Device for acquisition of doppler ultrasound data, and associated method

Abstract

The invention relates to a device for assistance in acquisition of ultrasound data, the device being configured to receive, in real time, ultrasound images captured by an ultrasound probe connected to the device, and comprising:  a first computer, associated with a first memory, for implementing a classifier configured to classify, in real time, first received ultrasound images (11) and to associate a class (14) with each image in order to generate a quality indicator (13) as a function of the associated class (14);  a second computer, associated with a second memory and configured to transmit an activation command (63) to the ultrasound probe (SECH) to capture an ultrasound image with a Doppler ultrasound mode.

Description

[0001] DEVICE FOR ACQUIRING DOPPLER ULTRASOUND DATA AND ASSOCIATED METHOD

[0002] Scope of the invention

[0003] The present invention relates to a device and a method for automating and assisting an operator in acquiring ultrasound images. Advantageously, the present invention allows for the automatic determination of a Doppler measurement location on a received ultrasound image.

[0004] State of the art

[0005] Ultrasound imaging, also known as sonography, is a medical imaging technique that uses high-frequency waves to visualize two-dimensional structures inside a living person's body. Because ultrasound images are taken in real time, they show the movement of internal organs, such as the heart's movements during its beat.

[0006] To acquire such images, an ultrasound probe is placed directly on the subject's skin. A thin layer of gel may be applied to the skin to allow the ultrasound waves to pass through the skin from the probe to the subject's body. Ultrasound images are produced by measuring the reflection of these ultrasound waves off the subject's organs. The amplitude of the reflected waves and their reflection time provide the information needed to reconstruct the ultrasound image.

[0007] During a cardiac ultrasound examination, the operator must know where to place the probe and how to orient it to obtain the desired image. Generally, the desired images from a cardiac ultrasound are those that allow for specific measurements, particularly of organ or tissue dimensions.

[0008] To increase the capacity for ultrasound examinations, the idea arose of relieving cardiologists of the image acquisition burden. For example, images can be taken by an operator who records them and then transmits them to a cardiologist who would then only need to interpret them. Such an arrangement allows a cardiologist to perform more ultrasound examinations. However, it is then essential to ensure that the ultrasound images are of sufficient quality for interpretation by a cardiologist. The operator may have ultrasound knowledge, but when they will not be performing the image analysis themselves, there is a need for a device to assist in acquiring such ultrasound images. Furthermore, the operator must not just acquire a single image of sufficient quality, but a sufficiently long sequence of images, for example, covering a complete heartbeat cycle, so that the cardiologist can make their diagnosis and / or perform an analysis.

[0009] Software-implemented methods exist to guide the operator in acquiring echocardiographic images of the heart. These methods include evaluating the quality of an image sequence against a particular view and recording such a sequence.

[0010] One drawback of these methods is that it remains difficult for the operator to assess the quality of the image acquired and whether it is sufficient. Another drawback is that it can be difficult to obtain a sufficiently long sequence of images meeting the required quality, further complicating the operator's task, as their image capture will not be validated by the software.

[0011] Furthermore, spectral analysis is often used for certain diagnostic purposes to assess blood flow direction and measure the velocity of heart valves, vascular lesions, and other pathological lesions (an area of ​​an organ or tissue damaged by injury or disease). Spectral analysis can be performed using continuous wave (CW) or pulsed wave (PW) Doppler, color Doppler, or tissue Doppler imaging (TDI).

[0012] To measure blood flow velocity using spectral analysis, the ultrasound operator defines a region of the ultrasound cone within which the measurement is to be taken, using measurement localization controls that vary depending on the Doppler mode. For CW mode, the operator specifies a Doppler line containing the origin at the apex of the ultrasound cone and a point at the curved base of the cone, and the highest velocity at any point on the line is determined. For PW mode, the operator specifies both a Doppler line and a segment of that line, defined by placing a gate to delimit it, and the velocity along that segment of the line is determined.For color Doppler, the operator specifies a window (or region of interest) delimiting a sample volume within a range of depths across a contiguous sequence of scan lines, in order to display the velocity and direction of flow within that sample volume using color. The optimal placement of these Doppler measurement controls may vary depending on the structure being examined, the suspected pathology, and the cardiac view.

[0013] An additional challenge for the operator is therefore correctly positioning the Doppler measurement site. This difficulty is particularly detrimental to tissue Doppler, commonly used in cardiac examinations, where the quality and accuracy of measurements can be influenced by the operator's experience.

[0014] Therefore, there is a need for a new device to facilitate the operator in taking ultrasound images.

[0015] The invention therefore aims to provide a device and an associated method enabling operators to acquire ultrasound image sequences of sufficient quality and including Doppler measurements more easily and more quickly.

[0016] Summary of the invention

[0017] The invention relates to a device for assisting in the acquisition of ultrasound data, configured to receive in real time ultrasound images acquired by an ultrasound probe connected to said device.

[0018] The said device includes a first computer, associated with a first memory, to implement a classifier configured to classify in real time the first ultrasound images received and to associate each image with a class to generate a quality indicator according to said associated class.

[0019] The said device includes a second calculator, associated with a second memory.

[0020] In one embodiment, the first calculator and the second calculator are the same calculator.

[0021] The second calculator is configured to perform the following steps: ■ when a sequence of a predefined number of received images includes a rate of images associated with the same class greater than a predefined rate, the determination of the location of at least one predefined anatomical structure detected in a received ultrasound image;

[0022] ■ the determination of a measurement location based on the determined location of said detected anatomical structure;

[0023] ■ the transmission to the ultrasound probe connected to said device of an activation command for one or more Doppler ultrasound modes using the determined measurement location.

