System and method for assessing placenta

Abstract

This invention provides a method for performing a placental assessment. The method includes obtaining a 3D ultrasound image of the uterus (210) and segmenting the placenta (220). A 3D map (200) of the uterus is then generated, wherein the generation includes: identifying the location of the placenta within the uterus relative to anatomical structures such as the cervix (250); obtaining anatomical reference data regarding potential risks associated with the location of the placenta within the uterus; and comparing the location of the placenta with the anatomical reference data. Generating the 3D map of the uterus includes a 3D map of the placenta marked with indicators that change based on the comparison of the placenta's location with the anatomical reference data. The appearance of the indicators can be changed according to, for example, risk type / severity.

Description

Technical Field

[0001] This invention relates to the field of uterine imaging, and more particularly to the field of placental assessment based on uterine images. Background Technology

[0002] As the organ that provides blood, oxygen, and nutrients to the fetus, the health of the placenta is crucial during pregnancy. Placental abnormalities can be benign or require the implementation of specific pregnancy management techniques. For example, the location of the placenta within the uterus can be highly variable. Furthermore, scarring resulting from an increased number of cesarean deliveries is a common source of placental abnormalities in subsequent pregnancies.

[0003] In some cases, placental assessment can reveal serious medical conditions, such as pathological adherent placenta, which requires a specific delivery plan to ensure the survival of both mother and child.

[0004] Typically, due to its large size, the placenta is difficult to observe as a whole, especially after the first trimester of pregnancy. Furthermore, ultrasound physicians require specific skills to accurately locate and assess placental abnormalities.

[0005] The use of three-dimensional ultrasound for placental assessment was proposed in the journal article “Three-dimensional ultrasound evaluation of the placenta” by T. Hata et al. (PLACENTA, Vol. 32, No. 2, pp. 105-115, 2010-11-03 (XP028360511)).

[0006] Therefore, a means of accurately assessing the placenta within uterine images is needed. Summary of the Invention

[0007] This invention is defined by the claims.

[0008] According to an example of one aspect of the present invention, a method for performing an assessment of the placenta is provided, the method comprising:

[0009] Obtain a 3D ultrasound image of the uterus, which contains the placenta;

[0010] The placenta is segmented from the 3D ultrasound image of the uterus using an image segmentation algorithm; and

[0011] A 3D model of the uterus is generated based on the results of the segmentation of the placenta and the 3D ultrasound image of the uterus, wherein generating the 3D model of the uterus includes:

[0012] Identify the location of the placenta within the uterus;

[0013] Obtain anatomical reference data regarding potential risks associated with the location of the placenta within the uterus;

[0014] The location of the placenta is compared with the anatomical reference data;

[0015] A 3D model of the uterus is generated based on the 3D ultrasound image of the uterus, the 3D model of the uterus including a 3D model of the placenta; and

[0016] The 3D rendering of the placenta is marked using indicators, wherein the indicators are changed based on a comparison of the placenta's location with the anatomical reference data.

[0017] This method provides a means of assessing the entire uterus and placenta, which are often too large to fit within a single ultrasound image (especially after the first trimester of pregnancy).

[0018] By generating a complete 3D map of the uterus based on image data and placental segmentation, it is possible to present a visual representation of the scanned area to the user, with automatically determined indicators to emphasize aspects of the placenta based on its location within the uterus.

[0019] In this way, placental assessment can be simplified and possible considerations can be determined based on the location of the placenta within the uterus.

[0020] The position of the placenta within the uterus is preferably a relative position, that is, the position of the placenta within the uterus relative to the anatomical structures in the 3D ultrasound image of the uterus.

[0021] In some embodiments, the method includes segmenting the anatomical structures within the 3D ultrasound image of the uterus. Alternatively, in other embodiments, the method includes using feature detection algorithms and / or classification algorithms to identify the anatomical structures.

[0022] In an embodiment, the method includes:

[0023] Obtain multiple 3D ultrasound images; and

[0024] The multiple 3D ultrasound images are combined to form the 3D ultrasound image of the uterus.

[0025] In this way, the method provides a means of assessing the uterus when it is too large to be fit within a single field of view.

[0026] In an embodiment, the method further includes segmenting additional anatomical structures within the 3D ultrasound image of the uterus, and wherein generating the 3D rendering of the uterus further includes:

[0027] Identify the position of the placenta relative to the additional anatomical structures;

[0028] To obtain additional anatomical reference data regarding potential risks associated with the position of the placenta relative to the additional anatomical structures;

[0029] The location of the placenta is compared with the additional anatomical reference data;

[0030] Additional indicators are used to mark one or more of the 3D rendering of the placenta and the 3D rendering of the uterus, wherein the additional indicators are changed based on the comparison of the placenta's location with the additional anatomical reference data.

[0031] In this way, the interaction between the placenta and another physical structure near the uterus can be determined and highlighted to the user in a 3D rendering of the uterus.

