Ultrasound system with custom unit

By using custom units in the ultrasound system, estimating skill levels based on real-time user interaction data and dynamically adjusting functions, the problems of low usage efficiency and high error rate caused by user skills diversity are solved, and efficient and accurate in different environments are achieved ultrasound measurement.

CN118974842APending Publication Date: 2024-11-15KONINKLIJKE PHILIPS NV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380030152.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-04-06
Filing Date
2023-03-16
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

When used in intensive care units and non-hospital settings, it is difficult to provide personalized user interfaces and guidance due to the diversity of user skills and goals, resulting in inefficient use and high error rates.

Method used

Through custom units, the user's skill level is estimated using real-time user interaction data, and the functions and user interface of the ultrasonic system are dynamically adjusted, providing user-oriented functions to improve operational efficiency and accuracy.

Benefits of technology

It improves the efficiency of ultrasound system use for users of different skill levels, reduces errors, and adapts to different environments and user needs, especially in non-hospital environments to provide convenient and accurate ultrasound measurement results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118974842A_ABST
    Figure CN118974842A_ABST
Patent Text Reader

Abstract

The present invention relates to an ultrasound system (200) operable by one or more users of the ultrasound system to process ultrasound measurements of a patient. The system includes an ultrasound unit operable to perform the ultrasound measurements of the patient. The system also includes a customization unit. The customization unit obtains user interaction data for one or more interactions of a current user performing a current task at the ultrasound system. The ultrasound system provides user-oriented functionality for performing the current task at a plurality of different skill levels. The customization unit determines a real-time skill estimate of the current user based on the user interaction data. The real-time skill estimate indicates a current capability of the current user to perform the current task. The customization unit customizes the ultrasound system to perform the current task at a skill level corresponding to the real-time skill estimate of the current user.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an ultrasound system, a customization unit for an ultrasound system and a computer-implemented method corresponding to the customization unit. The present invention also relates to a computer-readable medium. Background Art

[0002] Currently, ultrasound is often used as a standalone tool in the intensive care unit (ICU) to perform spot checks in patients and for specific tasks such as line placement. When a scan is desired, the machine is usually brought in by an intensivist, or an order is placed with the cardiology department or other related ultrasound scanning department. Typically, the results stay in the ultrasound ecosystem and are reviewed there. Therefore, most of the ultrasound in the intensive care unit is currently used by specialists.

[0003] However, as invasive tools for monitoring patient hemodynamics fall out of favor, ultrasound is increasingly being used as a monitoring tool in critical care, perioperative settings. As a result of this change in use, ultrasound in the ICU and other hospital units appears to be increasingly performed by users in different clinical roles (e.g., chief intensivist, junior clinician, resident, nurse, etc.), with different skill levels, and with different time constraints. Users in different roles may have different levels of training, and their goals for using ultrasound may also be different. Moreover, the level of training for the use of ultrasound varies significantly with clinical specialty and institutional preferences.

[0004] Another development is that ultrasound is increasingly used in non-hospital environments, such as in the home, in an ambulance or at a sports facility. Also in this context, ultrasound may have a wide range of users with different skills, goals and time constraints. Summary of the invention

[0005] It would be desirable to provide an ultrasound system that can be effectively used by a wide range of users.

[0006] According to a first aspect of the invention, there is provided an ultrasound system, as defined by claim 1. According to another aspect of the invention, there is provided a customization unit for an ultrasound system, as defined by claim 13; and, there is provided a computer-implemented method corresponding to the customization unit, as defined by claim 14. According to an aspect of the invention, there is described a computer-readable medium, as defined by claim 15.

[0007] Various measures discussed herein relate to an ultrasound system. The ultrasound system may be operated by one or more users of the ultrasound system to process ultrasound measurements of a patient. The processing may involve: preparation of the ultrasound scan; image acquisition; and / or analysis of acquired ultrasound measurements. The processing may be performed at various units of the ultrasound system, for example: image acquisition may be performed at the ultrasound unit that performs the ultrasound measurements; analysis may be performed at a patient monitor that allows inspection of the ultrasound image or other types of measurements of the ultrasound unit; and scan preparation may be performed at the ultrasound unit and / or the patient monitor. The ultrasound system may include at least the ultrasound unit.

[0008] In order to improve the usability of the ultrasound system, the inventors envision the use of a customization unit. The ultrasound system can provide user-oriented functionality, such as a user interface, for performing one or more tasks at the ultrasound system. Interestingly, the functionality can be provided at multiple different skill levels, for example, the difference being the amount of user guidance and / or automation provided. By appropriately customizing the ultrasound system to provide user-oriented functionality for the skill level of the user at hand, the ultrasound system can be used more effectively by different users. It should be noted that providing customizability typically involves not only adapting the user interface itself, but also providing other components that implement the customization, for example, a software unit that provides user guidance and / or automation at a corresponding skill level, and / or a trained machine learning model, sensor, storage, etc., for example, for providing the user guidance or automation.

[0009] Interestingly, the customization may be based on a real-time skill estimate of a current user performing a current task at the ultrasound system. The customization unit may be configured to obtain user interaction data of one or more interactions of the current user. For example, the user interaction data may include one or more corresponding interaction items representing corresponding individual interactions of the user with the system, such as button presses, selection of a scanning mode, and / or selection of an image quality estimation method. An interaction item may also represent a cumulative amount corresponding to a set of interactions, such as an amount of time and / or an amount of steps taken to complete a given task.

[0010] The customization unit may determine a real-time skill estimate of the current user based on the user interaction data. The ultrasound system may be customized by providing user-oriented functionality at a skill level corresponding to the determined real-time skill estimate. Interestingly, by basing the skill estimate on the user interaction data, a skill estimate may be obtained, the skill estimate indicating the current ability of the current user to perform the current task, for example indicating a skill level and a current physical and / or mental ability. In particular, the skill estimate may not be derived by simply identifying the user and / or user role and selecting a skill estimate for the identified user and / or role. In particular, in different situations, different skill estimates may be determined for the same user performing the same task based on the interaction data. For example, the same user may perform the same task a first time (wherein a first real-time skill estimate is determined) and a second time (wherein a second different real-time skill estimate is determined). This is despite the fact that there is no change in the user profile; in fact, the determination of the skill estimate may not be based on any identification of the current user. Rather, the skill estimate may be determined from the user interaction data based on how the current user interacts with the system, and thus may be indicative of the user's current ability to perform the task and in this sense may be considered real-time.

[0011] Using real-time skill assessments, rather than, for example, looking up an appropriate skill level given the identity of the user, has several advantages. As the inventors appreciate, the user's skills may be affected by their current mental and physical state, and therefore, may change over time. Specifically, the user's skill level may be affected by the context of the current task being performed, for example, by the urgency or pressure of the situation, and / or by the level of cooperation provided by the patient. In addition, without considering context-related factors, it may be difficult in practice to manually set an appropriate skill level for the user, for example, deriving the skill level from a paper questionnaire or similar questionnaire. Determining the skill level in an automated manner based on the current interaction avoids this difficulty. When manually setting the skill level, there is also a problem that user skills typically change over time, and it is difficult to keep the profile up to date to reflect the changing skills, and it is prone to errors. For example, over time, a user may gradually improve without himself or others even realizing it. Relying (only) on the profile of the identified user is also error-prone in situations involving several people, in which case, in practice, re-identification is usually not done when another user takes over; and it cannot be used to provide customization for new users inside and outside the hospital who do not yet have a profile.