[0024] One advantage of the invention is that it allows the sequential acquisition of ultrasound images from a desired view, followed by the acquisition of Doppler data at a predetermined position on that image. Another advantage is that it allows a non-specialist user to perform Doppler acquisitions of an organ without having to manually position the Doppler measurement area. This results in increased speed of the overall ultrasound examination, the possibility of performing the examination by non-specialist personnel, and increased reliability of the acquired measurements.

[0025] In one embodiment, at least one anatomical structure includes an identifiable point or linear landmark such as a tissue interface or an anatomical edge characteristic of an organ.

[0026] In one embodiment, at least one predefined anatomical structure comprises a first anatomical structure and a second anatomical structure different from the first anatomical structure.

[0027] Another advantage of the invention is that it detects the location of two predetermined anatomical structures, allowing for greater accuracy in the measurement area. Indeed, detecting two structures improves spatial referencing compared to detecting a single structure. Preferably, each anatomical structure comprises a point. Determining the location of two points on the ultrasound image advantageously enables a more reliable determination of the measurement area. In one embodiment, the second processor is further configured to perform a step of receiving at least one Doppler ultrasound data point produced by the probe using one or more Doppler ultrasound modes.

[0028] In one embodiment, the second computer is configured to, in addition, perform a predetermined two-anatomical structure detection step on the received ultrasound image.

[0029] In one embodiment, the predefined anatomical structures include a first anatomical structure defined by the junction between a first leaflet of the valve of a ventricle of the heart and the wall of said ventricle and a second anatomical structure defined by the junction between a second leaflet of said valve and the wall of said ventricle of the heart.

[0030] One advantage is to determine two precise anatomical structures, allowing a measurement area to be determined based on these two points more reliably and accurately than with the simple detection of an area such as the ventricle of the heart.

[0031] In one embodiment, the device includes a display for, in real time, displaying the first ultrasound images received, at least one Doppler ultrasound data point and / or the last generated quality indicator.

[0032] In one embodiment, the determined measurement location includes a measurement point, a measurement line, or a measurement surface on a first received ultrasound image.

[0033] In one embodiment, the device includes a third computer, associated with a third memory, and configured to automatically perform the recording of said image sequence in a fourth memory when a sequence of a predefined number of received images includes a rate of images associated with the same class greater than a predefined rate.

[0034] In one embodiment, the second computer is configured to perform the step of detecting at least two anatomical structures appearing in the received ultrasound image once the third computer has executed the step of automatically recording said image sequence. An advantage is the ability to trigger Doppler measurements once a non-Doppler ultrasound image sequence has already been acquired.

[0035] In one embodiment, the classifier is configured to classify an image among: a plurality of classes, each representing a particular view of the organ; and at least one class representing a view of insufficient quality.

[0036] In one embodiment, the classifier is implemented by means of a learning function configured from supervised machine learning.

[0037] In one embodiment, the at least one Doppler ultrasound data includes at least one second (or a second) ultrasound image such as a Doppler ultrasound image generated from ultrasound data acquired in Doppler mode.

[0038] In one embodiment, the second computer is further configured to perform an automatic recording step of a sequence of multiple images, each comprising the second ultrasound image superimposed on at least one received first ultrasound image. An advantage is the ability to superimpose a non-Doppler black and white ultrasound image with the received ultrasound data to improve visualization for the clinician.

[0039] In another aspect, the invention relates to a system comprising a device according to the invention and an ultrasound probe adapted to be connected to said device so as to transmit the ultrasound images captured by the probe to the device. The ultrasound probe comprises a Doppler ultrasound probe.

[0040] According to another aspect, the invention relates to a computer program product comprising instructions that lead a device according to the invention to perform the steps according to the invention.

[0041] In another aspect, the invention relates to a computer program product comprising instructions that lead a device to execute the following steps:

[0042] • the receipt of the first ultrasound images;

[0043] • real-time classification of the first ultrasound images received and association of each image with a class to generate a quality indicator based on said associated class;

[0044] • when a sequence of a predefined number of received images includes a rate of images associated with the same class greater than a predefined rate, the detection of at least two predefined anatomical structures appearing in the received ultrasound image;

[0045] • the localization to determine the location of each of the said anatomical structures detected in the received ultrasound image;

[0046] • the determination of a measurement location in relation to the determined location of said detected anatomical structures;

[0047] • the transmission to an ultrasound probe connected to said device of a command to activate one or more Doppler ultrasound modes using the determined measurement location;

[0048] • the reception of at least one Doppler ultrasound data produced by the probe using one or more Doppler ultrasound modes.

[0049] According to another aspect, the invention relates to a computer-readable medium, for example a non-transient medium such as a memory, on which the computer program according to the invention is recorded.

[0050] Brief description of the figures

[0051] Other features and advantages of the invention will become apparent from the detailed description that follows, with reference to the attached figures, which illustrate:

[0052] Figure 1: A schematic view of the image displayed by the display of the device according to a first embodiment of the invention.

[0053] Figure 2: A schematic view of a device according to one embodiment of the invention.

[0054] Figure 3: A flowchart representing the different steps of a process according to one embodiment of the invention. Figure 4: A schematic representation of the heart illustrating the two predefined anatomical structures to be located on the ultrasound image. These anatomical structures are represented by two points.

[0055] Figure 5: a first example of a measurement location generated by an embodiment of the invention in which said measurement location includes a point 51 equidistant from the two anatomical structures.

[0056] Figure 6: a second example of a measurement location generated by an embodiment of the invention in which said measurement location comprises a segment 52 defined by the position of the two anatomical structures according to a predefined rule.

[0057] Figure 7: a third example of a measurement location generated by an embodiment of the invention in which said measurement location comprises a surface 53 defined by the position of the two anatomical structures according to a predefined rule.