[0032] In yet another embodiment, the anatomical structure and / or the additional anatomical structure includes one or more of the following:

[0033] bladder;

[0034] Umbilical cord;

[0035] Cervix;

[0036] Cervical region; and

[0037] Anatomical structures near the uterus.

[0038] In an embodiment, the method further includes determining the physical properties of the placenta based on the result of the placenta's segmentation.

[0039] In this way, it is possible to perform additional assessments of the placenta and uterus based on various physical aspects of the placenta.

[0040] In yet another embodiment, the physical property includes one or more of the following:

[0041] Placental volume; and

[0042] Placental texture.

[0043] In one embodiment, the method further includes detecting anatomical features within the placenta, and wherein generating the 3D rendering of the uterus further includes using feature indicators to mark the detected anatomical features.

[0044] Anatomical features can be geometric features. In some embodiments, anatomical features can indicate placental blood pools.

[0045] In an embodiment, the detection of the anatomical features of the placenta is performed based on the results of the segmentation of the placenta and the 3D ultrasound image of the uterus.

[0046] In yet another embodiment, detecting the anatomical features preferably includes:

[0047] Determine the thickness of the uterine wall near the placenta; and

[0048] The anatomical features are identified based on the thickness of the uterine wall.

[0049] In an embodiment, the detection of the anatomical features is preferably performed using a machine learning algorithm.

[0050] In this way, additional physical measurements of the placenta can be automatically obtained from the image data and highlighted to the user in a 3D rendering of the uterus. Based on these additional physical measurements, the user can then assess the health of the placenta.

[0051] In one embodiment, the method further includes detecting placental blood pools within the placenta, and wherein generating the 3D rendering of the uterus further includes using blood pool indicators to mark the placental blood pools.

[0052] In an embodiment, the method further includes detecting placental implantation type based on the result of the segmentation of the placenta and the 3D ultrasound image of the uterus, wherein generating the 3D drawing of the uterus further includes using an implantation type indicator to mark the 3D drawing of the placenta.

[0053] In this way, additional measures of placental health can be automatically obtained from the image data and highlighted to the user in a 3D rendering of the uterus.

[0054] In yet another embodiment, detecting the placental implantation type includes:

[0055] Determine the thickness of the uterine wall near the placenta; and

[0056] The type of placenta accreta is identified based on the thickness of the uterine wall.

[0057] In this embodiment, the detection of the placental implantation type is performed using a machine learning algorithm.

[0058] In one embodiment, the method further includes performing a Doppler ultrasound-based vascular assessment of the placenta.

[0059] In this way, additional measures of placental health can be automatically obtained from the image data and highlighted to the user in a 3D rendering of the uterus.

[0060] According to an example of one aspect of the invention, a computer program including a computer program code module is provided, wherein when the computer program is run on a computer, the computer program code module is adapted to implement the method described above.

[0061] According to an example of one aspect of the present invention, an ultrasound system for evaluating the placenta is provided, the ultrasound system comprising:

[0062] An ultrasound probe suitable for acquiring 3D ultrasound images of the uterus and placenta;

[0063] Processor, which is suitable for:

[0064] The placenta is segmented from the 3D ultrasound image of the uterus using an image segmentation algorithm; and

[0065] A 3D model of the uterus is generated based on the results of the segmentation of the placenta and the 3D ultrasound image of the uterus, wherein generating the 3D model of the uterus includes:

[0066] Identify the location of the placenta within the uterus;

[0067] Obtain anatomical reference data regarding potential risks associated with the location of the placenta within the uterus;

[0068] The location of the placenta is compared with the anatomical reference data;

[0069] A 3D model of the uterus is generated based on the 3D ultrasound image of the uterus, the 3D model of the uterus including a 3D model of the placenta; and

[0070] The 3D rendering of the placenta is marked using indicators, wherein the indicators change based on a comparison of the placenta's location with anatomical reference data; and

[0071] Display unit adapted to display the 3D rendering of the uterus to the user.

[0072] The position of the placenta within the uterus is preferably a relative position, that is, the position of the placenta within the uterus relative to the anatomical structures in the 3D ultrasound image of the uterus.

[0073] In one embodiment, the ultrasound system further includes a user interface adapted to receive user input indicating a region of interest for further study of the 3D rendering of the uterus, and wherein the processor is adapted to:

[0074] A set of acquisition settings is determined based on the indicated region of interest; and

[0075] Apply the set of acquisition settings to the ultrasound system.

[0076] In this way, users can easily indicate areas that require further investigation, so that the system is ready to perform additional data collection using optimal settings.

[0077] In one embodiment, additional 3D ultrasound images of the uterus are acquired, and the processor is further adapted to update the 3D rendering of the uterus based on the additional 3D ultrasound images of the uterus.

[0078] In this way, any further research performed by the user can be automatically added to the 3D mapping of the uterus.