[0012] By providing user-oriented functions at a skill level appropriate to the context at hand, the user may be assisted in interacting with the ultrasound system (such as the ultrasound unit or patient monitor) in a more efficient manner. Given such a context, the user-oriented functions may focus on the most important added value for the user. For example, workflow suggestions may be provided only in areas relevant to the current context, thereby avoiding excessive guidance. Moreover, by providing guidance at a lower skill level, errors made by such users may be fewer; and examination time and the need for re-examinations may be minimized.

[0013] The user-oriented functionality may not be tailored to the subjective preferences of any particular user; rather, by collecting user interaction data and using the user interaction data to estimate skill levels, for example, without identifying the user at hand, and therefore independently of the user's subjective preferences, technical measures may be provided that allow automation and / or guidance to be provided in the most appropriate manner for the current situation. Thus, by providing customization in a data-driven manner based on the situation at hand, rather than based on user preferences, the efficiency of interacting with the entire ultrasound system may be improved, as opposed to only targeting certain specific usage modes and / or targeting certain specific users. For example, the number of required user actions measured in many interactions with different users having different roles and / or in different situations may be reduced.

[0014] The unit of the ultrasound system that provides the customization (e.g., which determines the real-time skill estimate and customizes the ultrasound system based thereon) may be referred to herein as a customization unit. The customization unit may generally take various forms. For example, the customization unit may be included in the ultrasound unit, the patient monitoring unit, the bridge unit, or any other unit of the ultrasound system. The customization unit may be part of the same device that provides the user-oriented functionality for the current task, but this is not required, for example, the customization device may customize the ultrasound system according to the real-time skill estimate by determining the skill level for the current task and providing instructions to the device that provides the user-oriented functionality to provide the functionality at that skill level.

[0015] Optionally, the ultrasound system can be used in critical care and / or perioperative settings, such as an intensive care unit (ICU). In particular, in such facilities, ultrasound is expected to be increasingly used by users of different clinical roles and skill levels, for example, to replace invasive tools to monitor patient hemodynamics. Therefore, in this facility, it is particularly relevant to provide appropriate guidance.

[0016] Optionally, the ultrasound system can be used outside the hospital, such as at home, in an ambulance, at a sports facility, or in a primary clinic. For example, the ultrasound system can be used in situations where fixed diagnostic systems may not be available, such as in underdeveloped counties or regions or rural areas. And, in such use, ultrasound can be used by a wide range of users, particularly by relatively inexperienced users. In addition, outside the hospital, it may be difficult or impractical to identify the user or access stored user profiles, and there may be great variability in environmental factors (other people present, stress of the situation, etc.), making it particularly relevant to use interactive data to determine real-time skill estimates. In such a scenario, field-measured signs or images (e.g., ultrasound images of sports injuries) and / or personal health data (e.g., reference ultrasound images) can be used for analysis.

[0017] Optionally, the customization unit may be configured to determine the real-time skill estimate independently of any user identification. For example, the customization unit may not obtain a user identifier of the user, or in any case, may not use any such user identifier when determining the real-time skill estimate. Thus, the customization may be performed to provide user-oriented functionality in the most effective manner for the situation at hand, regardless of any subjective user preferences that the particular user may have. This has several advantages. It allows customization to be provided without requiring access to user profile information, which makes the system easier to integrate into existing workflows where no identification currently occurs, particularly in non-hospital facilities; it also improves security because the source of profile information does not need to be accessed. Not performing user identification is also beneficial from a privacy perspective: a user may not be comfortable with their skill level being tracked, but can still benefit from customization performed in an anonymous manner. Using data-driven customization also allows the customization to be optimized in a feedback loop, for example, by collecting data about the effect of performing a particular customization based on specific interaction data; for example, determining the real-time skill estimate manually or by training a machine learning model.

[0018] Optionally, the ultrasound system is an ultrasound for monitoring (US4M) system. Generally, the term US4M system may refer to a system in which the ultrasound unit is functionally integrated with a patient monitoring unit, for example, an ultrasound image and / or measurement results are integrated into the patient monitoring ecosystem. As known per se, such a patient monitor (sometimes also referred to as a "medical monitor", "physiological monitor" or "vital sign monitor") may be configured to display current measurements of one or more physiological parameters of the patient, such as blood pressure, pulse oximetry and / or ECG. In place of or in addition to such measurements, the patient monitor of the US4M system may be configured to show information based on the measurements of the ultrasound unit. For example, the patient monitor may be configured to show the ultrasound image itself, or to show the current value of a physiological parameter derived from an ultrasound image, such as an anatomical dimension, such as the volume of a heart chamber or another body part. In such a US4M system, a user may perform ultrasound tasks at each unit of the US4M system, such as scan preparation, acquisition and analysis. In an embodiment, the current task may be performed at the patient monitoring unit. In other embodiments, the current task may be performed at a bridge device for information exchange between the ultrasound unit and the patient monitoring unit. Thus, the user-oriented ultrasound-related functions at these respective units of the ultrasound system may be customized, as described herein.

[0019] Optionally, the customization unit may also be configured to derive the current task being performed by the current user from the user interaction data. This has the advantage that this may also be performed automatically, without, for example, requiring the user to manually select the current task. However, the current task may also be derived in other ways. For example, an alternative way to derive the current task is to use data captured from an expert guidance subsystem of the ultrasound system, such as a system for providing remote expert guidance based on video and / or audio.

[0020] Optionally, the user-oriented functionality may be used to perform the current task at multiple different skill levels and according to multiple different roles. For example, the current task may be performed according to a first role, and the ultrasound system may be configured to provide user-oriented functionality for the current task according to multiple different skill levels of the first role. Also, the current task may be performed according to a second role, and the ultrasound system may be configured to provide user-oriented functionality for the current task according to multiple skill levels of the second role. User interfaces for different combinations of role skill levels may overlap partially, but are generally not identical. Thus, the customization unit may obtain the current role; determine the real-time skill estimate; and customize the ultrasound system according to the role and skill estimate. For example, the customization unit may be configured to customize the ultrasound system for at least one or two of the following roles: intensive care physician, chief intensive care physician, clinician, junior clinician, resident, nurse, general practitioner, midwife, sports physician, home caregiver, and the patient himself.

[0021] Typically, the current task may be part of a current stage of multiple stages of processing of the ultrasound measurement. For example, the current stage may be: preparation; image acquisition; or analysis. Typically, analysis may involve interpretation of the ultrasound image or other measurement results; management (e.g., selecting images or other measurement data for storage or further processing); and data transmission.

[0022] Optionally, the real-time skill estimate can be determined for the current stage based solely on the interaction of the current user at the current stage. Thus, the real-time skill estimate can be specific to the current stage and can therefore be better suited to the context of the current task. For practical purposes, in a given stage, it can be assumed that one or more users interacting with the ultrasound system remain unchanged. Therefore, using only data from the current stage can reduce the contamination of the user interaction data with data of other users and / or with less relevant data. This can be beneficial in particular when machine learning is used to determine the skill estimate. On the other hand, even when only the interaction data of the current stage is used, it is still possible to use the real-time skill estimate and / or role derived in an earlier stage as a clue, for example, as an initial guess for the current real-time skill estimate and / or role, which may be confirmed or refuted as the interaction data of the current stage is collected.

