Device and method for detecting bone fractures

Abstract

The present invention proposes an apparatus (120) and a method for detecting a fracture of an object based on an ultrasound image. The apparatus (120) includes a first fracture detector (122) and a second fracture detector (124). The first fracture detector (122) is configured to: receive a first ultrasound image of a region of the object, identify bones in the first ultrasound image, identify at least one focal zone in the region based on the identified bones, generate focal zone information indicating the position of the at least one focal zone, and guide the acquisition of a second ultrasound image based on the generated focal zone information. The second fracture detector (124) is configured to: receive the second ultrasound image, and detect a fracture based on the second ultrasound image. The second ultrasound image has a higher resolution in the at least one focal zone than in the rest of the region, and the resolution of the at least one focal zone in the second ultrasound image is higher than the resolution of the at least one focal zone in the first ultrasound image.

Description

Technical Field

[0001] The present invention relates to ultrasound imaging, and more particularly to an apparatus and method for detecting bone fractures based on ultrasound images. Background Art

[0002] Fractures resulting from traumatic injuries (injuries caused by trauma) are very common. For example, rib fractures are the most common injuries resulting from blunt force (chest) trauma. There are several different methods for diagnosing fractures, including physical examination, conventional radiography (e.g., diagnostic X-rays), advanced medical imaging modalities (e.g., CT, MRI, or SPECT (not generally recommended for this task)), and ultrasound imaging. Currently, physical examination and conventional radiography are the main diagnostic tools used in daily clinical practice, but they cannot detect all fractures. For example, only about 49% of rib fractures can be detected based on an accurate physical examination and conventional radiographs. CT and MRI are able to provide the most detailed information about the fracture, including location, fragment dislocation, etc. However, these more advanced medical imaging modalities are not always available in resource-limited settings (e.g., pre-hospitalization and some small hospitals in developing countries), and they are not suitable for evaluating fracture reduction, which requires repeated serial imaging throughout the treatment process.

[0003] Ultrasound imaging has been recognized as a good first alternative for diagnosing fractures in adults and especially in children and pregnant women. Ultrasound imaging allows identification of the exact location of the rib fracture to help determine the site of the surgical incision prior to rib fixation surgery, thereby eliminating the possibility of having to enlarge the incision. Ultrasound imaging has shown advantages in the evaluation of fractures in children because of the relatively thin chest wall (soft tissue) in children compared to adults. Further advantages of ultrasound imaging include its portability, noninvasiveness, and low cost-effectiveness, as well as the ability to perform repeated examinations at the bedside to evaluate fracture reduction throughout the entire procedure.

[0004] However, manual detection of fractures in ultrasound images is a highly operator-dependent, time-consuming process. This is always a challenging task for inexperienced users, especially when examination time is critical. For example, in the emergency room, the examination time for 12 pairs of ribs in the entire chest is preferably less than 12 minutes, but the shorter the better. In order to help doctors quickly identify bone surfaces or hyperechoic lines from bones, some methods for automatic bone segmentation from ultrasound B-mode images using intensity and gradient information or morphological features have been proposed. For example, in "Automatic bone localization and fracture detection from volumetric ultrasound images using 3-D local phase features" published by I. Hacihaliloglu et al. in "Ultrasound in Medicine and Biology" (Vol. 38, pp. 128-144) in 2012, a method for automatic bone localization and fracture detection using local phase features of 3D ultrasound data was developed. These existing methods have complex requirements for radio frequency (RF) ultrasound signals and require a high computational load in 3D local phase estimation.

[0005] US 2007 / 043290 A1 discloses a device for detecting bone fractures by comparing certain parameters of waves reflected from bones with threshold conditions. Lius Nascimento et al., "Computer aided bone fracture identification based on ultrasound image", propose a 3-stage process for automatically identifying bone fractures. However, further enhancements to computer-aided bone fracture detection are needed. Summary of the invention

[0006] It would be advantageous to provide improved methods for detecting bone fractures based on ultrasound images.

[0007] According to an embodiment of the first aspect of the present invention, a device for detecting a fracture of an object based on an ultrasound image is provided. The device includes a first fracture detector, which is configured to: receive a first ultrasound image of a region of the object, and identify bones in the first ultrasound image. The first fracture detector is also configured to: identify at least one focal zone in the region based on the identified bones, generate focal zone information indicating the position of the at least one focal zone, and guide the acquisition of a second ultrasound image based on the generated focal zone information. For example, the at least one focal zone can include at least part of the identified bones. The device also includes a second fracture detector, which is configured to: receive the second ultrasound image, and detect a fracture based on the second ultrasound image, and the second ultrasound image is acquired based on the generated focal zone information, and the second ultrasound image has a higher resolution in the at least one focal zone than in the rest of the region. The resolution of the at least one focal zone in the second ultrasound image is higher than the resolution of the at least one focal zone in the first ultrasound image.

[0008] In this way, the first ultrasound image is first used to identify the bone in the region, and the at least one focal area is identified as one or more sub-regions of the region near the bone surface and covering at least part of the identified bone; then the second ultrasound image is acquired in the following manner: the resolution of the at least one focal area is higher than the resolution of the rest of the region, and the resolution of at least one focal area in the second ultrasound image is higher than the resolution of at least one focal area in the first ultrasound image; and the second ultrasound image is used to detect the fracture. In the conventional method, an ultrasound image with an almost fixed resolution on the imaged region is used for fracture detection. Differently, in the method proposed above, a second ultrasound image with a higher axial resolution and lateral resolution in one or more sub-regions including the bone is acquired and used for fracture detection, and the higher axial resolution and lateral resolution will obtain a finer bone surface appearance, so that the detection is more accurate. In other words, the first ultrasound image is first used to identify one or more focal areas, and then the second ultrasound image is acquired in a manner of magnifying (high resolution) at least one or more identified focal areas. Therefore, some fractures (e.g., micro fractures) that cannot be detected in the first ultrasound image can be detected based on the second ultrasound image with high resolution.