[0079] These and other aspects of the invention will become apparent from the embodiments described below and will be set forth with reference to the embodiments described below. Attached Figure Description

[0080] Examples of the invention will now be described in detail with reference to the accompanying drawings, in which:

[0081] Figure 1 An ultrasound diagnostic imaging system is shown to explain its overall operation;

[0082] Figure 2 The method of the present invention is shown; and

[0083] Figure 3 An example of a 3D drawing of the uterus is shown. Detailed Implementation

[0084] The invention will be described with reference to the accompanying drawings.

[0085] It should be understood that the detailed descriptions and specific examples, while indicating exemplary embodiments of the apparatus, system, and method, are intended for illustrative purposes only and are not intended to limit the scope of the invention. These and other features, aspects, and advantages of the apparatus, system, and method of the invention will be better understood from the following description, the appended claims, and the accompanying drawings. It should be understood that the drawings are merely schematic and not drawn to scale. It should also be understood that the same reference numerals are used throughout the drawings to indicate the same or similar parts.

[0086] This invention provides a method for performing placental assessment. The method includes obtaining a 3D ultrasound image of the uterus containing the placenta, and segmenting the placenta from the 3D ultrasound image of the uterus using an image segmentation algorithm.

[0087] Then, a 3D map of the uterus is generated based on the results of the placental segmentation and the 3D ultrasound image of the uterus. Generating the 3D map of the uterus includes: identifying the location of the placenta within the uterus; obtaining anatomical reference data regarding potential risks associated with the location of the placenta within the uterus; and comparing the location of the placenta with the anatomical reference data. The 3D map of the uterus is generated based on the 3D ultrasound image of the uterus, including a 3D map of the placenta, and the 3D map of the placenta is labeled with an indicator, wherein the indicator changes based on the comparison of the location of the placenta with the anatomical reference data.

[0088] First, refer to Figure 1 Describe the general operation of an exemplary ultrasound system.

[0089] The system includes an array transducer probe 4 having a transducer array 6 for emitting ultrasonic waves and receiving echo information. The transducer array 6 may include a CMUT transducer; a piezoelectric transducer formed of a material such as PZT or PVDF; or any other suitable transducer technology. In this example, the transducer array 6 is a two-dimensional array of transducers 8 capable of scanning a 2D plane or a three-dimensional volume of a region of interest. In another example, the transducer array may be a 1D array.

[0090] Transducer array 6 is coupled to microwave beamformer 12, which controls the reception of signals by the transducer elements. The microwave beamformer is capable of performing at least partial beamforming on signals received by subarrays (often referred to as “groups” or “pieces”) of transducers, as described in U.S. Patents US 5,997,479 (Savord et al.), US 6,013,032 (Savord), and US 6,623,432 (Powers et al.).

[0091] It should be noted that the microwave beamformer is entirely optional. Furthermore, the system includes a transmit / receive (T / R) switch 16 to which the microwave beamformer 12 can be coupled, and this switch toggles the array between transmit and receive modes, and protects the main beamformer 20 from high-energy transmitted signals when the microwave beamformer is not used and the transducer array is directly operated by the main system beamformer. Transmission of the ultrasonic beam from the transducer array 6 is guided by a transducer controller 18 coupled to the microwave beamformer via the T / R switch 16 and the main transmit beamformer (not shown), which can receive input from user-to-user interface or control panel 38. The controller 18 may include transmit circuitry arranged to drive the transducer elements of array 6 during transmit mode (directly or via the microwave beamformer).

[0092] In a typical line-by-line imaging sequence, the beamforming system within the probe can operate as follows. During transmission, a beamformer (which, depending on the implementation, can be a microwave beamformer or a main system beamformer) activates the transducer array or sub-apertures of the transducer array. A sub-aperture can be a one-dimensional line of transducers within a larger array or a two-dimensional patch of transducers. In transmission mode, the focusing and manipulation of the ultrasonic beam generated by the array or its sub-apertures are controlled, as described below.

[0093] After receiving the backscattered echo signal from the object, the received signal undergoes receive beamforming (as described below) to align the received signal, and in the case of a sub-aperture, the sub-aperture is then shifted, for example, by a transducer element. The shifted sub-aperture is then activated, and this process is repeated until all transducer elements of the transducer array have been activated.

[0094] For each line (or sub-aperture), the total received signal of the associated lines used to form the final ultrasound image will be the sum of the voltage signals measured by the transducer elements of a given sub-aperture during the reception period. Following the beamforming process below, the resulting line signals are typically referred to as radio frequency (RF) data. Each line signal (RF dataset) generated from the individual sub-apertures then undergoes additional processing to generate the line of the final ultrasound image. The amplitude variation of the line signal over time will contribute to the brightness variation of the ultrasound image with depth, where high-amplitude peaks will correspond to bright pixels (or sets of pixels) in the final image. Peaks appearing near the beginning of the line signal will represent echoes from shallow structures, while peaks gradually appearing later in the line signal will represent echoes from structures at increasing depths within the object.