[0023] Typically, the user interaction data used to determine the real-time skill estimate may include multiple interaction items. The interaction items may include one or more interaction items representing the user's corresponding individual interactions with the system, for example, individual inputs provided by the user and / or individual actions performed by the user as part of the current task. For example, the set of interaction items may include interaction items representing one or more of the following: button presses, selection of a scan mode, selection of an image quality estimation method, access to previous images, review of an imaging sequence, access to previous image data, measurements performed on an ultrasound image, quantification performed on an ultrasound image, an image quality estimation method applied to an ultrasound image; access to patient information associated with the imaged patient; and, inspection of one or more signals at a patient monitor and / or viewing settings used. Alternatively or additionally, the interaction items may include an accumulation amount (e.g., calculated by the customization unit itself, or received in a calculated form) determined in multiple interactions of the user with the system. For example, the set of interaction items may include interaction items representing one or more of the following: time to obtain a first image, number of images saved, time to achieve a target view, and number and / or type of manual edits performed on the automatic quantification results.

[0024] In general, the interactive items may be obtained in various ways, for example, interactions with a user interface of the ultrasound system (such as with a keyboard, mouse, touch screen, etc.) may be recorded by a device providing the user interface or captured by a camera recording the user. In addition to the ultrasound measurements themselves, at least one of the interactive items may be based on sensor measurements of the ultrasound system. For example, one or more interactive items may be used that are based on sensors of the ultrasound probe and that indicate how to use the probe, for example, by indicating an area in the body that has been scanned; by indicating the pressure of insertion and / or by representing a movement pattern of the ultrasound probe.

[0025] Optionally, in addition to the user interaction data, the ultrasound unit may use context information about the current task when determining the real-time skill estimate. The context information may include information about the patient, such as an estimate of the difficulty of imagining the patient in his or her current condition based on patient information other than the ultrasound image itself. Alternatively or additionally, the context information may include information about the physical environment in which the ultrasound system is used, such as the number of people present in the room in which the ultrasound system is used, measurements of air quality and / or noise, etc. Such aspects may affect the user's physical and psychological state, which may interact with the real-time skill level.

[0026] Typically, the customization unit can customize the ultrasound system in various ways to operate at a given skill level. One way is to provide or not provide user guidance depending on the skill level, such as whether to show an acquisition guidance module (which itself can be known) for acquiring a specific type of image (such as an A4C image). Providing such guidance to users with lower skills can allow them to work faster and make fewer mistakes, while such guidance may only distract users with higher skill levels. Another way of customization is to automate or non-automate a certain operation, such as automatically preloading a set of predefined scan settings. And here, automation can prevent inexperienced users from making mistakes and allow them to work faster, while allowing experienced users to achieve a higher degree of quality by being able to customize the operation without distracting them.

[0027] As a specific example, the customization unit may be configured to customize the ultrasound system by automatically retrieving and showing or not automatically retrieving and showing information about previous scans for the current follow-up scan according to the skill level. As another example, the customization unit may be configured to customize the ultrasound system by automatically preloading or not automatically preloading scan settings (e.g., line placement) suitable for the current task according to the skill level. In yet another example, the current task may involve obtaining a given view of an organ. In this case, the customization unit may be configured to customize the ultrasound system by showing or not showing example images of the given view of the organ according to the skill level.

[0028] Optionally, the customization unit may allow the user to learn and / or modify the skill level at which the ultrasound system provides the user with the current task. In other words, the user may have a manual override that they can see and correct the performed customization as desired. Providing such a manual override makes it possible to use the automated customization even if it does not always derive a correct real-time skill estimate. The activation of the manual override may be recorded, for example, in conjunction with the user interaction data used, so that the determined real-time skill estimate may be reviewed, for example manually or automatically by retraining.

[0029] It will be appreciated by those skilled in the art that two or more of the above-mentioned embodiments, implementations and / or alternative aspects of the invention may be combined in any way deemed useful.

[0030] Modifications and variations of any computer-implemented method and / or any computer-readable medium corresponding to the described modifications and variations of the corresponding system, device or unit may be implemented by those skilled in the art based on this description. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] These and other aspects of the invention will become apparent and will be further elucidated with reference to the embodiments described by way of example in the following description and with reference to the accompanying drawings, in which:

[0032] Figure 1 A custom unit for an ultrasound system is shown;

[0033] Figure 2 An ultrasound system is shown;

[0034] Figure 3 shows a detailed example of how to customize an ultrasound system;

[0035] Figure 4 A computer-implemented method of customizing an ultrasound system is shown;

[0036] Figure 5 A computer readable medium including data is shown.

[0037] It should be noted that these figures are diagrammatic only and are not drawn to scale.In the figures, elements corresponding to elements already described may have the same reference numerals. DETAILED DESCRIPTION

[0038] Figure 1 A customized unit 100 is shown. The customized unit 100 may be used in an ultrasound system as described herein, for example, Figure 2 The ultrasound system 200 may be operated by one or more users of the ultrasound system to process ultrasound measurements of a patient, and the system may include an ultrasound unit that may be operated by a user to perform ultrasound measurements of a patient. Figure 2 As discussed, the customized unit 100 may be a separate unit of an ultrasound system, e.g., a separate device; however, the customized unit 100 may be combined with an existing unit of an ultrasound system. For example, the customized unit 100 may be an ultrasound unit, a patient monitoring unit, or a bridge unit of an ultrasound system. The customized unit may also be referred to as a customized system, or in the context of an ultrasound system, a customized subsystem.

[0039] The customization unit 100 may include a data interface 120 for accessing interaction data 010 of one or more interactions of a current user performing a current task, for example, at the ultrasound unit or other unit integrating the customization unit 100, or at a different unit of the ultrasound system. User-oriented functionality for performing the current task may be provided at a plurality of different skill levels. Alternatively or additionally, the data interface 120 may be used to access various other data, in particular model data of a machine learning model trained to determine real-time skill estimates, also described elsewhere.

[0040] For example, Figure 1As shown, the input interface can be composed of a data storage interface 120, which can access data 010 from a data storage 021. For example, the data storage interface 120 can be a memory interface or a persistent storage interface, such as a hard disk or SSD interface, and can be a personal, local area network or wide area network interface, such as a Bluetooth, Zigbee or Wi-Fi interface, or an Ethernet or fiber optic interface. The data storage 021 can be an internal data storage of the system 100, such as a hard disk or SSD, and can be an external data storage, such as a network accessible data storage. For example, at least part of the interaction data 010 can be accessed from a log file of an ultrasound unit of the ultrasound system, for example, via a network accessible data storage storing the log file. In some embodiments, the corresponding data (e.g., various parts of the interaction data) can be accessed from different data storages, for example, via different subsystems of the data storage interface 120. Each subsystem can be of the type described above for the data storage interface 120.

[0041] The customization unit 100 may also include a processor subsystem 140, which may be configured to determine a real-time skill estimate of a current user from the user interaction data 010 during operation of the system 100. The real-time skill estimate may indicate the current ability of the current user to perform a current task.

[0042] The processor subsystem 140 can also be configured to customize the ultrasound system for performing the current task at a skill level corresponding to the real-time skill estimate of the current user. In particular, the customization unit 100 may include an output interface (not shown) for outputting instructions for providing a user interface with the skill level to the user interface component for the current task. The output interface can provide instructions via internal data communication or via external data communication. For example, the output interface can be implemented as a communication interface, as discussed further below. As another possibility, the output interface can be composed of a data interface 120.