[0009] In some embodiments, the second ultrasound image is acquired by performing additional signal processing on the first ultrasound image only in at least one focal zone. That is, the second ultrasound image is acquired by performing signal processing on the first ultrasound image without transmitting any additional ultrasound signal into at least one focal zone. Advanced signal processing can improve image quality and / or accuracy of fracture detection, but at the cost of more computing time and / or workload. In addition, the larger the zone to be processed, the more computing time or workload it takes. Therefore, it is possible to reduce computing time and / or workload by performing additional signal processing only in at least one focal zone.

[0010] According to some embodiments, the second ultrasound image is acquired by emitting an additional ultrasound beam at least toward at least one focal zone. In an embodiment, the first fracture detector is further configured to instruct the ultrasound image acquisition unit to acquire the second ultrasound image by emitting an additional ultrasound beam at least toward the at least one focal zone, and the second fracture detector is further configured to receive the second ultrasound image acquired by the ultrasound image acquisition unit. The proposed device can be communicatively connected to the ultrasound image acquisition unit in a direct manner or an indirect manner. The device can send the generated focal zone information to the ultrasound image acquisition unit, which can then acquire the second ultrasound image based on the received focal zone information, and send the acquired second ultrasound image to the device (particularly the second fracture detector of the device).

[0011] In some embodiments, at least in at least one focal zone, a first ultrasound image is acquired using a first ultrasound beam setting, and a second ultrasound image is acquired using a second ultrasound beam setting, wherein the second ultrasound beam setting is different from the first ultrasound beam setting. The first fracture detection can be configured to direct the ultrasound image acquisition unit to acquire the second ultrasound image using the second ultrasound setting at least in at least one focal zone. The second ultrasound beam setting can be different from the first ultrasound beam setting in various ways. According to some embodiments, the first ultrasound beam setting and the second ultrasound beam setting differ in at least one of the following aspects: (a) focal band; (b) transmit frequency; (c) transmit pulse length; (d) steering angle.

[0012] According to some embodiments, the first fracture detector is also configured to detect a hematoma adjacent to the identified bone based on the second ultrasound image, and the at least one focal zone is generated based on the detected hematoma. For example, the at least one focal zone includes at least a portion of the detected hematoma. For example, the at least one focal zone can cover only a portion of the identified bone adjacent to the detected hematoma, rather than covering all of the identified bone. In this way, the at least one focal zone can be even smaller. Generally, the smaller the at least focal zone, the higher the resolution of at least one focal zone in the second ultrasound image can be, and / or the less work required to acquire the same resolution level for the at least one focal zone. When a fracture occurs, there is often a hematoma adjacent to the fracture. In an ultrasound image, a fresh hematoma appears as a low echo pattern.

[0013] According to some embodiments, the first fracture detector is further configured to detect fractures based on the first ultrasound image. Although the resolution of the first ultrasound image is lower than the resolution of the second ultrasound image at least in at least one focal region, some fractures (e.g., displaced fractures) can be detected based on the first ultrasound image.

[0014] In some specific embodiments, the first fracture detector is configured to first detect a fracture based on the first ultrasound image and then identify at least one focal region. In some further embodiments, the first fracture detector is configured to identify at least one focal region if the first fracture detector does not detect a fracture based on the first ultrasound image and / or not identify at least one focal region if the first fracture detector detects a fracture based on the first ultrasound image. In some further embodiments, the second fracture detector can be configured to receive a second ultrasound image of the region and detect a fracture based on the second ultrasound image only if the first fracture detector does not detect a fracture based on the first ultrasound image.

[0015] In this way, once a fracture is detected based on the first ultrasound image, there is no need to acquire a second ultrasound image or to detect a fracture based on the second ultrasound image. Sometimes (e.g., in an emergency situation), it is more important to confirm a fracture than to exclude a fracture, or in other words, once at least one fracture (e.g., a large bone fracture) is detected, the detection of fractures can be stopped. This is particularly advantageous in the case of a large area of ​​the object to be scanned. For example, in the case of rib fracture detection, some or all of the 12 pairs of ribs will be examined.

[0016] According to some embodiments, the apparatus further comprises a controller configured to switch between a first detection mode and a second detection mode, and to control the ultrasound image acquisition unit to acquire the first ultrasound image in the first detection mode, and to control the ultrasound image acquisition unit to acquire the second ultrasound image or both the first ultrasound image and the second ultrasound image in the second detection mode.

[0017] In some embodiments, the controller is further configured to switch between the first detection mode and the second detection mode based on a user input received from a user interface. The user input can take various forms, including physical contact with the user interface (e.g., operating a physical button, touching a touch panel, etc.) or contactless input (e.g., audio input, gesture input, etc.).

[0018] By allowing switching between the first detection mode and the second detection mode, the device can enable a more flexible detection mode or workflow to meet a variety of situations. For example, the first detection mode may be suitable for time-sensitive situations when small bone fractures are not very critical. In other words, it is more important to detect large bone fractures at this time. For example, in some cases, a clinician can first use the first detection mode for a quick and rough inspection, and then switch to the second detection mode for a detailed inspection if necessary. In some embodiments, in the first detection mode, the device can be configured to present any detected bones and / or any identified focal areas (e.g., those focal areas including hematomas).