[0095] One of the functions controlled by the transducer controller 18 is the direction of beam manipulation and focusing. The beam can be manipulated to be straight ahead (orthogonal to) the transducer array, or at different angles for a wider field of view. The manipulation and focusing of the transmitted beam can be controlled based on the actuation time of the transducer elements.

[0096] In general ultrasound data acquisition, two methods can be distinguished: plane wave imaging and beamforming imaging. The two methods are distinguished by the presence of beamforming in the transmit mode (beamforming imaging) and / or receive mode (plane wave imaging and beamforming imaging).

[0097] First, let's look at the focusing function. By simultaneously activating all transducer elements, the transducer array generates a plane wave that diverges as it travels past an object. In this case, the ultrasound beam remains unfocused. By introducing a position-dependent time delay into the transducer activation, the wavefront of the beam can be focused at a desired point, called the focal zone. The focal zone is defined as a point where the lateral beamwidth is less than half the width of the emitted beam. In this way, the lateral resolution of the final ultrasound image is improved.

[0098] For example, if the time delay causes the transducer elements to activate sequentially, starting from the outermost element and ending at one or more central elements of the transducer array, a focal zone will be formed at a given distance from the probe, coinciding with one or more central elements. The distance of the focal zone from the probe will vary depending on the time delay between each subsequent round of transducer element activation. After the beam passes through the focal zone, it will begin to diverge, thus forming a far-field imaging region. It should be noted that for focal zones positioned close to the transducer array, the ultrasonic beam will diverge rapidly in the far field, resulting in beamwidth artifacts in the final image. Typically, the near field between the transducer array and the focal zone shows little detail due to large overlap in the ultrasonic beam. Therefore, changing the position of the focal zone will result in a significant change in the quality of the final image.

[0099] It should be noted that in emission mode, unless the ultrasound image is divided into multiple focal zones (each of which may have a different emission focus), only one focus can be defined.

[0100] Furthermore, after receiving the echo signal from within the object, the reverse process described above can be executed to perform receiver focusing. In other words, the incoming signal can be received by the transducer elements and undergoes an electronic time delay before being transmitted to the system for signal processing. The simplest example of this is called delay-sum beamforming. The receiver focusing of the transducer array can be dynamically adjusted according to time.

[0101] Now consider the function of beam manipulation. By correctly applying a time delay to the transducer elements, a desired angle can be imparted to the ultrasonic beam as it leaves the transducer array. For example, by activating the transducers on the first side of the transducer array in a sequence ending on opposite sides of the array, followed by the remaining transducers, the wavefront of the beam will be angled toward the second side. The magnitude of the manipulation angle relative to the normal of the transducer array depends on the magnitude of the time delay between the subsequent activation of the transducer elements.

[0102] Furthermore, it is possible to focus and manipulate the beam, where the total time delay applied to each transducer element is the sum of the focusing and manipulation time delays. In this case, the transducer array is called a phased array.

[0103] In cases where a CMUT transducer requires a DC bias voltage for activation, the transducer controller 18 can be coupled to control a DC bias control 45 for the transducer array. The DC bias control 45 sets one or more DC bias voltages applied to the CMUT transducer elements.

[0104] For each transducer element in the transducer array, an analog ultrasonic signal, typically referred to as channel data, enters the system through a receiving channel. In the receiving channel, a partial beamforming signal is generated by a microwave beamformer 12 based on the channel data and then passed to a main receiving beamformer 20, where partial beamforming signals from individual transducer segments are combined into a complete beamforming signal, referred to as radio frequency (RF) data. Beamforming performed at each stage can be performed as described above, or may include additional functionality. For example, the main beamformer 20 may have 128 channels, each receiving partial beamforming signals from segments of dozens or hundreds of transducer elements. In this way, signals received from thousands of transducers in the transducer array can be effectively contributed to a single beamforming signal.

[0105] The beamforming received signal is coupled to signal processor 22. Signal processor 22 is capable of processing the received echo signal in various ways, such as bandpass filtering; decimation; I and Q component separation; and harmonic signal separation, which is used to separate linear and nonlinear signals, thereby enabling the identification of nonlinear (higher harmonics of the fundamental frequency) echo signals returning from tissue and microbubbles. Signal processor 22 can also perform additional signal enhancement, such as speckle reduction, signal synthesis, and noise cancellation. The bandpass filter in the signal processor can be a tracking filter whose passband slides from higher frequency bands to lower frequency bands as the echo signal is received from increasing depths, thereby rejecting noise from higher frequencies at greater depths, which typically lack anatomical information.

[0106] Beamformers for transmitting and receiving are implemented in different hardware and can have different functions. Of course, the receiver beamformer is designed to take into account the characteristics of the transmitting beamformer. For simplicity, in... Figure 1 Only receiver beamformers 12 and 20 are shown in the diagram. Throughout the system, there will also be a transmitter chain with both a transmitter microwave beamformer and a main transmitter beamformer.