[0043] As shown, the customized unit 100 may optionally include a sensor interface 160 for accessing sensor data 124 collected by one or more sensors 075. In particular, the customized unit may be incorporated into the ultrasound unit 100, in which case the sensor interface 160 may be used at least to access the ultrasound measurements 124. Conversely or additionally, the sensor interface may be used to access sensor measurements other than ultrasound measurements, for example, from one or more sensors located on an ultrasound probe, and provide sensor data indicating how to use the probe, such as accelerometers, gyroscopes, pressure sensors, optical sensors, and / or temperature sensors. As another example, the customized unit may be incorporated into a patient monitoring unit, in which case the sensor interface 160 may be used to access sensor data from, for example, a sensor measuring one or more vital signs. The sensor(s) 072 may, but need not, be part of the unit 100. The sensor interface 160 may have any suitable form corresponding in type to the type of sensor, including but not limited to a low-level communication interface, such as based on I2C or SPI data communication, or a data storage interface for the data interface 120 as described above.

[0044] The customization unit 100 may also include a communication interface (not shown) configured to communicate with other (one or more) units of the ultrasound system. The communication interface may be arranged for direct communication with the other (one or more) units, for example, using a USB, IEEE 1394 or similar interface. The communication interface may also communicate via a computer network, for example, a wireless personal area network, the Internet, an intranet, a LAN, a WLAN, etc. For example, the communication interface may include a connector such as a wireless connector, an Ethernet connector, a Wi-Fi, a 4G or 4G antenna, a ZigBee chip, etc. The communication interface may also be an internal communication interface, for example, a bus, an API, a storage interface, etc.

[0045] The customization unit 100 may also include an I / O interface 180 for communicating 182 with one or more input and / or output devices for enabling communication with a user. Examples of input devices may include, but are not limited to, a microphone, a keyboard, a pointing device such as a mouse or a touch screen, a game controller, a Bluetooth controller, a VR controller, and a gesture-based input device, etc. Examples of output devices may include, but are not limited to, a monitor or display, a speaker, etc. The processor subsystem 140 may be configured to provide user-oriented functions, such as a user interface, to the user via the I / O interface 180. This is illustrated in the accompanying drawings by showing a display 190 on which a user interface element 192 that may be customized based on the determined real-time skill estimate is shown. Typically, the input and / or output devices may be incorporated into the customization unit 100 or may be connected to the customization unit 100.

[0046] Generally, each system described in this specification, including but not limited to Figure 1 The system 100 may be implemented as a single device or apparatus, such as a workstation or server. The device may be an embedded device. The device or apparatus may include one or more microprocessors that execute appropriate software. For example, the processor subsystem of each system may be implemented by a single central processing unit (CPU), and may be implemented by a combination or system of such a CPU and / or other types of processing units. The software may have been downloaded and / or stored in a corresponding memory, for example, a volatile memory such as RAM or a non-volatile memory such as Flash. Alternatively, the processor subsystem of each system may be implemented in a device or apparatus in the form of programmable logic, for example, as a field programmable gate array (FPGA). Typically, each functional unit of each system may be implemented in the form of a circuit. Each system may also be implemented in a distributed manner, for example, involving different devices or apparatuses, such as distributed local or cloud-based servers.

[0047] Figure 2 An example ultrasound system 200 is shown. The ultrasound system can be operated by one or more users of the ultrasound system to process ultrasound measurements of a patient. The ultrasound system 200 may generally include various units that may be used at various stages of processing of ultrasound images, for example, in preparation, image acquisition, and / or analysis. The figure shows several such units that may be used alone or in combination. In general, each of the units described may be described with respect to Figure 1 Implemented as described.

[0048] Specifically, the ultrasound system 200 may include an ultrasound unit 210. Such an ultrasound unit may also be referred to as an ultrasound device or a medical ultrasound device. The ultrasound unit 210 may be operated by a user of the ultrasound system to perform ultrasound measurements on a patient. The ultrasound unit 210 may be used in the preparation and / or image acquisition phases for ultrasound use. Various types of ultrasound units are in use today and may be adjusted as described herein. Shown in the accompanying drawings is a portable ultrasound device, in this case with wheels, as on the market today. Examples of such devices include Philips EPIQ series devices, such as EPIQ Elite and EPIQ CVx*, or Philips Affiniti series. Another series of devices that may be used are wired or wireless portable ultrasound devices, such as Philips CX50 series or Philips Lumify. Typically, the ultrasound unit 210 may include: a probe 212, which is used to obtain ultrasound measurements, for example, using one or more CMUTs and / or piezoelectric transducers; and an input and / or output device, which is used to provide a user interface to the user to prepare for scanning and check ultrasound measurements, such as a display 211. When in image acquisition mode, the ultrasound unit 210 can be configured to continuously capture ultrasound images and / or derive measurement results therefrom, for example deriving new measurement results at a period of at most or at least 1 second, at most or at least 10 seconds, at most or at least 1 minute, or at most or at least 10 minutes, and display them on the screen 211 and / or store them.

[0049] As shown, the ultrasound system 200 may also include a patient monitoring unit 220, in other words, a patient monitor. The patient monitor 220 may be configured to show information based on the ultrasound measurements performed on a display 221, for example, an ultrasound image captured by the ultrasound unit 210, or measurements derived therefrom, such as measured anatomical dimensions, for example, the volume of a measured body part, etc. The patient monitor 220 may be used in the analysis phase of ultrasound use. As also shown, the patient monitor is typically configured to also show patient information and / or measurements other than the patient information and / or measurements provided by the ultrasound unit 210, for example, measurements of vital signs such as heart rate and / or body temperature. A system 200 comprising both an ultrasound unit 210 and a patient monitor 220, on which data from the ultrasound unit 210 is shown, may generally be referred to as an ultrasound US4UM system for monitoring.

[0050] The system 200 can be used, for example, in critical care and / or perioperative settings, such as, for example, in an intensive care unit. However, the system 200 can also be used in other clinical settings, which may also involve multiple users who may have potentially different goals to achieve with ultrasound. Another envisioned use is to use the US4M system in an ultrasound clinic. In this example, for example, ultrasound physicians can be assigned to specific tasks and / or examinations based on their experience level. Additionally, in such a situation, it becomes relevant and useful to automatically detect the experience level and / or task and customize the ultrasound system accordingly.

[0051] Another unit that may optionally be part of the ultrasound system 200 is a bridge device 230. The bridge device 230 may facilitate information exchange between the ultrasound unit 210 and the patient monitor 220. For example, the bridge device may provide a shared storage 241 and may optionally provide functions such as image processing, data retrieval, data storage, and / or sensor control. The ultrasound unit 210 may optionally be configured to store data such as ultrasound measurements and / or log files on the storage 241, and the patient monitor 220 may retrieve data therefrom. The bridge device 230 may optionally be configured to store information collected from the patient monitor 220 and passed to the ultrasound unit 210 on the storage 241, such as, but not limited to, patient demographics and vital signs. The bridge device 230 may optionally provide a user interface for facilitating information exchange, such as for accessing previously stored data and / or viewing previously stored instructions or imaging commands. The bridge device 230 may be used in the image acquisition and / or analysis phase of ultrasound use. For example, during image acquisition, the bridge device can be used for scan guidance, such as calling previous exams and / or views from storage 241. During analysis, the bridge device 230 can be used, for example, to push images and / or loops to the bridge device 230 and analyze them on the device.