[0019] According to some embodiments, the first fracture detector is further configured to receive a plurality of first ultrasound images acquired when the ultrasound transducer is moving relative to the object and first position information about the ultrasound transducer relative to the object corresponding to each of the plurality of first ultrasound images. The position information about the ultrasound transducer relative to the object corresponding to a particular first ultrasound image is information about a position relative to the ultrasound transducer at which the particular first ultrasound image was acquired. The first fracture detector is further configured to: identify bones in each of the plurality of first ultrasound images and detect the presence of hematomas adjacent to the bones, and identify a subset of the plurality of first ultrasound images in which the hematomas adjacent to the bones exist. The first fracture detector is further configured to generate second position information about the ultrasound transducer relative to the object corresponding to each of the subsets in the plurality of first ultrasound images. For example, the plurality of first ultrasound images are acquired while the ultrasound transducer is moved along the chest of the object or along the ribs of the object.

[0020] In some embodiments, the user interface is configured to present the second location information.

[0021] According to an embodiment of the second aspect of the present invention, an ultrasound system for detecting a fracture of an object is provided. The ultrasound system comprises: an ultrasound image acquisition unit, which comprises an ultrasound transducer; the above-mentioned device for detecting a fracture, which is coupled to the ultrasound image acquisition unit; and the ultrasound image acquisition unit is configured to acquire the first ultrasound image and the second ultrasound image. The ultrasound transducer is configured to: transmit an ultrasound signal to a region of the object, and receive an ultrasound echo signal from the region.

[0022] According to an embodiment of the second aspect of the present invention, a method for detecting bone fractures of an object based on an ultrasound image is proposed. The method comprises the following steps: receiving a first ultrasound image of a region of the object; identifying bones in the first ultrasound image; identifying at least one focal zone within the region based on the identified bones. For example, the at least one focal zone includes at least part of the identified bones. The method further comprises the following steps: guiding the acquisition of a second ultrasound image of the region, and receiving the acquired second ultrasound image, wherein the second ultrasound image has a higher resolution in the at least one focal zone than in the rest of the region, and the resolution of the at least one focal zone in the second ultrasound image is higher than the resolution of the at least one focal zone in the first ultrasound image. The method further comprises the following steps: detecting bone fractures based on the second ultrasound image.

[0023] According to an embodiment of the third aspect of the present invention, a computer-readable medium comprising executable instructions is provided, wherein the executable instructions, when executed, cause a computer processor to perform the above method.

[0024] Other objects and advantages of the present invention will become more apparent and can be easily understood with reference to the description taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Hereinafter, the present invention will be described and explained in more detail with reference to the accompanying drawings in which:

[0026] Figure 1 An ultrasound system for detecting a bone fracture in a subject according to some embodiments of the present invention is illustrated;

[0027] Figure 2 A method of detecting a bone fracture of an object based on an ultrasound image according to some embodiments of the present invention is illustrated;

[0028] Figure 3 illustrates two ultrasound B-mode images according to some embodiments of the present invention;

[0029] Figures 4A-4Cillustrates at least one focal region of an imaging region according to various embodiments of the present invention;

[0030] Figure 5 illustrates a plurality of ultrasound images acquired while an ultrasound transducer is moving relative to an object according to some embodiments of the present invention;

[0031] Figure 6 illustrates a schematic diagram of a user interface according to some embodiments of the present invention; and

[0032] Fig. 7A The diagram shows Figure 3 A sub-region selected in an ultrasound B-mode image; Figures 7B-7F Pictured Fig. 7A Five different texture analyses of sub-regions in .

[0033] In the accompanying drawings, the same reference numerals indicate similar or corresponding features and / or functions. DETAILED DESCRIPTION

[0034] The present invention will be described with respect to particular embodiments and with reference to certain drawings, but the invention is not limited thereto but only by the claims. The drawings described are merely illustrative and non-limiting. In the drawings, the size of some of the elements may be exaggerated and not drawn to scale for illustrative purposes.

[0035] Figure 1 An ultrasound system 100 for detecting a bone fracture in a subject is illustrated in accordance with some embodiments of the present invention. Figure 2 A method 200 of detecting a bone fracture of a subject based on ultrasound images according to some embodiments of the present invention is illustrated.

[0036] The ultrasound system 100 includes an apparatus 110 for detecting a fracture of an object, an ultrasound image acquisition unit 110, and a user interface 130. The ultrasound image acquisition unit 110 can be an ultrasound probe. The apparatus 120 is communicatively connected to the ultrasound image acquisition unit 110 and the user interface 130 in various ways (including wired or wireless connection, local or remote connection, etc.). The user interface 130 can also be communicatively connected to the ultrasound image acquisition unit 110 as appropriate. The apparatus 120, the ultrasound image acquisition unit 110, and the user interface 130 can be integrated into a single device, or can be physically separated from each other. Each of the apparatus 120, the ultrasound image acquisition unit 110, and the user interface 130 can include one or more devices. In some embodiments, the ultrasound system 100 can include an ultrasound probe that can be connected to a computer system, and the computer system includes a processor and a memory. For example, the computer system can be a personal computer, a smart phone, or a tablet computer. The apparatus 120 includes a processor and a memory of the computer system, wherein the memory includes executable instructions that, when executed, cause the computer processor to perform the method 200. The user interface 130 can also be part of the computer system or can be connected to the ultrasound probe and / or the computer system.

[0037] The ultrasound image acquisition unit 110 can be any kind of device including an ultrasound transducer capable of transmitting and / or receiving ultrasound signals. For example, the ultrasound image acquisition unit 110 is configured to both transmit ultrasound signals toward the region of interest and receive echoes of the transmitted ultrasound signals from the region of interest. The ultrasound image acquisition unit 110 can be configured to acquire two-dimensional (2D) or three-dimensional (3D) ultrasound images. In some examples, the ultrasound transducer can include an array of transducer elements, so that a 2D plane can be scanned. In some examples, the ultrasound transducer can include a matrix of transducer elements, so that a 3D volume can be directly scanned. In some other examples, the ultrasound transducer includes an array of transducer elements, the array of which is repositioned and / or reoriented to automatically or manually scan multiple different planes, and the two-dimensional ultrasound data of the multiple planes can be assembled into three-dimensional ultrasound data of the volume of interest. For example, the ultrasound image acquisition unit 110 can include a mechanical three-dimensional ultrasound probe.