[0107] The function of microwave beamformer 12 is to provide an initial combination of signals in order to reduce the number of analog signal paths. This is typically performed in the analog domain.

[0108] The final beamforming is performed in the main beamformer 20, and is typically done after digitization.

[0109] The transmit and receive channels use the same transducer array 6 with a fixed frequency band. However, the bandwidth occupied by the transmit pulse can vary depending on the transmit beamforming used. The receive channel can capture the entire transducer bandwidth (which is the classical method), or it can use bandpass processing, which extracts only the bandwidth containing the desired information (e.g., harmonics of the main harmonic).

[0110] The RF signal can then be coupled to a B-mode (i.e., brightness mode or 2D imaging mode) processor 26 and a Doppler processor 28. The B-mode processor 26 performs amplitude detection on the received ultrasound signal to image structures in the body, such as organs, tissues, and blood vessels. In the case of line-by-line imaging, each line (beam) is represented by an associated RF signal whose amplitude is used to generate a brightness value to be assigned to a pixel in the B-mode image. The exact location of a pixel within the image is determined by the position of the associated amplitude measurement along the RF signal and the number of lines (beams) of the RF signal. B-mode images of such structures can be formed in harmonic or fundamental image modes or a combination of both, as described in U.S. Patent 6,283,919 (Roundhill et al.) and U.S. Patent 6,458,083 (Jago et al.). The Doppler processor 28 processes temporally different signals generated by tissue movement and blood flow to detect moving material, such as the flow of blood cells in the image field. Doppler processor 28 typically includes a wall filter having parameters set to allow or reject echoes returning from a selected type of material within the body.

[0111] The structural and motion signals generated by the B-mode and Doppler processors are coupled to a scan converter 32 and a multiplane reformer 44. The scan converter 32 arranges the echo signals in a spatial relationship, receiving them in a desired image format according to this spatial relationship. In other words, the scan converter is used to transform RF data from a cylindrical coordinate system to a Cartesian coordinate system suitable for displaying ultrasound images on an image display 40. In the case of B-mode imaging, the brightness of a pixel at a given coordinate is proportional to the amplitude of the RF signal received from that location. For example, the scan converter can arrange the echo signals into a two-dimensional (2D) fan-shaped format or a pyramidal three-dimensional (3D) image. The scan converter can superimpose colors corresponding to motion at points in the image field onto the B-mode structural image, where Doppler-estimated velocity produces the given color. The combined B-mode structural image and color Doppler image depict the motion of tissue and blood flow within the structural image field. As described in U.S. Patent US 6443896 (Detmer), the multiplane reformer converts echoes received from points in a common plane within a volumetric region of the body into an ultrasound image of that plane. Volume plotter 42 converts the echo signal of the 3D dataset into a projected 3D image, as seen from a given reference point, as described in U.S. Patent US 6530885 (Entrekin et al.).

[0112] 2D or 3D images are coupled from the scan converter 32, multiplane reformer 44, and volume renderer 42 to the image processor 30 for further enhancement, buffering, and temporary storage for display on the image display 40. The image processor may be adapted to remove certain imaging artifacts from the final ultrasound image, such as: acoustic shadowing caused by strong attenuators or refraction; post-enhancement caused by weak attenuators; reverberation artifacts, for example, where highly reflective tissue interfaces are located adjacent to each other; and so on. Additionally, the image processor may be adapted to perform certain speckle reduction functions to improve the contrast of the final ultrasound image.

[0113] In addition to their use in imaging, the blood flow values ​​generated by the Doppler processor 28 and the tissue structure information generated by the B-mode processor 26 are also coupled to the quantization processor 34. This quantization processor generates measurements of different flow conditions, such as the volumetric rate of blood flow in addition to structural measurements like organ size and gestational age. The quantization processor can receive input from the user control panel 38, such as the points to be measured in the anatomical structures of the image.

[0114] Output data from the quantization processor is coupled to a graphics processor 36, which is used to reproduce measurement graphs and values ​​along with images on a display 40, and for audio output from the display device 40. The graphics processor 36 can also generate graphic overlays for display alongside ultrasound images. These overlays may contain standard identification information such as patient name, date and time of the image, imaging parameters, etc. For these purposes, the graphics processor receives input, such as patient name, from a user interface 38. The user interface is also coupled to a transmission controller 18 to control the generation of ultrasound signals from the transducer array 6, and thus the images produced by the transducer array and the ultrasound system. The transmission control function of the controller 18 is only one of the functions performed. The controller 18 also considers the operating mode (given by the user) and the corresponding desired transmitter and bandpass configurations in the receiver analog-to-digital converter. The controller 18 may be a state machine with fixed states.

[0115] The user interface is also coupled to a multiplane reformer 44 for selecting and controlling planes of multiple multiplane reformatted (MPR) images, which can be used to perform quantization measurements in the image field of the MPR image.

[0116] Figure 2 A method 100 for performing placental assessment according to one aspect of the present invention is shown.