[0058] Detailed description

[0059] The invention relates to a method for assisting in the generation of an ultrasound image sequence. The invention is particularly advantageous for generating an ultrasound image sequence of the heart. Indeed, the heart is an organ subject to a cycle.

[0060] The cardiac cycle consists of two phases: one during which the heart muscle relaxes and fills with blood, called diastole, followed by a phase of vigorous contraction and pumping of blood, called systole. After emptying, the heart immediately relaxes and expands to receive another influx of blood returning from the lungs and other body systems, before contracting again to pump blood back to the lungs and these systems. A normally functioning heart must be fully expanded before it can pump efficiently again.

[0061] To analyze potential pathologies, it is important for the practitioner to obtain a sequence of images (or a video) of the entire cardiac cycle. The invention advantageously assists the practitioner in acquiring such a sequence of images.

[0062] However, the invention may also find advantages in the ultrasound acquisition of other organs or where a sequence of images is desirable and / or when Doppler ultrasound image acquisitions are required for diagnosis, for example in the field of obstetric ultrasound.

[0063] The invention also relates to an associated device. An example of such a device 1 is illustrated in Figure 2.

[0064] Device 1 may include a tablet, a smartphone, a computer, or any other device comprising at least a display and a processor associated with memory. In another aspect, the invention also relates to a system 2 comprising a device 1 according to the invention and a SECH ultrasound probe connected to said device 1.

[0065] In one embodiment, the device 1 comprises software and hardware means for implementing the process according to the invention described below. Said device preferably comprises at least one REC receiver, a classifier, and a display.

[0066] RECEPTION

[0067] In one embodiment, the method includes a step of receiving 100 real-time ultrasound images 11. These ultrasound images 11 are received by a REC receiver of the device. In one embodiment, the REC receiver may include a buffer in which the images received by the REC receiver are temporarily stored before transmission.

[0068] The received ultrasound images 11 are preferentially displayed in real time, for example on an AFF display of the device 1. The received ultrasound images 11 can therefore be transmitted to an AFF display for their real-time display.

[0069] In a preferred embodiment, the received ultrasound images are not ultrasound images acquired using a Doppler ultrasound mode. In another embodiment, the acquired ultrasound images include two-dimensional ultrasound images, also known as B-mode images.

[0070] Device 1 may include an AFF display such as a screen, for example, a monitor screen, a touchscreen tablet, or a smartphone. The AFF display is connected to the REC receiver and / or the CLASS classifier and / or the CALC processor. The AFF display is configured to show in real time the ultrasound images 11 received by the REC receiver. The AFF display advantageously allows the operator to have real-time feedback on the ultrasound images being acquired. The AFF display is configured to show additional indicators or data, which will be described later in this document. The AFF display shows an image 1 comprising an ultrasound image received 11 by the REC receiver. Preferably, the ultrasound image displayed is the last ultrasound image received by the REC receiver.The AFF display is thus configured to display in real time the images captured by the SECH ultrasound probe connected to the device according to the invention.

[0071] CLASSIFICATION

[0072] In one execution mode, the method includes a classification step of 200 ultrasound images 11 received by the REC receiver in real time. The classification is performed by a CLASS classifier of the device.

[0073] The classification of an ultrasound image includes the association of said ultrasound image with a class 14, preferably a class from a predefined group of classes.

[0074] In one execution mode, the classification further includes the generation of a quality indicator based on the class associated with the ultrasound image. The quality indicator may be representative of a single class or a predefined set of classes.

[0075] In one execution mode, the classification step also includes the generation of a confidence value. This confidence value can represent the confidence level of the classification performed. The confidence value is generated by the CLASS classifier.

[0076] The classification step is preferably performed by a CLASS classifier within the device. A classifier is defined as an algorithmic function executed by a processor or computer connected to a memory. The CLASS classifier is configured to receive images from the REC receiver.

[0077] The classifier is configured to classify in real time at least a portion of the images received by the REC receiver. The classifier is configured to associate a class with each ultrasound image. The classifier generates a quality indicator 13 based on the associated class. The quality indicator 13 is preferentially associated with the classified ultrasound image.

[0078] Preferably, the device includes image processing means to process the received images before providing them to the CLASS classifier. These processing means may include image filters or contrast functions.

[0079] Preferably, the CLASS classifier is configured to classify an ultrasound image of the heart among a plurality of classes comprising, on the one hand, classes each representing a particular view of the heart and, on the other hand, a class representing a view of insufficient quality.

[0080] Thus, when an image is classified into a class representing a particular view of the heart, that image is considered to be of sufficient quality to allow the clinician to advantageously perform predefined measurements from these images. Indeed, cardiologists use specific, well-identified views for their diagnoses.

[0081] In one embodiment, each class representing a particular view of the core represents at least one of the following views:

[0082] • a view of a parasternal long-axis section

[0083] • a view of a parasternal short-axis section

[0084] • a view of an apical section

[0085] • a view of a subcostal section

[0086] • a view of a suprasternal section

[0087] • a view of a right parasternal section

[0088] Each of these views is well known to cardiologists and is used to visualize different parts of the heart, calculate or measure specific data, and identify certain pathologies.

[0089] Each of these views can be characterized by the presence of one or more specific parts of the heart. For example, a parasternal short-axis view is characterized by the presence on the image of the left and right ventricles. From this view, the cardiologist can calculate the fractional shortening and pulmonary pressures.

[0090] In one embodiment, the class representing a view of insufficient quality represents ultrasound views of the heart not belonging to one of the specific views of the other classes or the specific ultrasound views listed above, but whose image quality does not allow the necessary measurements to be carried out related to that specific view.

[0091] In one embodiment, the CLASS classifier is implemented using a learning function trained through supervised and / or automatic learning. The learning function preferably comprises a neural network. The learning function is preferably trained on a series of labeled echocardiographic images of the heart.