[0038] Typically, the ultrasound echo signals received by the transducer are beamformed, the beamformed signals are processed by various signal processing (e.g., bandpass filtering, extraction, I and Q component separation, harmonic signal separation, speckle reduction, signal compounding and / or noise elimination), and the processed signals can be further processed to obtain B-mode data, Doppler data, strain data, motion data, etc. Various ultrasound data can be further processed to form different types of ultrasound images. These signal and / or image processing can be performed by the same device or multiple devices. In some examples, the ultrasound probe can include an ultrasound transducer and one or more components configured to generate ultrasound images. In some examples, the ultrasound probe can include an ultrasound transducer and one or more components for performing some of the processing, and then another device is configured to perform the rest to generate ultrasound images based on the output of the ultrasound probe. Such a device can be an independent device or part of a computer system.

[0039] like Figure 1 As shown, the device 120 includes a first fracture detector 122 , a second fracture detector 124 , and optionally a controller 126 .

[0040] The first fracture detector 122 is configured to receive a first ultrasound image of a region of the object. Here, the region of the object is sometimes referred to as an imaged region. The first ultrasound image can be a 2D image or a 3D image. The region of the object can be a 2D plane of the object or a 3D volume of the object.

[0041] A region of the object can be pre-selected based on prior knowledge about the skeleton of the object. Based on prior knowledge of the skeleton of the object, the region of the object can be selected as a region that includes a particular bone of interest. For example, in the case where the object is a human, if the ribs are the bones of interest, the region of the object can be a portion of the chest of the object, or, if the humerus is the bone of interest, the region of the object can be a portion of the upper arm of the object. In some embodiments, a first ultrasound image is acquired based on one or more predetermined anatomical parameters of the object. In this way, the first ultrasound image can be acquired by setting the focal area at the depth of the bone of interest to enhance the resolution at that depth. For example, if a rib fracture of a human is to be detected, the anatomical parameters include a parameter indicating the depth between the surface of the skin of the chest and the top of the ribs. Such a parameter can be determined as the thickness of the chest wall (CWT) minus a predetermined value (e.g., 0.5 cm) indicating the typical distance between the pleural line and the ribs of a human. For example, the CWT can be a value determined based on a body mass index, or a value extracted from a historical medical examination (e.g., a previously acquired ultrasound image of a particular person). Additionally or alternatively, several different acquisition modes can be automatically or manually selected based on the CWT. For example, for a large CWT (e.g., ), a penetration mode is selected in which the focal region is set to a predetermined large depth value; for a medium CWT (e.g., ), a general mode is selected in which the focal region is set to a predetermined medium depth value; for small CWT (e.g., ), selecting a resolution that sets the focus area to a predetermined small depth value).

[0042] The first bone fracture detector 122 is also configured to identify any bones in the received first ultrasound image. The identification of bones can be implemented in any suitable manner (including known or future developed manners). In some embodiments, the identified bones can be presented to the user, for example, via the user interface 130. For example, the first ultrasound image is displayed, and the identified bones are outlined in the displayed first ultrasound image. Figure 3 , the ultrasound B-mode image 310 is an example of a first ultrasound image received by the first fracture detector. The ultrasound B-mode image 310 is an ultrasound B-mode image 310 overlaid with a solid line that outlines the identified bone 322. Figure 3 , the y-axis extends along the depth direction and the x-axis extends along the lateral direction.

[0043] In some embodiments, if the first fracture detector does not identify the bone, an alarm signal can be generated. In addition, the first fracture detector 122 can be configured to: receive user input indicating a bone manually identified by a user and / or user input (e.g., a seed point) to assist bone detection, and identify the bone based on the user input. In an example, the user input can indicate one or more seed points, and the first fracture detector can be configured to identify the bone based on the indicated seed points. In another example, the user input can indicate a spatial area, and the first fracture detector can be configured to identify the bone based on the indicated spatial area. For example, the first fracture detector can be configured to: identify bones within a spatial area, or identify bones having the spatial area and bones having an area adjacent to the spatial area.

[0044] The first bone fracture detector 122 is further configured to: identify at least one focal zone within the region based on the first ultrasound image, and generate focal zone information indicating a location of the at least one focal zone. The at least one focal zone includes at least a portion of the identified bone. Figures 4A-4C At least one focal region of an imaging region according to various embodiments of the present invention is illustrated.

[0045] The at least one focal region can have any regular or irregular shape. Figure 4A , the at least one focus area 430 can be a rectangular area including or covering the identified bone 420. For example, referring to Figure 4B, the at least one focal region 430 can be a band region centered on the identified bone 420. The focal region information can indicate the position of the at least one focal region along at least one dimension. For example, the focal region information can indicate the depth of the at least one focal region, or the focal region information can indicate the lateral position and depth of the at least one focal region in the two-dimensional imaged area.

[0046] In some embodiments, the first fracture detector 122 is further configured to detect a hematoma adjacent to the identified bone based on the detected hematoma. The at least one focal region can be generated to include or cover a portion of the identified bone adjacent to the detected hematoma. For example, referring to Figure 4C , a hematoma 440 adjacent to the identified bone 420 is detected, and at least one focus area 430 is identified as an area covering the hematoma 440 and the portion of the bone located below the hematoma 440. For example, the at least one focus area may include a portion of the identified bone whose distance from the detected hematoma is less than a predetermined threshold. In addition, the user can define one or more focus areas via a user interface, or in other words, the first fracture detector can also be configured to: receive user input indicating one or more focus areas, and identify at least one focus area based on the user input. In an example, the user input can indicate a spatial point, and the first fracture detector can be configured to identify the focus area as a region having a predetermined shape and size and centered on the spatial point indicated by the user input. In another example, the user input can indicate a spatial region by means of a point, a line, etc., and the first fracture detector can be configured to identify the focus area based on the spatial region indicated by the user input. The identified focus area corresponds exactly to the indicated spatial region, or can be a region of a predetermined shape covering the indicated spatial region.