[0052] Another example of a unit that may optionally be part of the ultrasound system 200 is an inspection device 240 that may allow the ultrasound measurements of the ultrasound unit 210 to be inspected by a user. As shown, the inspection device 240 may be a mobile device, such as a smartphone, tablet, smart glasses, virtual reality headset, hologram, etc. In particular, such inspection devices are useful for use at the point of care outside of a hospital (e.g., at home, in an ambulance, or at a sports facility). These are typical examples where the amount of experience may vary greatly between users, making customization beneficial.

[0053] The figure also shows a computer network 299, through which different devices of the ultrasound system 200 can be configured to interact. For example, the computer network can be the Internet or an internal communication network of a hospital.

[0054] The ultrasound system 200 can provide user-oriented functionality for performing a current task at one of the units of the ultrasound system, for example, at the ultrasound unit 210; at the patient monitor 220; at the bridge device 230; or at the examination device 240. For example, the task is related to performing a specific measurement or imaging a specific body part and / or in a specific view. Various examples of such tasks are provided throughout this specification. The user-oriented functionality can be provided at multiple different skill levels, for example, at multiple different skill levels for each of a set of multiple different roles. For example, the system can provide at least two or at least three different skill levels or at least two or three levels of functionality for a given role.

[0055] Interestingly, the ultrasound system 200 may include a custom unit. The custom unit may be as described with respect to Figure 1 As discussed. The customization unit can be used to enable the ultrasound system 200 to provide user-oriented functionality for a current task at the correct skill level for the situation at hand. To this end, the customization unit can be configured to obtain user interaction data of one or more interactions of a current user performing the current task at the ultrasound system 200. The customization unit can also be configured to determine a real-time skill estimate of the current user based on the user interaction data. The real-time skill estimate can indicate the current ability of the current user to perform the current task. The customization unit can also be configured to derive the current task being performed by the current user based on the user interaction data, but the current task can also be derived in other ways. Thereafter, the customization unit can be configured to customize the ultrasound system to perform the current task at a skill level corresponding to the real-time skill estimate of the current user.

[0056] Typically, the customization unit may be integrated in one of the previously discussed units, for example, may be combined with: ultrasound unit 210; patient monitor 220; bridge device 230; or examination device 240. However, the customization unit may also be implemented as a separate device. The customization unit may include a user interface configured to enable a user to learn and / or modify the skill level of a customized ultrasound system, in other words, to provide a manual override. Preferably, the user interface is provided at a device that also provides user-oriented functionality.

[0057] Depending on how the customization unit is implemented, the customization unit may obtain the user interaction data through internal or external data communication. For example, the user interaction data may include or be based on, for example, a device log of the ultrasound unit 210 stored at the shared storage 241. Another possibility is to implement the customization unit at the same device that provides the user-facing functionality, in which case the user interaction data may be obtained by internally accessing the log file, by receiving a callback from the user-facing functionality, by internally inspecting the user interface using an accessibility API, etc.

[0058] The customization unit may also customize the ultrasound system in various ways, e.g., using internal or external data communications, depending on how the customization unit is implemented. For example, the ultrasound unit 210 or another unit may be configured to operate at a given skill level via an externally accessible API such as a REST API, or may periodically check for settings for a given skill level at the shared storage 241, etc.

[0059] The customization unit may be configured to determine the real-time skill estimate independently of any user identification. Although user identification may be performed in the ultrasound system 200, for example for authorization purposes and / or to derive the role of the current user, it is interesting that the real-time skill estimate may be determined independently of it. As also discussed elsewhere, this may have various advantages, including improved privacy and better adaptation to the current context.

[0060] One or more of the units of the system 200 (e.g., the ultrasound unit and / or the patient monitor 220) may include an expert guidance subsystem, such as a system for providing remote expert guidance based on video and / or audio, also referred to as a telehealth-based support interface. Data from such a subsystem may, for example, be used to derive the current task, and interactions with the subsystem may be captured as user interaction data, e.g., the use of expert guidance may imply a lower real-time skill estimate.

[0061] Figure 3 Detailed but non-limiting examples of how to customize an ultrasound system are shown. The ultrasound system may be operated by one or more users of the ultrasound system to process ultrasound measurements of a patient. For example, the techniques described with respect to this figure may be used by Figure 1 and / or Figure 2 The custom unit described is implemented.

[0062] Typically, customization of the ultrasound system may be performed by estimating a user's profile utilizing user interactions with units of the ultrasound system; specifically, their real-time skill estimate (which may indicate experience level and / or current physical and / or mental ability to perform the task at hand, with the intent of using this information to provide a real-time skill estimate) to customize the ultrasound system for the current user, for example, by providing relevant assistance to the user and / or automatically performing actions to streamline workflow on one or more of the system components.

[0063] Such customization may involve one or more of the following steps, which are discussed in more detail below:

[0064] 1) Identify the current stage of processing of ultrasound measurement, for example, ultrasound scan preparation, image acquisition, analysis of ultrasound measurement results, etc.;

[0065] 2) using user interactions with the system, preferably based on the current stage of the workflow, to estimate the user's profile, specifically their real-time skills;

[0066] 3) optionally, refining the user profile estimate by incorporating historical information for the user;

[0067] 4) Estimate the current task at hand; and

[0068] 5) Given a user profile and a task to be performed, customize the ultrasound system by, for example, providing relevant assistance to the user and / or automating one or more aspects of the workflow.

[0069] Below, an example of how the above steps can be implemented is given:

[0070] 1) Identify the stages of the workflow

[0071] Typically, the processing workflow for ultrasound measurements can be divided into various stages. The tasks of a given stage can typically be performed by a single user and / or at a single unit of ultrasound equipment. Therefore, as described herein, a real-time skill estimate for a current stage can be determined based solely on the current user's interactions at the current stage, possibly using real-time skills and / or roles derived at earlier stages as additional input.

[0072] Typically, clinical workflows involving ultrasound scanning can be performed from the user Prepare The US system starts, for example, by changing modes, establishing presets, etc. Once the desired settings have been set, the workflow can proceed to Image acquisition Once the desired image has been acquired, various types of analyze (e.g., quantification or other analytical processing).

[0073] In general, the phases of the workflow can be identified based on the use of corresponding functions at the ultrasound system, as indicated, for example, by interaction data at the various units of the ultrasound unit. For example, the transition from the image acquisition to the analysis phase can be initiated by transferring ultrasound images and / or measurements from the ultrasound unit to the patient monitoring unit via a study / image output or other button function on the ultrasound unit. Based on these button presses and other system transitions in the system, the current phase of the clinical workflow can be determined.

[0074] 2) Evaluation of user interaction with the system

[0075] Typically, use of an ultrasound system, such as a US4M (Ultrasound for Monitoring) system, may include user interactions during various stages, many of which may be used to assess the profile and / or capabilities of the user. This is illustrated in the accompanying figure, which schematically illustrates a user interface UI 310 used by the ultrasound system to provide user-oriented functionality to a user U 301 to perform a current task at the ultrasound system. The figure also illustrates an extraction operation Extr 320, in which user interaction data UID 330 of one or more interactions of the user U with the ultrasound system may be obtained. Examples of user interactions on the ultrasound unit that may be represented by the user interaction data UID may include badge / fingerprint scanning, button pressing, scan mode selection, access to previous images, and the like. Interactions on the bridge / patient monitor may include reviewing imaging commands set by an expert and accessing previous data about the patient, and the like.