[0092] In one example, the learning function was configured using a training method that involved submitting a plurality of ultrasound images, each associated with a label, to the classifier. In another embodiment, the classifier was trained with ultrasound images of a specific view of the heart of sufficient quality, each image being associated with a label representing that specific view. The classifier was also trained with ultrasound images of views of the heart other than those mentioned above, or not of sufficient quality to allow the cardiologist to perform measurements from those images, each of these images being associated with a quality label representing a view of insufficient quality.Labels may also include information characteristic of an image of an organ, such as its viewing angle and / or size, image quality, the presence of a specific portion of the heart, or the presence of a measurable anatomical contour. Preferably, labels include the name of the particular view of the heart or a name associated with a poor-quality view.

[0093] The CLASS classifier can be executed in a processor, which is itself associated with memory. The CLASS classifier can be stored in a computer-readable medium, such as memory associated with said processor. In one execution mode, the ultrasound image classification step includes generating a matching score for each class. The matching score may include a probability that the ultrasound image belongs to each class. The process then includes selecting the class with the highest score. Preferably, the confidence value is generated from said matching score of the class associated with said ultrasound image.

[0094] In one embodiment, the CLASS classifier classifies 100% of the images received in real time.

[0095] In an alternative embodiment, the CLASS classifier is configured to classify only a portion of the received images. The classifier is then configured to classify a subset of the ultrasound images received in real time. This embodiment is particularly advantageous when the classification speed is lower than the image acquisition speed (in images per unit of time).

[0096] In one execution mode, the classification of an ultrasound image includes the classification of the last ultrasound image received. Once that image has been classified, the process again uses the last ultrasound image received. Thus, it is possible that some ultrasound images may not be classified. However, this execution mode advantageously allows for real-time classification of ultrasound images, regardless of the speed of the classifier or the frequency of ultrasound image reception.

[0097] QUALITY INDICATOR

[0098] In one embodiment, the method according to the invention comprises generating and preferably displaying in real time on the AFF display the generated quality indicator 13. Preferably, the displayed quality indicator 13 corresponds to the last generated quality indicator 13 and / or corresponds to a value representative of a predefined number of the last generated quality indicators.

[0099] In one execution mode, this quality indicator 13 is generated by the CLASS classifier. Such a classifier is configured to, upon receiving an ultrasound image of the organ, classify said image into one of the classes mentioned above. The method then includes generating a quality indicator representative of the classification of this image. This quality indicator can be associated with the classified ultrasound image.

[0100] Alternatively, the quality indicator 13 can be generated by a remote processor connected to the AFF display and receiving information from the classifier.

[0101] The display of the quality indicator 13 may include a color indicator whose color depends on the class associated with the classified ultrasound image. The indicator color may represent a single class. Preferably, the color indicator may have two distinct colors: a first color representing a group of classes including the classes representing a particular view of the organ, such as one of the specific views of the heart mentioned above, and a second color representing a class representing a view of insufficient quality. This allows the operator to quickly determine whether the ultrasound image being acquired is of sufficient quality.

[0102] In one embodiment, the classifier is also configured to generate a confidence value 21 upon receiving an ultrasound image of the organ. This confidence value 21 can represent the level of certainty that the image has been correctly classified. In one embodiment, the quality indicator may include this confidence value. The confidence indicator 21 is preferably displayed by the AFF display.

[0103] The quality indicator 13 associated with the classified image may include said confidence indicator. In another alternative mode, a confidence indicator representing the confidence value is generated and displayed in real time.

[0104] In the first example shown in Figure 1, the displayed quality indicator 13 includes a colored frame that extends preferentially around the displayed ultrasound image. This allows the operator to easily visualize whether the image being acquired is indeed the expected one without having to look away from the image. The confidence indicator can be displayed as a numerical value, as illustrated in Figure 1.

[0105] In a second example illustrated in Figure 3, the quality indicator 13 includes a color indicator. The color of the color indicator represents the classification of the last ultrasound image. At least one dimension of this color indicator is a function of the confidence level. In the example shown in Figure 3, the quality indicator 13 is a bar whose length varies according to the confidence level and whose color depends on the class associated with the last ultrasound image.

[0106] Such an indicator advantageously allows the operator to visualize whether the image being acquired is of sufficient quality and also allows them to see, without looking away, whether the indicator is degraded or improved depending on the movement of the SECH ultrasound probe. The benefit of such an indicator is that it allows the operator, when slightly moving the probe, to visualize whether this movement increases or decreases the confidence level, and consequently increases or decreases their chances of obtaining a validated image sequence as described below.

[0107] SEQUENCE

[0108] In one embodiment, the method includes the automatic recording 400 of a sequence 20 of images in a MEM memory. This sequence 20 is automatically recorded when a sequence of a predefined number of received or classified images contains a proportion of images associated with the same class 14 greater than a predefined proportion. Preferably, said class is a class representing a particular view of the core.

[0109] The term "image sequence" refers to a series of ultrasound images that follow one another in the chronological order of their acquisition. Therefore, the term "video sequence" can also be used to refer to such a sequence of images.

[0110] Thus, when a sufficiently long video contains, for example, a majority of ultrasound images classified in the same class representing a particular view of the organ, the video sequence is automatically recorded in memory. This automatic recording advantageously generates videos of a specific view of the organ that can be analyzed by a cardiologist without operator validation.

[0111] The predefined number of received or classified images can be understood as a predefined minimum duration, provided the SECH ultrasound probe's image acquisition rate is constant. The advantage of this threshold is that it ensures the video sequence is long enough for analysis by a cardiologist. In an alternative implementation, the predefined number of images can be replaced by a number of received or classified images. This predefined number can be configured to correspond to a predefined number of cardiac cycles in the subject's heart.