[0047] A hematoma adjacent to a bone can be detected in any suitable manner, such as by means of texture analysis. In some embodiments, a sub-region of the imaged area can be selected to perform texture analysis. The sub-region can be an area above the identified bone. Here, the term "above" refers to a more superficial or shallower place along the depth direction. Reference Figure 3 In the ultrasound image 320 in FIG. 1 , the sub-region can be selected as a region between two curves 322 and 324 , wherein the curve 322 illustrates the identified bones and the curve 324 is spaced apart from the curve 322 by a predetermined value.

[0048] Texture analysis can include calculating any of the co-occurrence matrices and their properties (e.g., correlation, entropy, contrast, energy, and uniformity). Energy is defined as the sum of the squares of each matrix element and reflects the uniformity of the grayscale distribution and texture roughness of the image. The same values ​​for all co-occurrence matrices result in a smaller energy profile; while unequal values ​​between co-occurrence matrices may be expected to have high energy. Contrast reflects the sharpness of the image and the depth of the texture grooves. Deeper texture grooves are associated with higher contrast and better visual acuity, while lower contrast is attributed to shallower grooves and blurred images. A higher number of pixels with higher grayscale differences (e.g., contrast profiles) are associated with higher contrast values. Entropy reflects the inhomogeneity and complexity of the image texture. Correlation reflects the consistency of the image texture. Uniformity reflects the uniformity of the image texture and scales the local variations of the image texture. A high uniformity value indicates that there is no intra-regional variation in the image texture and that it is locally uniformly distributed.

[0049] Fig. 7A The diagram shows Figure 3 A region is selected from a sub-region between two curves 322 and 324 in the ultrasound B-mode image 320 of B. Figures 7B-7F Picture shows Fig. 7A Five different texture analyses of the area shown in Figures 7B-7F In each figure, the x-axis represents the lateral direction, and the y-axis represents the magnitude of the corresponding texture attribute. The five texture attributes are Figure 7B is entropy, Figure 7C is the correlation, Fig.7D In is the contrast, in Fig. 7E is energy, and Figure 7F The middle is uniformity.

[0050] It is known that there is a hematoma around the lateral position 200. Figures 7B-7F ,At least Figure 7B The entropy value in Fig. 7E The energy value and Figure 7F The uniformity values ​​in show significant variations around the lateral position 200 where the hematoma is located.

[0051] The analysis shows that for normal subperiosteal soft tissue, the entropy value is low, while for subperiosteal hematoma, the entropy value is high. It can be noticed that user settings such as gain and TGC affect some parameters (e.g., contrast and energy). However, these settings have less impact on parameters such as correlation and entropy. Entropy reflects the inhomogeneity and complexity of the image texture, so high entropy indicates more randomness within the image.

[0052] Several other features can also be used, such as: classical backscattering, echogenicity, attenuation coefficients, scattering parameters, parameters of amplitude or intensity histograms (e.g., Nakagami shape parameters), spectral features in RF signal-based analysis, and motion features. Since there are significant differences in ultrasound images of normal subperiosteal soft tissue and subperiosteal hematoma, it is recommended to use texture features for measuring speckle patterns in our case. Many research results have shown that statistical texture analysis methods can well characterize tissue properties and provide complementary information to echogenicity.

[0053] The first fracture detector 122 is also configured to issue an instruction for acquiring a second ultrasound image of the region based on the generated focal zone information, and the second fracture detector 124 is configured to receive the acquired second ultrasound image of the region and detect fracture based on the second ultrasound image. The second ultrasound image is acquired based on the generated focal zone information, the second ultrasound image has a higher resolution in at least one focal zone than the rest of the region, and the resolution of at least one focal zone in the second ultrasound image is higher than the resolution of at least one focal zone in the first ultrasound image. In some embodiments, the second fracture detector 124 can be configured to detect fracture by combining the first ultrasound image with the second ultrasound image.

[0054] Fractures can be detected in any suitable manner. In some examples, the bone surface is first detected, and then the features of the detected bone (e.g., smoothness and continuity) are extracted. In addition, other features (e.g., detected hematoma, or texture features, strain of tissue adjacent to the bone) can also be used to detect fractures. In some examples, fractures can be detected using an artificial intelligence-based algorithm. Additionally or alternatively, fracture detection can be assisted by one or more user inputs. For example, if a user detects a fracture by observing an ultrasound image, the user can indicate the location of the detected fracture via a user interface.

[0055] According to some embodiments of the present invention, the second ultrasound image is acquired by performing additional signal processing on the first ultrasound image only in at least one focal zone. In some examples, the first fracture detector can be configured to perform signal processing on the first ultrasound image to acquire the second ultrasound image by itself, or can be configured to instruct another processing unit for performing signal processing to acquire the second ultrasound image. The first fracture detector can send the generated focal zone information to another processing unit, and the other processing unit is configured to perform additional signal processing on the first ultrasound image only in at least one focal zone based on the generated focal zone information. This other processing unit can be part of the device 120, or can be communicatively connected to the device 120.

[0056] According to some embodiments of the present invention, the second ultrasound image is acquired by transmitting an additional ultrasound beam at least toward at least one focal region. The first fracture detector 122 can be communicatively connected to the ultrasound image acquisition unit 110 and can be configured to direct the ultrasound image acquisition unit 110 to acquire the second ultrasound image by transmitting an additional ultrasound beam at least toward at least one focal region.