[0117] The method begins in step 110 with obtaining a 3D ultrasound image of the uterus, which contains the placenta.

[0118] In the early stages of pregnancy, it is possible to acquire the entire uterus and placenta in a single scan using a 3D ultrasound probe. Later, as the pregnancy progresses to more advanced stages, the uterus may not be suitable for a single ultrasound acquisition volume; however, it is possible to use a series of multiple volume acquisitions to cover the entire area of ​​the uterus.

[0119] In other words, multiple 3D ultrasound images can be obtained to assess a larger area. In this case, the method may include step 120, wherein the multiple 3D ultrasound images are combined to form a 3D ultrasound image of the uterus.

[0120] Using multiple 3D ultrasound images, and in some cases combined with spatial ultrasound probe position tracking obtained through electromagnetic tracking, volumetric fusion algorithms can be used to reconstruct a 3D image of the entire uterus. Such reconstruction is possible even if the fetus is moving during the acquisition of multiple 3D ultrasound images, as the uterus and placenta will remain stable.

[0121] In step 130, the placenta is segmented from the 3D ultrasound image of the uterus using an image segmentation algorithm. Any suitable image segmentation algorithm can be used.

[0122] In the context of this invention, segmenting an object or structure preferably refers to finding the outline or boundary of the object or structure in a 3D volume. Therefore, the expression "segmenting the placenta" preferably means finding the outline or boundary of the placenta in a representation of the placenta (such as, in particular, an image, a mask (e.g., a binary mask), or a mesh).

[0123] For example, once multiple 3D ultrasound images of the uterus have been combined, the placenta is segmented from the 3D ultrasound images of the uterus. Placental segmentation can be performed, for example, using 3D deep learning segmentation algorithms (such as the U-net architecture described in O. Ronneberger et al., U-Net: Convolutional Networks for Biomedical Image Segmentation (Proceedings of MICCAI'15, Vol. 9351, pp. 234-241 (2015))).

[0124] In step 140, a 3D model of the uterus is generated based on the results of placental segmentation and a 3D ultrasound image of the uterus. Generating the 3D model of the uterus includes the following steps.

[0125] In step 150, the location of the placenta within the uterus is identified. The location of the placenta within the uterus can be identified based on the results of placental segmentation.

[0126] In addition, the position of the placenta relative to the cervix is ​​identified, and the cervix can be detected within a 3D ultrasound image of the uterus (e.g., through shape recognition algorithms or machine learning algorithms). The position of the cervix within the uterus and the position of the placenta relative to the cervix can then be compared with anatomical reference data in the following steps.

[0127] In step 160, anatomical reference data relating to the potential risks associated with the placenta's location within the uterus is obtained. This anatomical reference data can be obtained from the ultrasound system's local storage or from remote storage, for example, maintained on a server.

[0128] In step 170, the location of the placenta is compared with anatomical reference data in order to determine the condition of the placenta and uterus based on the reference data.

[0129] In step 180, a 3D map of the uterus is generated based on a 3D ultrasound image of the uterus, the 3D map of the uterus including a 3D map of the placenta, and in step 190, the 3D map of the placenta is marked with indicators, wherein the indicators are changed based on a comparison of the placenta's location and anatomical reference data.

[0130] In other words, a 3D rendering engine is used to generate a global 3D visualization of the placenta within the uterus. On this global 3D visualization, regions of interest for clinicians, already segmented in previous steps, are marked using indicators, such as highlighting using color or specific lighting effects. These indicators can include, for example, color overlays on regions of interest from a 3D rendering of the uterus or a composite structure rendering (such as a mesh on a 3D rendering of the uterus based on segmentation results).

[0131] Indicators can be any type of indicator suitable for conveying information to a user. For example, an indicator can utilize various colors with associated keys to convey information to a user. Alternatively, or in addition to various colors, an indicator can have variable brightness, saturation, or patterns to convey the desired information.

[0132] Figure 3 An example of a 3D drawing of the uterus 210 is shown.

[0133] exist Figure 3 In the example shown, it has already been compared with the above regarding Figure 2 A similar method described herein generates a 3D map, wherein the method further includes segmenting one or more additional anatomical structures within a 3D ultrasound image of the uterus. In this case, generating the 3D map of the uterus also includes identifying the position of the placenta 220 relative to one or more additional anatomical structures. Again, anatomical reference data regarding potential risks associated with the position of the placenta relative to the additional anatomical structures and the position of the placenta are compared with anatomical reference data. Additional indicators can then be used to label the 3D map of the placenta and / or the 3D map of the uterus, wherein the additional indicators change based on comparisons of the placental position and additional anatomical reference data.

[0134] Looking Figure 3 Additional anatomical structures include: bladder 230; umbilical cord 240; and cervix 250. In other words, the bladder, umbilical cord, and cervix are segmented from the 3D ultrasound image of the uterus and mapped as part of the 3D mapping 200.