[0112] The percentage of images associated with the same class in the video sequence exceeding a predefined threshold allows for automatic recording despite a negligible number of images lacking sufficient quality. Indeed, this negligible number of images in a video sequence may stem from a classification error or noise related to a particular image. This tolerance represents a good compromise between the ease of automatic sequence generation and the quality of the resulting sequence.

[0113] In one execution mode, the predefined rate is at least greater than 50%. In another execution mode, the predefined rate can be set by the operator, for example via a user interface of the device.

[0114] In one execution mode, the automatic recording step includes buffering continuously received ultrasound images along with generated quality indices associated with those images. In this way, when a video sequence meeting the criteria stated above is detected, the images belonging to that sequence can be transferred from the buffer to another memory location and / or grouped into a single file to generate a video sequence.

[0115] For the sake of clarity, in the rest of this description, the generation and recording of such a sequence will be referred to as the "validation" of such a sequence.

[0116] TIME INDICATOR

[0117] In one execution mode, the process includes a step of generating and displaying in real time a time indicator 12.

[0118] The displayed time indicator 12 represents the number of images remaining to be received or classified to achieve a sequence of a predefined number of received images, including a percentage of images associated with the same class exceeding a predefined rate. This indicator advantageously allows the operator to visualize the time remaining to validate a video sequence of a particular view of the organ. This indicator also advantageously allows the operator to visualize the remaining time during which they must maintain an ultrasound image of sufficient quality, this quality being displayed by the quality indicator 13 and / or the confidence indicator.

[0119] Time indicator 12 can include a numerical value such as a countdown. Time indicator 12 can include a gauge that fills or empties depending on the time remaining to validate a video sequence.

[0120] In one execution mode, the method includes, when a first image is classified into a class representing a particular view of the organ, the automatic generation and display of the temporal indicator 12.

[0121] With each new classified image, the temporal indicator 12 is preferentially updated in real time:

[0122] - if the number of images classified since the first image includes a rate of images associated with the same class which is greater than the predefined rate, then the time indicator 12 is updated by indicating the time remaining or the number of images remaining to be received to reach the predefined number of images and the recording of the video sequence 20;

[0123] - if the number of images classified since the first image includes a rate of images associated with the same class which is less than the predefined rate, then the time indicator 12 is updated to indicate the failure of the validation, for example by resetting the counter or gauge or by the disappearance of the display of such a counter or gauge.

[0124] The method includes the real-time display of received ultrasound images, the quality indicator 13, and optionally the time indicator 12. Such a display is described below with reference to Figure 2. The image displayed by the AFF display includes the ultrasound image 11 received in real time or taken in real time by the SECH ultrasound probe, the quality indicator 13, and the time indicator 12.

[0125] DOPPLER MEASUREMENT The method includes generating a command for activating one or more Doppler ultrasound modes.

[0126] By "Doppler ultrasound mode" (or "Doppler mode"), we mean the acquisition of ultrasound data to measure the speed and / or direction of blood flow based on the Doppler effect, specifically the change in frequency of sound and ultrasound waves reflected by a moving object (in this case, blood cells or heart walls).

[0127] The various Doppler modes that can be used include continuous wave Doppler, pulsed wave Doppler, color Doppler, power Doppler, or tissue Doppler, or a combination of each of the preceding modes. In one embodiment, the method includes a step of detecting at least one, or at least two, predefined anatomical structures appearing on the received ultrasound image.

[0128] In one embodiment, an "anatomical structure" refers to an identifiable morphological landmark in an ultrasound image, used to precisely define a measurement location, particularly during Doppler acquisition. An anatomical structure may be a point landmark, such as an identifiable junction between a valve leaflet and the ventricle wall, or a linear landmark, such as a segment representing a characteristic anatomical border or tissue interface. These landmarks must be sufficiently stable, localizable, and reproducible to allow for reliable automatic detection and precise spatial localization of the measurement area.

[0129] In one embodiment, organs in their entirety, such as the heart, the fetal heart, or valves considered as complete entities, are not considered anatomical structures within the meaning of the present invention. This distinction ensures that the location of the Doppler measurement is determined from fine and discrete landmarks, and not from anatomical masses whose overall position does not guarantee sufficient accuracy for reliable automatic positioning.

[0130] The step of detecting on the image at least one or at least two predefined anatomical structures can be implemented by any computer-implemented algorithm or process known to those skilled in the art, such as an image segmentation algorithm, a feature detection algorithm, or a shape analysis algorithm.

[0131] The process further includes a LOC determination step of the location of each of at least one or at least two anatomical structures in the received ultrasound image.

[0132] In one execution mode, the detection step and / or the location determination step, on the image, of at least one or at least two predefined anatomical structures can be implemented by any computer-implemented algorithm or process known to those skilled in the art, such as an image segmentation algorithm, a feature detection algorithm, or a shape analysis algorithm.

[0133] In one embodiment, the detection and location determination steps are implemented simultaneously, for example by a single algorithm.

[0134] Preferably, the determination of the location includes the generation of at least one coordinate for each anatomical structure. These coordinates advantageously represent the position of said anatomical structure on the acquired image.

[0135] Preferably, the received ultrasound image 11 represents one of the particular views of the heart described above on which at least one or at least two predefined anatomical structures appear.

[0136] In one embodiment, the localization step of at least one or at least two anatomical structures is implemented when at least one first received ultrasound image is classified into a predetermined class.

[0137] In an alternative embodiment, the localization step of at least one or at least two anatomical structures is implemented when the initial ultrasound image sequence 20 is automatically generated and / or recorded. In an example illustrated in Figure 3, the recording or acquisition of said sequence 20 automatically triggers a trigger instruction 62 for the localization step LOC.