[0057] In some embodiments, the first ultrasound image is acquired using a first ultrasound beam setting in at least at least one focal zone, and the second ultrasound image is acquired using a second ultrasound beam setting in at least at least at least one focal zone, wherein the second ultrasound beam setting is different from the first ultrasound beam setting. The second ultrasound beam setting can be different from the first ultrasound beam setting in various ways. According to some embodiments, the first ultrasound beam setting and the second ultrasound beam setting differ in at least one of the following aspects: (a) focal zone; (b) transmit frequency; (c) transmit pulse length; (d) steering angle.

[0058] In an embodiment, the second ultrasound image is acquired by focusing the transmitted and / or received ultrasound beam on at least one focal zone (e.g., focusing on the depth of at least one focal zone). For example, the second ultrasound image is acquired by automatically setting the focal zone of the ultrasound beam to the at least one focal zone.

[0059] Ultrasound focusing is a well-known technique. With this technique, a phased array transducer can adjust the narrowest point of an ultrasound beam at any position along its length. This is called a focal zone, and the highest resolution is usually obtained within a 2-4 cm portion of the ultrasound beam. Some existing ultrasound machines can effectively focus in multiple focal zones along the ultrasound beam, but at the expense of affecting other factors (e.g., frame rate) because more than one scan or pulse is required. More details of this technique can be found in the prior art, for example, B. Haider's "Power drive circuits for Diagnostic Medical Ultrasound" (Proceedings of the IEEE International Symposium on Power Semiconductor Devices and ICs, June 4-8, 2006). Conventionally, a user can manually adjust the focal zone at various depths in order to see each structure most clearly. Advantageously, according to some embodiments of the present invention, the device 120 can automatically identify at least one focal zone and automatically set the focal zone at the at least one focal zone without user intervention. Therefore, an ultrasound image with higher resolution at the desired focal zone can be acquired without any user input.

[0060] In further embodiments, in the second ultrasound beam setting, the transmit frequency can be increased and / or the transmit pulse length can be reduced in order to obtain a higher resolution.

[0061] In another embodiment, in the second ultrasound beam setting, multiple steering angles can be used. In addition, one or more steering angles can be dynamically determined. For example, one or more steering angles can be determined based on the orientation of the detected bone.

[0062] The first fracture detector 122 can also be configured to detect fractures based on the first ultrasound image. Although the resolution of the first ultrasound image is lower than that of the second ultrasound image at least in at least one focal region, some fractures (e.g., displaced fractures) can be detected based on the first ultrasound image.

[0063] In addition, the controller 126 is configured to switch between the first detection mode and the second detection mode, and control the ultrasound image acquisition unit 110 to acquire the first ultrasound image in the first detection mode, and control the ultrasound image acquisition unit 110 to acquire the second ultrasound image or acquire both the first ultrasound image and the second ultrasound image in the second detection mode. For example, the device 120 can be configured to operate in the first detection mode first, and then operate in the second detection mode. Since the resolution of the second ultrasound image is higher than the resolution of the first ultrasound image at least in at least the focal region, some fractures that cannot be detected in the first mode can be detected in the second mode. In some embodiments, in the second mode, the second ultrasound image is acquired based on the focal region information generated in the first mode. In some other embodiments, in the second mode, the first ultrasound image is first acquired to generate the focal region information, and then the second ultrasound image is acquired based on the focal region information.

[0064] In some embodiments, controller 126 is further configured to switch between the first detection mode and the second detection mode based on user input received from user interface 130. For example, a user can first perform a relatively coarse scan by selecting the first detection mode, and then switch to a relatively fine scan by selecting the second detection mode.

[0065] refer to Figure 2 , a method 200 for detecting a bone fracture of an object based on an ultrasound image is illustrated. In step 210, a first ultrasound image of a region of the object is received. In step 220, a bone is identified in the first ultrasound image. In step 240, at least one focal region within the region is identified based on the identified bone. In step 250, a second ultrasound image of the region is received, wherein the second ultrasound image has a higher resolution in at least one focal region than in the rest of the region. In step 260, a bone fracture is detected based on the second ultrasound image.

[0066] In some embodiments, in step 240, a hematoma adjacent to the identified bone is detected in the first ultrasound image, and at least one focal zone within the region can be identified based on the detected hematoma.

[0067] In some further embodiments, the method 200 may further include step 230. In step 230, a fracture is detected based on the first ultrasound image. The method 200 may proceed in various ways after step 230. In some embodiments, if no fracture is detected in step 230, the method 200 proceeds to step 240, and if a fracture is detected in step 230, the method 200 skips steps 240-260. For example, when a fracture is to be identified, detection can be stopped once any fracture is detected. In some other embodiments, the method 200 proceeds to step 240 regardless of whether any fracture is detected in step 230.

[0068] The field of view of an ultrasound transducer is typically small compared to the area being scanned or examined for fracture detection, so typically the ultrasound transducer is swept over the area of ​​interest, such as the chest area when detecting rib fractures.

[0069] Figure 5 Illustrated are a plurality of ultrasound images 540 acquired while an ultrasound transducer is moving relative to an object, according to some embodiments of the present invention. Figure 5 A bone 510 with a fracture 530 is shown beneath skin 520. As the ultrasound transducer is swept across the area of ​​the subject, the corresponding positions P1, ..., P n Collect multiple ultrasonic images F1, ..., F n The movement of the ultrasonic transducer relative to the object can be directly controlled by the user. Alternatively, the ultrasonic transducer can be connected to a drive device capable of moving the ultrasonic transducer relative to the object. The ultrasonic transducer can be freely swept across the object, so that multiple positions P1, ..., P n It can be any position, and if the ultrasound transducer remains at the same position for a period of time, two or more ultrasound images can be acquired at the same position. These positions can be recorded by tracking the position of the ultrasound transducer in various ways (e.g., EM sensor, motion sensor, acceleration sensor, etc.).

