment 9. (Aspect 11) The acquired medical data includes Data obtained from subjects in real time, Medical test data, and Data acquired by at least one medical device, optionally including an ultrasound system (10). At least one of 11. The system of any one of aspects 1 to 10. (Aspect 12) the dynamic inspection process (100) being adaptive in real time; 12. The system of any one of aspects 1 to 11. (Aspect 13) said pathological hypothesis comprises a hypothesis of a pathology family and / or said dynamic inspection process (100) defines at least one entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) configured to evaluate said pathological hypothesis or a particular member of a pathology family as a function of said acquired medical data, 13. The system of any one of aspects 1 to 12. (Aspect 14) the predefined algorithm is or includes an artificial intelligence (AI) model; and / or at least one entity is or includes an artificial intelligence (AI) model; 14. The system of any one of aspects 1 to 13. (Aspect 15) 1. A computer-implemented method for determining a dynamic examination process (100) for a medical examination of a subject, the method comprising: determining said dynamic inspection process (100) based on a pathological hypothesis using a predefined algorithm; The dynamic inspection process (100) is associated with acquiring medical data; the dynamic inspection process (100) is adapted on the fly based on the acquired medical data; Computer-implemented method. (Aspect 16) A computer program comprising instructions that, when the program is executed by a computer, cause the computer to perform the method according to claim 15. Computer program.

Claims

1. A system (10) for determining a dynamic inspection process, comprising: a processing unit (30) configured to determine a dynamic inspection process (100) based on a pathological hypothesis using a predefined algorithm, The dynamic inspection process (100) is associated with the acquisition of medical data, the dynamic inspection process (100) being adaptable on the fly based on the acquired medical data; system.

2. said dynamic inspection process (100) defining at least one entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) capable of performing or controlling a task; The system of claim 1 .

3. said at least one entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) being adaptable during execution of said dynamic inspection process (100) based on said acquired medical data; The system of claim 2.

4. The tasks and / or entities (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) Biomarkers, Calculation tools, Analyzing and / or processing the components; Measurement tasks and / or measurements, Acquisition mode, Probe type, Machine parameters, scan parameters, the acquisition process for obtaining the medical data; Processing tools, Reporting tools, and Selecting medical data associated with at least one of A system according to any one of claims 2 and 3.

5. said dynamic inspection process (100) defining at least one predefined acquisition mode (400) of said medical device, said acquisition mode (400) being associated with a group of entities (401, 402); A system according to any one of claims 2 to 4.

6. Adapting the dynamic inspection process (100) comprises: re-executing said first entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) or first group of entities (401, 402) using at least one adapted machine parameter based on said medical data obtained by said first entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) or first group of entities (401, 402); and / or selecting and / or adapting a second entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8), or a second group of entities (401, 402), based on medical data acquired by the first entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8), or a first group of entities (401, 402); A system according to any one of claims 2 to 5.

7. Adapting the second entity includes: removing the second entity from the dynamic inspection process (100); adding said second entity to said dynamic inspection process (100); and Adapting at least one parameter of the second entity. At least one of: A system according to any one of claims 2 to 6.

8. The at least one entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) is executable using a medical device; A system according to any one of claims 2 to 8.

9. said dynamic inspection process (100) defining a group (401, 402) of interconnected entities, at least two entities (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) of said group being mutually dependent; A system according to any one of claims 2 to 9.

10. at least two entities (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) of said group of interconnected entities are capable of communicating with each other via a digital communication channel (300), optionally established according to said dynamic inspection process (100), The system of claim 9.

11. The acquired medical data includes data obtained in real time from said subject; Medical test data, and Data acquired by at least one medical device, optionally including an ultrasound system (10). At least one of: A system according to any one of claims 1 to 10.

12. the dynamic inspection process (100) being adaptive in real time; A system according to any one of claims 1 to 11.

13. said pathology hypothesis comprises a hypothesis of a pathology family and / or said dynamic inspection process (100) defines at least one entity (200, 201, 202, 203, 204, S1, S2, S3, S4, S5, S6, S7, S8) configured to evaluate said pathology hypothesis or a particular member of a pathology family as a function of said acquired medical data, A system according to any one of claims 1 to 12.

14. the predefined algorithm is or includes an artificial intelligence (AI) model; and / or At least one entity is or includes an artificial intelligence (AI) model; A system according to any one of claims 1 to 13.

15. 1. A computer-implemented method for determining a dynamic testing process (100) for a medical test of a subject, the method comprising: determining a dynamic inspection process (100) based on a pathological hypothesis using a predefined algorithm, The dynamic inspection process (100) is associated with acquiring medical data; the dynamic inspection process (100) is adapted on the fly based on the acquired medical data; Computer-implemented method.

16. A computer program comprising instructions that, when said program is executed by a computer, cause said computer to carry out the method according to claim 15. Computer program.

Citation Information

Patent Citations

  • User interface system and method for creating, organizing and configuring ultrasound imaging protocols

    JP2009509615A

  • Adaptive Ultrasound Scanning

    JP2021520939A

  • Breast Ultrasound Workflow Applications

    JP2021522875A

  • Ultrasound diagnostic apparatus and ultrasound examination assisting method

    JP2022127808A

  • System and method for standardized MRI examinations with patient-centric scan workflow adaptations

    US20220068472A1