[0138] In an alternative or cumulative embodiment also illustrated in Figure 3, the LOC step of at least one or at least two anatomical structures is implemented by the transmission of an instruction 61 by the user via a user interface of the device. In a preferred execution mode, each anatomical structure is defined by a characteristic point of the predefined organ, such as a junction between two anatomical elements.

[0139] The advantage of determining on an ultrasound image the position on said image of one or at least two anatomical structures, each defined by a point, is to allow the determination of the location and orientation of a Doppler measurement location.

[0140] In a preferred example illustrated in Figure 4, a first anatomical structure 35 is defined by the point of junction between a first leaflet 33 of the mitral valve 31 of the left ventricle 39 of the heart and the wall 36 of said ventricle. Similarly, the second anatomical structure 34 is preferentially defined by the point of junction between a second leaflet 32 ​​of said valve of said ventricle 39 of the heart and the wall 37 of said ventricle.

[0141] One advantage of determining the location on the received ultrasound image 11 of two anatomical structures, each defined by a point, is to allow the determination of a Doppler measurement location more precisely, particularly regarding the orientation of a measurement segment or measurement surface.

[0142] Indeed, this localization of two specific points advantageously leads to a better precision in defining the measurement location.

[0143] In another alternative mode, the at least two characteristic points of the predefined organ define a surface, preferably a surface of an anatomical element such as a wall.

[0144] In a particularly preferred execution mode, the location of at least one or at least two anatomical structures is determined using a learning function trained or configured to receive as input an ultrasound image and generate as output the locations on said image of the various predefined anatomical structures to be detected and located. The learning function preferably comprises a neural network.

[0145] In one example, the learning function was configured using supervised and / or automatic learning, which involved submitting to the classifier a plurality of ultrasound images, each associated with at least one label. The at least one label associated with each training ultrasound image includes the location, on the training ultrasound image, of each predefined anatomical structure.

[0146] In one embodiment, said neural network of the learning function includes a convolutional neural network (better known by the English acronym "CNN" for "Convolutional Neural Network") or more precisely an II-NET.

[0147] Preferably, the learning function is the same as that implemented by the classifier. Thus, each training image includes both at least one label corresponding to a class representing a particular view of the organ and at least one label including the location, on the training ultrasound image, of each predefined anatomical structure.

[0148] The classifier's learning function is configured in this case to classify an ultrasound image into a plurality of classes and to generate the location of each predefined anatomical structure.

[0149] One advantage of such a function is to allow the user to know if the acquired ultrasound image is of sufficient quality and at the same time to determine the Doppler measurement location on said image with a faster calculation speed and adapted to real time.

[0150] DETERMINATION OF MEASUREMENT LOCATION

[0151] In another step, the process includes determining a Doppler measurement location.

[0152] The measurement location may include a measurement point, a measurement segment and / or a Doppler measurement area on said ultrasound image.

[0153] The measurement location is determined from the position or location of each anatomical structure previously determined on the received ultrasound image 11.

[0154] The measurement location is defined by coordinates on the ultrasound image, preferably by coordinates relative to the positions on the ultrasound image of the detected structure or at least two anatomical structures. Alternatively, the measurement location is defined by coordinates based on the positions on the ultrasound image of the detected anatomical structure or at least two anatomical structures.

[0155] In one example, the measurement location is determined based on a point equidistant between the two anatomical structures detected on the ultrasound image.

[0156] One advantage is that it allows the measurement location to be positioned at much more precise anatomical locations on the ultrasound image.

[0157] In the first example illustrated in Figure 5, the measurement location includes a measurement point 51. The relative position of said measurement point 51 is determined by a relative position with respect to the anatomical structure or the two anatomical structures detected 34, 35.

[0158] In a second example illustrated in Figure 6, the measurement location comprises a measurement segment or line 52 whose position and orientation are predefined relative to the position of the two anatomical points. The measurement segment 52 is defined by two endpoints 521 and 522 such that the Doppler measurement allows the blood velocity to be measured between these two endpoints.

[0159] In a third example illustrated in Figure 7, the measurement location includes a Doppler color box 53, that is, a measurement area on the ultrasound image. The size, position, and orientation of this Doppler color box 53 are determined based on the relative positions of the two detected anatomical structures.

[0160] The present invention is particularly effective in the second and third examples, since determining the position of two anatomical structures on the ultrasound image 11 allows for the precise determination of the Doppler measurement location's orientation. A traditional method using only the detection of one organ (e.g., the mitral valve or the ventricle) cannot generate the measurement location's orientation with similar accuracy and requires specialist intervention or additional measurements to correct the position and orientation of the Doppler measurement location.

[0161] In one embodiment, the method includes a step of transmitting TR information based on the Doppler measurement location to the SECH ultrasound probe. This transmission allows the ultrasound probe to generate a Doppler measurement based on the previously determined measurement location.

[0162] This information may include a control signal transmitted to the ultrasound probe, triggering the activation of one or more Doppler ultrasound modes using the measurement location. In this way, the ultrasound probe uses the measurement location to perform ultrasound measurements in Doppler mode.

[0163] In one embodiment, the method includes the acquisition of Doppler ultrasound data. Preferably, this step is implemented by the SECH ultrasound probe using at least one Doppler ultrasound mode. The SECH ultrasound probe acquires Doppler signals according to the transmitted control signal.

[0164] Doppler ultrasound data is acquired from the emission of ultrasonic signals by the ultrasound probe and the reception of these reflected ultrasonic signals. The emitted signals are generated according to the control signal so that the ultrasonic signals are reflected within an area corresponding to the predetermined measurement location.

[0165] Doppler ultrasound data preferably includes Doppler images, a color map, or any type of data from Doppler measurement such as speed and / or direction information.