[0135] Using the obtained anatomical reference data, the location of additional anatomical structures within the uterus 210 and relative to the placenta 220 can be assessed to evaluate potential health risks or conditions associated with the uterine layout. In response to this assessment, important areas can be marked in a manner similar to the markings used in the 3D mapping of the placenta as described above.

[0136] exist Figure 3In the example shown, the 3D rendering 200 of the uterus 210 includes three regions of interest (ROIs) based on additional anatomical structures segmented from a 3D ultrasound image of the uterus. Specifically, there are a bladder / uterus interface 260, an umbilical cord insertion area 270, and a cervical region 280. Each of the ROIs can be marked using an individual indicator, the appearance of which varies according to, for example, region type or risk type / severity.

[0137] Additional anatomical structures can be segmented using methods similar to those described above for placental segmentation.

[0138] Examples of risk assessments for each additional anatomical structure may include: for the cervical region 280, if the placenta 210 is close to or covers the cervix 250 (a condition known as placenta previa), specific pregnancy management (which may include cesarean delivery) may be identified as necessary. In the case of the umbilical cord insertion area, if the umbilical cord is inserted close to the edge of the placenta, specific vascular conditions may have occurred and require attention during or after pregnancy.

[0139] In cases involving the bladder / uterus interface, if the placenta is covering the area, specific Doppler examinations can be performed to assess the vascular distribution around the bladder. Increased vascular distribution can be an indicator of possible myometrial invasion and pathologically apposed placenta, which requires specific pregnancy management and delivery planning.

[0140] In addition, 3D mapping of the uterus can be used for anomaly detection and quantification purposes.

[0141] For example, 3D mapping of the uterus, and especially the placenta, can be used to assess placenta accreta, which involves the area of ​​the uterus where the placenta is located.

[0142] In other words, the above method may also include detecting placental implantation type based on the results of placental segmentation and 3D ultrasound images of the uterus, and wherein generating a 3D map of the uterus also includes marking the 3D map of the placenta with an implantation type indicator.

[0143] In other words, 3D visualization can reveal suspicious placental implantation in the uterine wall.

[0144] For example, the thickness and appearance (texture) of the myometrium (uterine wall) in the placenta accreta region can indicate possible abnormal placenta accreta. Image classification deep learning algorithms can be used to classify the texture of the myometrium as normal or abnormal.

[0145] In addition, placenta 220 may include one or more placental blood pools 290, which are enlarged spaces within the placenta filled with blood.

[0146] Therefore, the above method may also include detecting placental blood pools within the placenta and using blood pool indicators to mark placental blood pools in a 3D rendering of the uterus.

[0147] The detection of placental blood pools can be combined with determining additional physical properties of the placenta, such as placental volume (which can be derived from placental segmentation), in addition to determining the placental location within the uterus. Similarly, the volume of placental blood pools can be determined, for example, through texture analysis and region-based segmentation, and thus the proportion of the placenta occupied by the placental blood pools can be determined.

[0148] It will be recognized that the above image processing methods can utilize various machine learning algorithms, especially for the purposes of feature recognition and image segmentation.

[0149] The above about Figure 1 The described system, or any other system capable of performing the methods described above, may include a user interface adapted to receive user input indicating a region of interest for further study of the 3D rendering of the uterus. For example, the 3D rendering may be displayed on a touchscreen display, which the user can interact with by touching a given region of interest for further study.

[0150] In the example, to improve the quantitative assessment of the uterus, each area of ​​the 3D visualization can be indicated by the user or manually selected to display the corresponding quantification and suggest additional acquisitions (e.g., vascular Doppler) to the clinician.

[0151] Furthermore, 3D visualization can be used to initiate additional acquisitions, such as vascular Doppler assessments of the placental blood pool or at the boundary with the uterine wall. For example, a user can manually select a region of interest, which can initiate additional acquisitions, such as vascular Doppler acquisitions.

[0152] In other words, when a user points to a region of interest on a 3D map of the uterus, the system can suggest performing vascular Doppler acquisition to show, for example, specific flow patterns or increased vascular distribution in the placental blood pool, particularly at the interface between the uterus and the bladder.

[0153] It can store the spatial coordinates of the region of interest and can automatically prepare the ultrasound system for additional acquisitions based on the selected optimal settings for additional acquisitions of the region.

[0154] Once additional acquisitions have been performed, the results (such as flow visualizations from vascular Doppler acquisitions) are displayed on the regular screen and the 3D rendering volume. This can be done by registering the additional acquisition results using previously stored region of interest coordinates and updating the global 3D rendering.

[0155] By studying the accompanying drawings, the disclosure, and the dependent claims, those skilled in the art, in practicing the claimed invention, will be able to understand and implement other variations of the disclosed embodiments. In the claims, the word "comprising" does not exclude other elements or steps, and the words "a" or "an" do not exclude a plurality. Although certain measures are described in dissimilar dependent claims, this does not indicate that combinations of these measures cannot be advantageously used. No reference numerals in the claims should be construed as limiting the scope.