[0070] In some embodiments, at each location, it is possible to perform Figure 2 For example, a first ultrasound image of a corresponding area of ​​an object is acquired, bones in the first ultrasound image are identified, at least one focal area in the area is identified based on the identified bones, a second ultrasound image of the area is acquired, and then a fracture is detected based on the second ultrasound image.

[0071] According to some other embodiments of the present invention, the first fracture detector 122 is further configured to receive a plurality of first ultrasound images acquired when the ultrasound transducer is moving relative to the object and first position information about the ultrasound transducer relative to the object corresponding to each of the plurality of first ultrasound images. In other words, the ultrasound image F i The first position information indicates the corresponding position P i The first fracture detector 122 is also configured to identify a bone and detect the presence of a hematoma adjacent to the bone in each of the plurality of first ultrasound images, and to identify a subset 550 of the plurality of first ultrasound images in which a hematoma adjacent to the bone is present.

[0072] In some embodiments, the first fracture detector 122 can also be configured to detect fractured bones and can identify different subsets of the plurality of first ultrasound images. For example, the first subset includes first ultrasound images in which fractures are detected, and the second subset includes first ultrasound images in which fractures are not detected by the first fracture detector but hematomas are detected.

[0073] The first fracture detector 122 is also configured to generate second position information about the ultrasonic transducer relative to the object corresponding to a subset of the plurality of first ultrasound images. Such second position information can be presented via a user interface. Given such second position information, a user can operate the ultrasound image acquisition unit 110 to scan one or more of those positions. In some examples, a user can trigger the device 120 to switch from the first detection mode to the second detection mode and scan one or more of those positions in the second detection mode.

[0074] Figure 6 An exemplary view 600 of a display 600 of a user interface according to some embodiments of the present invention is illustrated. The view 600 may include a portion 610 presenting a bone(s) on which a fracture is to be detected (e.g., Figure 6ribs shown). Image 612 can be a general anatomical model of bones, or can be an image of bones of a specific object (e.g., a 3D image reconstructed from a previously acquired CT image, MR image, or ultrasound image). In some embodiments, the second position information can be presented as a mark 614 in image 612. For example, whenever a fracture and / or hematoma is detected in an ultrasound image, the location where the ultrasound image is acquired is marked in the corresponding location in image 612. Different marks can be used to indicate fractures and hematomas, respectively. View 600 can include a portion 620 presenting an ultrasound image. The ultrasound image can be a real-time image, or an ultrasound image previously acquired at a position selected by a user. In some examples, a detected fracture and / or a detected hematoma can be presented in an ultrasound image. View 600 can include one or more additional portions 630 for presenting various information and / or receiving various user inputs. The sizes and arrangements of portions 610-630 are for illustrative purposes only. Any suitable size and arrangement can be implemented. In some embodiments, view 600 can be dynamic and / or user-configurable.

[0075] The device 120 can also include a report creator configured to record detected fractures and / or detected hematomas and optional location information, and generate a report based on the recorded information. In some examples, the report can include one or more of the following information:

[0076] (a) whether any fractures were detected;

[0077] (b) the number of fractures detected;

[0078] (c) location of detected fracture;

[0079] (d) whether any hematoma or abnormal soft tissue was detected adjacent to the bone;

[0080] (e) Recommending further steps, such as more advanced medical imaging such as CT or repeated ultrasound examinations if no fractures are detected but the subject is still experiencing pain.

[0081] An exemplary workflow for rib fractures according to an embodiment of the present invention will be described below.

[0082] 1) The user selects a rib to start ultrasound scanning, and marks the rib on a 3D CT chest image or a 3D MRI chest image or a 3D model of the chest;

[0083] 2) Use a broadband linear ultrasound probe to determine the depth from the skin surface to the ribs, with specific settings determined and set based on the depth of the specific subject;

[0084] 3) Gently move the ultrasound probe along the rib axis (from medial to lateral) to acquire a 2D ultrasound image sequence. The user can assist in rib identification by providing seed points for rib surface detection. Alternatively, shadows or loss of correlation beneath the suspicious rib can be used for a more automated detection.

[0085] 4) Extract the echo texture of the rib attributes (e.g., any gaps / steps / smoothness) and the subperiosteal soft tissue (i.e., the soft tissue adjacent to the bone) to determine if there are any gaps / steps on the rib and / or any subperiosteal hematomas (i.e., hematomas adjacent to the bone). If any rib has a gap / step / lower smoothness, the rib is considered to be fractured; otherwise the rib may be normal or indeterminate. For example, if the bone surface is smooth and the subperiosteal soft tissue is normal, the rib is considered normal, or in other words, there is no rib fracture. If the rib surface is smooth but the adjacent soft tissue looks like a hematoma or swelling based on its texture pattern, the rib fracture can be considered indeterminate.

[0086] 5) If a rib fracture is detected or is uncertain, the corresponding position is determined with the help of a motion / positioning sensor that can track the position of the ultrasound transducer, and then the ultrasound probe is gently moved to the next rib;

[0087] 6) If the rib is normal, gently move the ultrasound probe along the axis of the rib and check for the following events: (a) any gaps / steps / smoothness; and (b) any subperiosteal hematoma up to the distal side of the rib;

[0088] 7) Repeat steps 4) to 6) until all 12 pairs of ribs are covered.

[0089] The technical process described herein can be implemented by various means. For example, these technologies can be implemented in hardware, software or a combination thereof. For hardware implementation, the technical process can be implemented in one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described herein or a combination thereof. With software, it is possible to implement by modules (e.g., processes, functions, etc.) that perform the functions described herein. The software code can be stored in a volatile or non-volatile storage medium, and the software code is executed by a processor.