[0166] In one embodiment, the device is configured to acquire a succession of ultrasound images 11 and a succession of Doppler data intermittently. An advantage is that it allows Doppler data to be acquired over a significant time period while simultaneously redefining the measurement location using the acquired ultrasound images.

[0167] In one embodiment, the device is configured to acquire a second sequence. This second sequence comprises a plurality of second ultrasound images formed from the Doppler ultrasound data. The device is preferably configured to record such a second sequence in device memory and / or to transmit said sequence to a remote device. In one embodiment, the device is configured so that the recording of said sequence 20 automatically generates an alert to the user. In one example, said recording can generate a message inviting the user to initiate the various steps required to generate the TR transmission of the Doppler ultrasound mode activation command. The user can then confirm the invitation via a user interface to generate the simplified instruction 61.

[0168] PROGRAM

[0169] According to one embodiment, the method according to the invention comprises selecting a first class from among the classes representing a particular view of the heart such as one of the views mentioned above.

[0170] Selecting the first class generates the display on the AFF display of a first pre-recorded echocardiographic image 15 of the heart representing such a view. Preferably, the label 17, which may include the name of said particular view, is also displayed on the AFF display. Selecting the first class also generates the display of a first instruction image 14. The first instruction image 14 is preferably pre-recorded. The instruction image illustrates a position and / or orientation instruction for the SECH echocardiographic probe on a patient to obtain said view corresponding to the selected class. The operator is thus advantageously guided and assisted in acquiring such a view. On the same AFF display, they can view an example of the view they must acquire, as well as the position and orientation of the probe to achieve it.Finally, the operator can view in real time whether the view which is acquired is of sufficiently good quality by the quality indicator 13, and can advantageously view the influence of the movements which he makes on the quality of the image acquired thanks to the confidence indicator.

[0171] Finally, once the operator has found a satisfactory quality image, they are assisted in real time by the time indicator 12, which represents the time they must hold the probe to acquire images of sufficient quality. The operator is also assured, by viewing the time indicator 12, that a first sequence 20 will be recorded while maintaining its position. Once the first sequence is recorded or validated, the process may include the automatic selection of a second class from among the classes representing a particular view of the heart, such as one of the views mentioned above. Again, the selection of this second class automatically generates the display of a second pre-recorded reference image and the display of a second pre-recorded ultrasound image illustrating an ultrasound view.

[0172] According to one embodiment, the method further comprises generating and displaying a progress indicator 18. The progress indicator may be representative of the number of sequences that have been recorded. For example, the progress indicator may be representative of the number of classes representing a particular view of the core for which an image sequence 20 has been automatically generated and / or recorded according to the method of the invention.

[0173] According to one execution mode, the process includes displaying the recorded sequence and further includes a second manual validation by the operator after displaying said sequence.

[0174] According to one embodiment, the method includes locating, on the received ultrasound image, at least two predefined anatomical structures and determining the location of each of said anatomical structures.

[0175] The process includes determining a measurement location based on the location of said anatomical structures.

[0176] According to one embodiment, the method comprises transmitting to an ultrasound probe a command to activate at least one Doppler ultrasound mode. This command is generated based on the determined measurement location so as to cause the probe to perform a Doppler ultrasound at the determined measurement location.

[0177] According to one execution mode, the method includes a Doppler ultrasound measurement performed using the ultrasound probe from the previously generated command.

[0178] According to one embodiment, the method comprises receiving at least one Doppler ultrasound data point from the signals acquired by the ultrasound probe during the Doppler ultrasound measurement. DEVICE

[0179] According to one aspect, the device according to the invention includes software and hardware means for implementing the process as described above.

[0180] An embodiment of the device according to the invention is now described with reference to Figure 2.

[0181] The device includes a REC receiver. The REC receiver is intended to be connected to a SECH ultrasound probe so as to continuously and in real time receive ultrasound images 11 acquired by said SECH ultrasound probe.

[0182] The REC receiver can be connected to the SECH ultrasound probe by a wired or wireless connection, for example by a Bluetooth connection or a WI-FI connection or any other data exchange protocol known to those skilled in the art.

[0183] The REC receiver may include or be associated with one or more memory units to temporarily store received images. The REC receiver is directly or indirectly connected to the AFF display to transmit the acquired ultrasound images to the AFF display.

[0184] The device further includes means for implementing a CLASS classifier as described above. The classifier is configured to receive ultrasound images 11, associate a class 14 with the ultrasound images in real time, and generate the quality indicator 13 in real time. The classifier is directly or indirectly connected to the AFF display to transmit the generated indicators to the AFF display.

[0185] Alternatively, the classifier is implemented by remote electronic equipment, such as a remote server. In this scenario, the device includes an interface for exchanging data with the remote equipment in order to transmit data and retrieve the results of the processed, i.e., classified, data.

[0186] Finally, in the present invention, it is understood that when classification is implemented wholly or partly by remote equipment, the device of the invention can be interpreted as the system comprising, on the one hand, the local device described in this application and, on the other hand, the remote means for implementing the classification function. The device further comprises at least one processor or CALC computer associated with memory to execute at least some of the steps of the process according to the invention. For example, a first processor can be configured to execute the steps of classification 200, generation and display of the quality indicator 300 and the time indicator 500, and / or automatic recording 300 of the image sequence 20.For example, a second processor can be configured to perform the steps for generating and transmitting an activation command 63 and receiving at least one Doppler ultrasound data point. In one embodiment, the first and second processors are the same processor. This processor can be connected to the AFF display to transmit the time indicator 12 to the AFF display. In one embodiment, the at least one processor or computer includes means for transmitting and receiving information with a remote device, for example via an internet network, enabling the implementation of the steps of the method according to the invention.