Claims

1. A method (100) for performing an assessment of the placenta, the method comprising: Obtain a 3D ultrasound image of (110) the uterus, which contains the placenta; The placenta is segmented (130) from the 3D ultrasound image of the uterus using an image segmentation algorithm; and (140) A 3D drawing of the uterus is generated based on the result of the segmentation of the placenta and the 3D ultrasound image of the uterus, wherein generating the 3D drawing of the uterus includes: Identify (150) the position of the placenta within the uterus relative to the anatomical structures within the 3D ultrasound image of the uterus; Obtain (160) anatomical reference data regarding potential risks associated with the position of the placenta within the uterus relative to the anatomical structure; The location of the placenta is compared with the anatomical reference data (170). (180) A 3D rendering of the uterus is generated based on the 3D ultrasound image of the uterus, the 3D rendering of the uterus including a 3D rendering of the placenta; and The 3D rendering of the placenta is marked (190) using an indicator, wherein the indicator is changed based on the comparison of the placenta’s location with the anatomical reference data.

2. The method according to claim 1, wherein, The method includes: Obtain multiple 3D ultrasound images; and The multiple 3D ultrasound images (120) are combined to form the 3D ultrasound image of the uterus.

3. The method (100) according to any one of claims 1 to 2, wherein, The method further includes segmenting additional anatomical structures within the 3D ultrasound image of the uterus, and wherein generating the 3D rendering of the uterus further includes: Identify the position of the placenta relative to the additional anatomical structures; To obtain additional anatomical reference data regarding potential risks associated with the position of the placenta relative to the additional anatomical structures; The location of the placenta is compared with the additional anatomical reference data; Additional indicators are used to mark one or more of the 3D rendering of the placenta and the 3D rendering of the uterus, wherein the additional indicators are changed based on the comparison of the placenta's location with the additional anatomical reference data.

4. The method (100) according to claim 3, wherein, The additional anatomical structures include one or more of the following: bladder; Umbilical cord; Cervix; Cervical region; and Anatomical structures near the uterus.

5. The method (100) according to claim 1 or 2, wherein, The method also includes determining the physical properties of the placenta based on the segmentation of the placenta.

6. The method (100) according to claim 5, wherein, The physical properties include one or more of the following: Placental volume; and Placental texture.

7. The method (100) according to claim 1 or 2, wherein, The method further includes detecting anatomical features within the placenta, and wherein generating the 3D rendering of the uterus further includes using feature indicators to mark the detected anatomical features.

8. The method (100) according to claim 7, wherein, The detection of the anatomical features of the placenta is performed based on the results of the segmentation of the placenta and the 3D ultrasound images of the uterus.

9. The method (100) according to claim 8, wherein, Detecting the anatomical features includes: Determine the thickness of the uterine wall near the placenta; and The anatomical features are identified based on the thickness of the walls of the uterus.

10. The method (100) according to claim 8, wherein, The detection of the anatomical features was performed using machine learning algorithms.

11. The method (100) according to claim 1 or 2, wherein, The method also includes performing a Doppler ultrasound-based vascular assessment of the placenta.

12. A computer program product comprising a computer program code module, wherein when the computer program is run on a computer, the computer program code module is adapted to implement the method according to any one of claims 1 to 11.

13. An ultrasound system (2) for evaluating the placenta, said ultrasound system comprising: An ultrasound probe (4) is suitable for acquiring 3D ultrasound images of the uterus and placenta; Processor, which is suitable for: The placenta is segmented from the 3D ultrasound image of the uterus using an image segmentation algorithm; and A 3D model of the uterus is generated based on the results of the segmentation of the placenta and the 3D ultrasound image of the uterus, wherein generating the 3D model of the uterus includes: Identify the position of the placenta within the uterus relative to the anatomical structures within the 3D ultrasound image of the uterus; Obtain anatomical reference data regarding potential risks associated with the position of the placenta within the uterus relative to the anatomical structure; The location of the placenta is compared with the anatomical reference data; A 3D model of the uterus is generated based on the 3D ultrasound image of the uterus, the 3D model of the uterus including a 3D model of the placenta; and The 3D rendering of the placenta is marked using indicators, wherein the indicators change based on a comparison of the placenta's location with anatomical reference data; and Display unit (40) adapted to display the 3D rendering of the uterus to the user.

14. The ultrasound system according to claim 13, wherein, The ultrasound system further includes a user interface adapted to receive user input indicating a region of interest for further study of the 3D rendering of the uterus, and wherein the processor is adapted to: A set of acquisition settings is determined based on the indicated region of interest; and Apply the set of acquisition settings to the ultrasound system.

15. The ultrasound system according to claim 14, wherein, Additional 3D ultrasound images of the uterus are acquired, and the processor is further adapted to update the 3D rendering of the uterus based on the additional 3D ultrasound images of the uterus.

Images