[0076] The figure also shows a determination operation Det 340, wherein a real-time skill estimate RTSE 350 of the current user U may be determined based on the user interaction data UID 330. The real-time skill estimate RTSE may indicate the current ability of the current user to perform the current task. Once a workflow stage has been identified, for that step in the workflow it may be assessed whether one or more user interactions with the system may be utilized to estimate the user's profile. Preferably, only interaction data from the current stage is used to determine the real-time skill estimate, so that customization may be provided for the current user at the current stage. The interaction data may include items representing cumulative features over individual interactions and / or multiple interactions.

[0077] For example, if the workflow is currently in the scan preparation phase, examples of user interactions might include:

[0078] - the sequence and / or timing of button presses on the ultrasound unit, e.g. represented as a button press count for each button, or encoded by a variational autoencoder;

[0079] - the scanning mode used to start the scan, e.g., represented as a categorical variable;

[0080] For example, if the workflow is at the image acquisition stage, examples of user interactions could include:

[0081] - Image scanning mode, including: what intermediate images / features are passed before determining the image of interest, for example, represented by a variational autoencoder; cluster identification for intermediate images according to a clustering algorithm; or manually configured features, such as the count of each type of intermediate image or the time to achieve the target view;

[0082] - data representing the usage pattern of the ultrasound probe, collected from available sensors in the ultrasound unit itself or elsewhere in the ultrasound system, e.g. movement patterns (which are represented as signals or signal features, e.g. 2D signals such as images; image features; 1D signals or other); insertion pressure; or scanned body area;

[0083] - one or more metrics derived from individual interactions, such as time to first image, or how many images a user saved;

[0084] - access to previous scans to use as reference for obtaining the correct image, which is for example expressed as the number of accesses;

[0085] - applying an image quality (IQ) estimation method to estimate the quality of an initially acquired image, e.g. which is represented as a bitmap to which the method has been applied;

[0086] For example, if the workflow is in the analysis phase (including image interpretation, data curation, and / or data transfer), examples of user interactions include:

[0087] - quantification / measurement performed on the saved images, for example expressed as the number and / or type of actions performed;

[0088] - if an automatic quantization method is used, manual editing is performed on the result of the automatic quantization, e.g. which is represented as the number and / or kind of manual edits, e.g. zoom-in and / or zoom-out actions;

[0089] - access to previous scans to use as a reference for obtaining the correct quantification, e.g. the device is expressed as the number of accesses;

[0090] - analyses run on previously acquired images, and features extracted from those analyses, e.g. initiated by the user (features may include the type of analysis and the results of previous analyses), or automated (features may include the results of previous analyses);

[0091] - images curated / selected by the user to be sent to the patient monitor, for example represented as the number of images specified by an earlier expert user and / or the number of images not specified;

[0092] - User interactions on the transmitted image on the patient monitor, for example represented as the number and / or type of operations performed.

[0093] The user interaction data described above may be enhanced by contextual information about the patient being imaged and / or the physical environment in which the ultrasound system is used. Product information of the ultrasound system may also be used, for example, hardware and / or software information such as device type, manufacturer, version, year of manufacture, etc.

[0094] For example, the context information may include characteristics representing the roles and / or number of people present at the ultrasound system, the user's workflow load and / or schedule and / or activities prior to using ultrasound, the location of the ultrasound system, air quality, noise levels, time of day or time relative to the start of a shift. Such aspects may affect the user's physical and / or mental state and, therefore, may affect real-time skill levels.

[0095] For example, information about the patient being imaged may indicate the difficulty in imaging the patient. Patients who are inherently difficult to image, or whose current condition prevents easy image acquisition, may be considered as such, and this may be used to appropriately weight the previously mentioned user interaction data. More generally, using patient information as additional input may be used to better identify patterns followed by people with a particular training level compared to people of a different training level or category.

[0096] The above data can be used to determine an estimate of Det user / user category / user experience level RTSE. The estimate or the skill level derived therefrom can be optionally displayed in a user interface. In an embodiment, the real-time skill level can be determined based on interaction data including at least 5, at least 10 or at least 50 interaction items. The interaction data can be enhanced by a feature vector of at least 2, at least 5 or at least 10 features of, for example, background information. Typically, the determined real-time skill estimate RTSE can be expressed as discrete categories, such as notifications, experts; or on a continuous scale (e.g., from 0 to 100).

[0097] In an embodiment, a trained machine learning model may be used to determine the real-time skill estimate RTSE. The advantage of using machine learning is that high accuracy can generally be achieved. Another advantage is that learning can be automated and continuously updated, and the explicit definition of features to be extracted from interaction data can be avoided. The model can receive interaction data and / or background information as input. Individual interaction items can be represented as individual feature vectors. For example, a feature vector may include features corresponding to the various types of interaction items discussed above, wherein the feature vector of a particular type of interaction item has a non-zero feature vector only for features corresponding to its type. The individual feature vectors can be input into the machine learning model in cumulative form, for example, by summing the individual feature vectors, or sequentially, for example, by using a recursive model, such as a recursive neural network.

[0098] The machine learning model can be parameterized by a set of parameters. For example, the model can be a neural network, a support vector machine, a random forest, etc. Neural networks are also called artificial neural networks. Examples include deep neural networks and convolutional neural networks. In this case, the parameter set can include the weights of the nodes of the neural network. For example, the number of layers of the model can be at least 5 or at least 10, and the number of nodes and / or weights can be at least 1000 or at least 10000. Various known architectures for neural networks and other types of machine learnable models can be used.

[0099] Training of the machine learning model can be performed using stochastic methods, such as stochastic gradient descent, for example, using the Adam optimizer as disclosed in "Adam: A Method for Stochastic Optimization" by Kingma and Ba (available at https: / / arxiv.org / abs / 1412.6980 and incorporated herein by reference). As is known, such optimization methods can be heuristic and / or reach local optima. Training can be performed instance by instance or in batches, for example, up to or at least 64 instances or up to or at least 256 instances. Training of the model can include applying the model to obtain model outputs, and optimizing trainable parameters of the model based on the model outputs, for example, optimizing a loss function based on determined and actual model outputs. Training can be performed by a separate training system based on labeled training data, for example, based on Figure 1 Hardware configuration.

[0100] As an alternative to using a training model for determining Det, a rule-based system (e.g., with manually defined rules) can also be used. For example, individual real-time skill scores can be assigned to individual interaction items according to individual mappings for individual types of interaction items. The mapping can be manually defined. For example, the real-time skill score can be a binary score (indicating or not indicating high real-time skill), or more generally, an N-point score (indicating various degrees of real-time skill), where, for example, N=3 or N=5. As an illustrative example, for the time T at which the first image is obtained, if T<10 seconds, a score indicating high real-time skill can be assigned; if T>1 minute, a score indicating low real-time skill can be assigned; otherwise, a score indicating medium real-time skill can be assigned. Thresholds can be manually defined, but can also be automatically determined based on training data. Individual real-time skill scores can be accumulated to determine an overall real-time skill estimate, for example, as an average, etc. The advantage of using a rule-based system is that interpretability is improved, for example, it is easier for humans to visualize / understand how a certain conclusion is reached.

[0101] In some embodiments, the determination of the real-time skill estimate RTSE also involves determining a confidence level of the output. For example, the machine learning model may output a confidence level that is known per se, e.g., in the form of a classification score. For a rule-based system, the confidence level may be based on, for example, determining the variance of the real-time skill score. For example, if the confidence level exceeds a given threshold, the confidence level may be used by customizing the ultrasound system based only on the real-time skill estimate RTSE.