[0187] In some cases, the processor or computer can communicate with one or more external devices over the network. The processor or computer can be connected to the network via a wired connection (e.g., via an Ethernet cable) and / or a wireless connection (e.g., via a Wi-Fi network). These external devices can include servers, workstations, and / or databases. The processor or computer can communicate with these devices to, for example, offload computationally intensive tasks. For instance, the processor or computer can send an ultrasound image over the network to the server for analysis (e.g., to identify or classify an anatomical feature in the ultrasound, to detect or determine the position of one or more anatomical structures, or to determine the measurement location) and receive the analysis results from the server.In addition (or alternatively), the processor or computer can communicate with these devices to access information that is not available locally and / or update a central information repository.

[0188] Device 1 may also include a plurality of processors, each associated with one or more memories, and configured to perform such steps together. In one embodiment, the processor(s) may be remote and connected to the display via a data network.

[0189] The device further includes one or more memories for storing or recording the sequences generated by the method according to the invention and / or for storing the computer programs that, when executed by one or more processors, implement the method according to the invention. In one embodiment, the device further includes an EMM transmitter connected to said MEM memory for transmitting said sequences recorded on said MEM memory to a data network. In another embodiment, the device further includes a second EMM2 transmitter connected to the SECH ultrasound probe for transmitting said activation command 63 to said probe.

[0190] The device may also include means of communication such as transmitters and receivers for exchanging information with the probe and / or a remote device.

[0191] The AFF display may include means for receiving the various information 11, 13, 21, 12, 63, 64 received by the various means REC, CLASS, PROC of the device to generate the final image to be displayed.

Claims

DEMANDS 1. A device (1) for assisting in the acquisition of ultrasound data, configured to receive in real time ultrasound images (11) acquired by an ultrasound probe (SECH) connected to said device (1), and comprising: ■ a first computer, associated with a first memory, to implement a classifier (CLASS) configured to classify in real time the first ultrasound images received (11) and associate each image with a class (14) to generate a quality indicator (13) according to said class (14) associated; ■ a second calculator (CALC2), associated with a second memory, and configured to execute the following steps: ■ when a sequence of a predefined number of received images includes a rate of images associated with the same class greater than a predefined rate, the determination of the location of at least one predefined anatomical structure (35, 34) detected in a received ultrasound image; ■ the determination (DET2) of a measurement location (51, 52, 53) as a function of the determined location of said anatomical structure (34, 35) detected; ■ the transmission (TR) to the ultrasound probe (SECH) connected to said device (1) of an activation command (63) of one or more Doppler ultrasound modes using the determined measurement location (51, 52, 53). ■ the reception (RC) of at least one Doppler ultrasound data (64) produced by the probe using one or more Doppler ultrasound modes.

2. Device according to claim 1 in which at least one anatomical structure includes an identifiable point or linear landmark such as a tissue interface or an anatomical edge characteristic of an organ.

3. Device according to claim 1 or claim 2, wherein at least one predefined anatomical structure comprises a first anatomical structure and a second anatomical structure different from the first anatomical structure.

4. Device according to any one of claims 1 to 3, wherein the first anatomical structure is defined by the junction between a first leaflet of the valve of a ventricle of the heart and the wall of said ventricle and the second anatomical structure is defined by the junction between a second leaflet of said valve and the wall of said ventricle of the heart.

5. Device according to any one of claims 1 to 4, further comprising a display (AFF) for, in real time, displaying the first ultrasound images, at least one Doppler ultrasound data (64) and / or the last generated quality indicator.

6. Device according to any one of the preceding claims, characterized in that the determined measurement location comprises a measurement point (51), a measurement line (52) or a measurement surface (53) on a first received ultrasound image.

7. Device according to any one of claims 1 to 6, further comprising a third computer (CALC3), associated with a third memory, and configured to automatically perform the recording of said image sequence (20) in a fourth memory (MEM) when a sequence of a predefined number of received images includes a rate of images associated with the same class greater than a predefined rate.

8. Device according to claim 7, wherein the second computer (CALC2) is configured to implement the step of detecting at least two anatomical structures appearing in the received ultrasound image once the third computer has executed the step of automatically recording said image sequence.

9. Device according to claim 1, wherein the classifier (CLASS) is configured to classify an image among: ■ a plurality of classes, each representing a particular view of the organ; and ■ at least one class representing a view of insufficient quality.

10. Device according to claim 9, wherein the classifier (CLASS) is implemented by means of a learning function configured from supervised machine learning.

11. Device according to any one of the preceding claims, characterized in that the at least one Doppler ultrasound data point (64) comprises at least one second ultrasound image.

12. System (2) comprising a device (1) according to any one of claims 1 to 11 and an ultrasound probe (SECH) capable of being connected to said device (1) so as to transmit the ultrasound images captured by the probe to the device.

13. A computer program product comprising instructions that lead a device to perform the following steps: ■ the reception of first ultrasound images (11); ■ real-time classification of the first ultrasound images received (11) and association of each image with a class (14) to generate a quality indicator (13) according to said class (14) associated; ■ when a sequence of a predefined number of received images includes a rate of images associated with the same class greater than a predefined rate, the detection (DET) of at least one predefined anatomical structure appearing in a received ultrasound image; ■ Localization (LOC) to determine the location of at least one anatomical structure detected in the received ultrasound image; ■ the determination (DET2) of a measurement location relative to the determined location of said at least one detected anatomical structure; ■ the transmission (TR) to an ultrasound probe (SECH) connected to said device (1) of an activation command for one or more Doppler ultrasound modes using the determined measurement location; ■ the reception (RC) of at least one Doppler ultrasound data (64) produced by the probe using one or more Doppler ultrasound modes.

14. Computer-readable medium on which the computer program according to claim 13 is recorded.

Images