[0102] In some embodiments, the determined real-time skill estimate RTSE may be specific to a particular role that the user has, for example, regarding the treatment of the current patient (e.g., intensivist, clinician, resident, nurse, etc.). User-oriented functionality may be used to perform the current task at multiple different skill levels and according to multiple different roles. In such cases, for example, individual training models or individual rule-based systems may be used to determine a real-time skill estimate specific to the respective role, and the role-specific real-time skill estimate may then be used to customize the ultrasound system, as also described elsewhere.

[0103] 3) Incorporate historical trends to increase confidence in user profile estimates

[0104] In typical embodiments, the real-time skill estimate RTSE is determined independently of any user identification.In some embodiments, for example, in embodiments where user identification has been done anyway, however, such user identification may be employed to refine the real-time skill estimate.

[0105] Specifically, the user can be identified and the determined real-time skill estimate RTSE can be confirmed based on the user's user profile data. For example, once the initial estimated RTSE of the user has been completed, the confidence of the estimate can be increased by polling the source of patient data (e.g., an institutional database or a US4M database, as applicable) to obtain historical information related to the user. For example, if the initial estimate indicates a certain level of experience / expertise (e.g., based on a sequence of button presses), historical data for that level of experience and / or expertise can be polled to find a close match with a specific user. The match can increase the confidence in the determined real-time skill estimate RTSE, for example, the degree of confidence can be increased as discussed above.

[0106] In the case where user identification is performed, the user profile can also be updated based on the determined real-time skill estimate, for example, the skill level stored in the user profile can be updated. Typically, as the user experience level progresses over time, the provided technology enables the ultrasound system to change the type of feedback provided to the user due to the changing real-time skill estimate RTSE. Thereafter, this information can also be used to update the user profile. Thus, the user's expertise level in the profile can be gradually estimated. For example, at the beginning, the system starts in a fully automated mode, and as data is concentrated, the level of automation and feedback can be changed based on the real-time skill estimate.

[0107] The identification can be performed using conventional means, for example, by obtaining a user identifier from a scanned badge. Alternatively, it is also possible to identify the user based on user interaction data UID, for example, based on usage patterns, such as button press sequences (for example, the user habitually often scans using a specific combination of settings) and / or selected scanning modes (for example, the user often follows the same sequence of actions to achieve the desired system state), and / or based on sensor measurements, for example, measurements of a camera and / or microphone. In this way, a manual identification step by the user is not required. When there is no manual identification of the user and the identification has been performed, the user can be asked to confirm the identification via the user interface. Optionally, the user profile associated with the identified user can indicate whether the user has opted in or out of the profile for the purpose of customizing the ultrasound system. In this way, only information about users who have opted in can be presented via the user interface, which improves privacy.

[0108] 4) Estimate the current task

[0109] The customization unit may obtain the current task CT360 being executed and whose execution may be customized in various ways. In some embodiments, for example, the current task may be manually selected and / or explicitly initiated by a user via a user interface.

[0110] In other embodiments, as also shown in the figure, the customization unit can also be configured to derive the current task CT that the current user is performing based on the user interaction data UID. For example, the current task can be estimated based on analyzing the initial images from the scan and / or from the button pressing sequence and system conversion of the user performing the task.

[0111] To determine the current task CT, the customization unit may also use data captured from an expert guidance subsystem of the ultrasound system (such as a Reacts framework or the like). For example, an expert user may be located at a remote location and may guide a user at the bedside who is determined to have a low real-time skill estimate. In such a case, the system may supplement the guidance provided by the expert user, such as by presenting the guidance in a manner that is easy for the bedside user to interpret, or by enhancing the guidance with data or information relevant to the task at hand. For example, if the remote expert user suggests the acquisition of an A4C image, the system may call up an acquisition guidance module for A4C ​​acquisition and / or previous A4C images, which may provide a template for an inexperienced bedside user.

[0112] 5) Provide relevant assistance and / or simplify workflow

[0113] Based on the real-time skill estimate RTSE and the current task CT, the customized operation Cust 370 can customize the ultrasound system to perform the current task at a skill level corresponding to the real-time skill estimate of the current user. Depending on the skill level, customization can include providing or not providing user guidance. Depending on the skill level, customization can also include automating or non-automating the operation. Customization can optionally further consider the user's role (e.g., chief intensivist, junior clinician, resident, nurse, general practitioner, midwife, sports doctor) to perform customization and provide user-oriented functions accordingly. Similar to the current task CT, the current role can be derived in various ways, for example, based on user interaction data UID; or by manual selection.

[0114] Now some specific examples of customization are provided.

[0115] In one example, depending on the skill level, the customization unit may automatically retrieve and display, or not automatically retrieve and display information about previous scans for the current follow-up scan. For example, the system may determine that the current task is associated with a clinical order for a follow-up scan and identify the previous scan from the clinical order. The information may be tagged with user information.

[0116] In one example, depending on the skill level, the customization unit may or may not automatically preload scan settings appropriate for the current task. For example, the current task may involve wire placement, in which case the wire placement settings may be preloaded, for example, system settings for device visualization, such as tissue-specific presets. For example, the settings may include one or more of depth, focus, harmonics, NeedleViz mode activation, transmit frequency, gain, and dynamic range. For example, by identifying a needle, wire, or sheath in an ultrasound image, the wire position may be automatically detected. This customization may be performed specifically if the current user is a nurse.

[0117] In one example, the current task may involve obtaining a given view of an organ. In such a case, depending on the skill level, the customization unit may or may not show example images of the given view of the organ, such as schematic images or images of actual previous scans. For example, if the initial images from the scan indicate that the user is trying to obtain an apical view of the heart (e.g., A2C / A4C), the ultrasound system may be configured to provide assistance by calling up examples of A2C / A4C views from previous patients that match the physical habits of the current patient.

[0118] In one example, depending on the skill level, the customization unit may automatically retrieve one or more relevant ultrasound images and / or measurements from a database.

[0119] In one example, depending on skill level, the customized unit may or may not provide feedback regarding the status of the ultrasound scan being performed.

[0120] In one example, depending on the skill level, the customization unit may or may not activate guidance in the form of assistance provided by the telehealth application, and / or determine the nature (eg, type and / or intensity) of the assistance provided by the telehealth application.

[0121] In an embodiment where user identification is performed, the customization unit can be configured to customize the ultrasound system based on a user profile. Specifically, the Custom Cust can use manually provided customization settings. For example, once it recognizes that the current user is Nurse #1, Nurse #1 can program the system to convert itself to certain settings they like. In another example, the Director of Critical Care can decide what feedback needs to be provided by any user classified as a "Junior Resident". The Custom Cust can also use usage patterns learned about users. In this way, the system can learn to provide user-specific assistance based on knowing what the user needed when performing similar scans in the past. For example, if a user continues to ignore or turn off a certain instruction, the system can be configured not to provide that instruction the next time the user performs the same task.

[0122] Typically, extracting Extr, determining Det, and customizing Cust can be performed continuously in the background, for example, determining Det or customizing Cust can be performed periodically and / or based on the availability of new user interaction data UID. Specifically, for a given user, the first real-time skill estimate RTSE can be updated to an updated real-time skill estimate based on new user interaction data, for example, can be updated once or more than once. For example, once it is determined that the real-time skill estimate RTSE has a confidence level exceeding a threshold, customizing Cust can be executed. For a given task, customization can be performed only once, but customization may also be performed multiple times when determining an updated estimate of the real-time skill estimate RTSE. When the user manually modifies the skill level, further automatic customization for the current task is generally not performed.

[0123] Figure 4 A block diagram of a computer-implemented method 400 for customizing an ultrasound system is shown. The ultrasound system may be operated by one or more users of the ultrasound system to process ultrasound measurements of a patient. The ultrasound system may include an ultrasound unit operable to perform ultrasound measurements of a patient. The method 400 may correspond to Figure 1 Customized units 100 and / or Figure 2 However, this is not a limitation, as the method 400 may also be performed using another system, device, or apparatus.

[0124] The method 400 may include, in an operation entitled "Obtain User Interactions," obtaining 410 user interaction data of one or more interactions of a current user performing a current task at the ultrasound system. The ultrasound system may provide user-oriented functionality at multiple different skill levels to perform the current task.

[0125] Method 400 may include, in an operation entitled “Estimate Real-Time Skills,” determining 420 a real-time skill estimate for a current user based on the user interaction data. The real-time skill estimate may indicate a current ability of the current user to perform a current task.

[0126] The method 400 may include, in an operation entitled "Customize," customizing 430 the ultrasound system for performing a current task at a skill level corresponding to a real-time skill estimate of a current user.

[0127] It will be understood that, generally, Figure 4 The operations of method 400 may be performed in any suitable order, eg, serially, concurrently, or a combination thereof, where applicable, where a particular order is necessary, eg, by input / output relationships.

[0128] The method(s) may be implemented on a computer as a computer-implemented method, dedicated hardware, or a combination of both. Figure 5As shown, instructions for a computer, e.g., executable code, can be stored on a computer readable medium 500, e.g., in the form of a series 510 of machine-readable physical indicia and / or as a series of elements having different electrical (e.g., magnetic) or optical properties or values. The medium 500 can be transient or non-transient. Examples of computer readable media include memory devices, optical storage devices, integrated circuits, servers, online software, etc. Figure 5 An optical disc 500 is shown. Alternatively, the computer readable medium 500 may include data 510 representing a machine learnable model trained to determine real-time skill estimates from user interaction data.

[0129] Examples, embodiments or optional features, whether or not indicated as non-limiting, should not be construed as limiting the invention as claimed.

[0130] It should be noted that the embodiments mentioned above illustrate rather than limit the present invention, and that a person skilled in the art will be able to design many alternative embodiments without departing from the scope of the claims. In the claims, any reference numerals placed between brackets should not be interpreted as limiting the claims. The use of the verb "comprise" and its verbal changes does not exclude the presence of elements or stages other than those listed in the claims. The quantifier "one" or "an" in front of an element does not exclude the presence of multiple such elements. Expressions such as "at least one of..." when preceding a list or group of elements indicate that all or any subset of elements are selected from the list or group. For example, the expression "at least one of A, B, and C" should be understood to include only A, only B, only C, both A and B, both A and C, both B and C, or all of A, B, and C. The present invention can be implemented by hardware including several different elements and by means of a suitably programmed computer. In a device claim that lists several devices, several of these devices may be embodied by the same hardware. The mere fact that specific measures are recorded in mutually different dependent claims does not indicate that a combination of these measures cannot be used advantageously.

Claims

1. An ultrasound system (200), in, The ultrasound system is operable by one or more users of the ultrasound system to process ultrasound measurements of a patient, the system comprising: - an ultrasound unit (210), wherein the ultrasound unit is operable to perform the ultrasound measurements on the patient; - Customization units (210, 220, 230, 240) configured to: - obtaining user interaction data of one or more interactions of a current user performing a current task at the ultrasound system, wherein the ultrasound system provides user-oriented functionality for performing the current task, wherein the user-oriented functionality is provided at a plurality of different skill levels, the plurality of different skill levels differing in the amount of user guidance and / or automation provided; - determining a real-time skill estimate of the current user based on the user interaction data, wherein the real-time skill estimate indicates a current ability of the current user to perform the current task; - customizing the ultrasound system to perform the current task at a skill level corresponding to the real-time skill estimate of the current user.

2. The ultrasound system (200) according to claim 1, in, The customization unit is configured to determine the real-time skill estimate independently of any user identification.

3. The ultrasound system (200) according to any preceding claim, in, The ultrasound system is an ultrasound system for monitoring, wherein the ultrasound system for monitoring further comprises a patient monitoring unit, the patient monitoring unit being configured to display information based on the ultrasound measurements performed, and wherein the user performs the current task at the patient monitoring unit or at a bridge device of the ultrasound system for monitoring, the bridge device being used for information exchange between the ultrasound unit and the patient monitoring unit.

4. The ultrasound system (200) according to any preceding claim, in, The user-facing functionality is provided at a number of different skill levels and according to a number of different clinical roles.

5. The ultrasound system (200) according to any preceding claim, in, The customization unit is further configured to derive the current task being performed by the current user from the user interaction data.

6. The ultrasound system (200) according to any preceding claim, in, The current task is part of a current stage of multiple stages of the processing of the ultrasound measurement, wherein the real-time skill estimate for the current stage is determined solely based on the interaction of the current user in the current stage, optionally using real-time skill estimates and / or roles derived in earlier stages.

7. The ultrasound system (200) according to any preceding claim, in, The user interaction data includes multiple interaction items, including one or more interaction items representing individual interactions of the user in performing the current task and / or one or more interaction items representing cumulative quantities determined during multiple interactions of the user with the system in performing the current task and / or one or more interaction items representing sensor measurements of the ultrasound system in addition to the ultrasound measurements performed by the ultrasound unit.

8. The ultrasound system (200) according to any preceding claim, in, The customization unit is configured to determine the real-time skill estimate based on contextual information about the physical environment in which the ultrasound system is used.

9. The ultrasound system (200) according to any preceding claim, in, The customization unit is configured to customize the ultrasound system by providing or not providing user guidance according to the skill level and / or automating or non-automating operations according to the skill level.

10. The ultrasound system (200) according to any preceding claim, in, The customization unit comprises a user interface configured to enable the user to override the skill level at which the ultrasound system was customized.

11. The ultrasound system (200) according to any one of the preceding claims, in, The ultrasound system is used in critical care and / or perioperative settings.

12. The ultrasound system (200) according to any one of claims 1-10, in, The ultrasound system is intended for use outside of a hospital, such as at home, in an ambulance, or at a sports facility.

13. A custom unit (100) for an ultrasound system according to any one of the preceding claims, in, The customization unit comprises a processor subsystem (140), wherein the processor subsystem is configured to: obtain the user interaction data (010); determining the real-time skill estimate; and customizing the ultrasound system.

14. A computer-implemented method (400) for customizing an ultrasound system, in, The ultrasound system is operable by one or more users of the ultrasound system to process ultrasound measurements on a patient, wherein the ultrasound system comprises an ultrasound unit operable to perform the ultrasound measurements on the patient, wherein the method comprises: - obtaining (410) user interaction data of one or more interactions of a current user performing a current task at the ultrasound system, wherein the ultrasound system provides user-oriented functionality for performing the current task, wherein the user-oriented functionality is provided at a plurality of different skill levels that differ in the amount of user guidance and / or automation provided; - determining (420) a real-time skill estimate of the current user based on the user interaction data, wherein the real-time skill estimate indicates a current ability of the current user to perform the current task; and - customizing (430) the ultrasound system to perform the current task at a skill level corresponding to the real-time skill estimate of the current user.

15. A transitory or non-transitory computer readable medium (500) comprising data representing instructions (510) which, when executed by a processor system, cause the processor system to perform the computer-implemented method of claim 14.