[0090] In addition, standard programming and / or engineering techniques can be used to generate software, firmware, hardware or any combination thereof to control computers or computing components to implement various aspects of the claimed subject matter, thereby implementing various aspects of the claimed subject matter as methods, devices, systems or products. The term "product" as used herein is intended to cover computer programs that can be accessed from any computer-readable device, carrier or medium. For example, computer-readable media can include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical disks (e.g., compact disks (CDs), digital versatile disks (DVDs), etc.), smart cards, and flash memory devices (e.g., cards, sticks, key drives, etc.). Of course, those skilled in the art will recognize that many modifications can be made to this configuration without departing from the scope or spirit of what is described herein.

[0091] The term "detector" (e.g., a first fracture detector and a second fracture detector, "processor," "controller") used in this application refers to a general-purpose processor, a special-purpose processor, a computer processor, or a computer-related entity (which can be hardware, a combination of hardware and software, software, or software in execution). For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As an illustration, an application running on a server and a server can both be components. One or more components can reside in an executing process and / or thread, and a component can be located on one computer and / or distributed between two or more computers.

[0092] What is described above includes examples of one or more embodiments. Of course, it is not possible to describe every possible combination of components or methods for the purpose of describing the aforementioned embodiments, but it will be appreciated by those of ordinary skill in the art that many additional combinations and permutations of the various embodiments are possible. Therefore, the described embodiments are intended to encompass all such changes, modifications and variations that fall within the spirit and scope of the claims. In addition, to the extent that the term "comprising" is used in a specific embodiment or claim, such term is intended to be inclusive in a manner similar to the term "comprising" when interpreted as a transitional word in a claim.

Claims

1. A device (120) for detecting a bone fracture of an object based on an ultrasound image, comprising: A first bone fracture detector (122) configured to: receive a first ultrasound image of a region of the subject and identify bones in the first ultrasound image; It is characterized in that The first fracture detector (122) is further configured to: detect a hematoma adjacent to the identified bone based on the first ultrasound image, identify at least one focal zone within the region based on the detected hematoma, generate focal zone information indicating a location of the at least one focal zone, and guide acquisition of a second ultrasound image based on the generated focal zone information; The apparatus further comprises a second fracture detector (124) configured to: receive the second ultrasound image of the region, and detect a fracture based on the second ultrasound image; and The second ultrasound image has a higher resolution in the at least one focal zone than in a rest of the region, and a resolution of the at least one focal zone in the second ultrasound image is higher than a resolution of the at least one focal zone in the first ultrasound image.

2. The device according to claim 1, wherein: The first fracture detector is further configured to guide the ultrasound image acquisition unit to acquire the second ultrasound image by transmitting an additional ultrasound beam at least toward the at least one focal area, and the second fracture detector is further configured to receive the second ultrasound image from the ultrasound image acquisition unit.

3. The device according to claim 1, wherein: The first fracture detection is also configured to guide the ultrasound image acquisition unit to acquire the second ultrasound image using a second ultrasound setting different from the first ultrasound beam setting used to acquire the first ultrasound image, at least in the at least one focal zone, and the first ultrasound beam setting and the second ultrasound beam setting are different in at least one of the following aspects: (a) transmission frequency; (b) transmission pulse length; (c) steering angle; (d) focal zone.

4. The device according to claim 1, wherein: The first bone fracture detector (122) is also configured to detect a bone fracture based on the first ultrasound image.

5. The device according to claim 1, further comprising a controller (126), which is configured to: switch between a first detection mode and a second detection mode, and control the ultrasound image acquisition unit to acquire the first ultrasound image in the first detection mode, and control the ultrasound image acquisition unit to acquire the second ultrasound image or acquire both the first ultrasound image and the second ultrasound image in the second detection mode.

6. The device according to claim 5, wherein: The controller (126) is also configured to switch between the first detection mode and the second detection mode based on user input received from a user interface (130).

7. The device according to claim 1, wherein: The first bone fracture detector (122) is further configured to receive a plurality of first ultrasound images acquired when the ultrasound transducer is moving relative to the object and first position information about the ultrasound transducer relative to the object corresponding to each of the plurality of first ultrasound images; The first fracture detector (122) is further configured to: identify a bone in each of the plurality of first ultrasound images and detect the presence of a hematoma adjacent to the bone, and identify a subset of the plurality of first ultrasound images in which the hematoma is adjacent to the bone; and The first fracture detector (122) is further configured to generate second position information about the ultrasound transducer relative to the object corresponding to each of the subsets in the plurality of first ultrasound images.

8. The apparatus of claim 7, further comprising a user interface (130) configured to present the second location information.

9. An ultrasound system (100) for detecting a bone fracture in a subject, comprising: An ultrasonic image acquisition unit (110), comprising an ultrasonic transducer, wherein the ultrasonic transducer is configured to: transmit an ultrasonic signal to a region of the object, and receive an ultrasonic echo signal from the region; The device (120) according to claim 1, which is coupled to the ultrasound image acquisition unit; and The ultrasound image acquisition unit is configured to acquire the first ultrasound image and the second ultrasound image.

10. A method for detecting bone fracture of an object based on an ultrasound image, comprising: receiving a first ultrasound image of a region of the object; identifying bones in the first ultrasound image; It is characterized in that detecting a hematoma adjacent to the identified bone based on the first ultrasound image; identifying at least one focal region within the region based on the detected hematoma; generating focus region information indicating a position of the at least one focus region; guiding acquisition of a second ultrasound image of the region based on the generated focal zone information, the second ultrasound image having a higher resolution in the at least one focal zone than in a remaining portion of the region, the resolution of the at least one focal zone in the second ultrasound image being higher than the resolution of the at least one focal zone in the first ultrasound image; receiving the second ultrasound image; and A bone fracture is detected based on the second ultrasound image.

11. The method according to claim 10, wherein: The second ultrasound image is acquired by transmitting an additional ultrasound beam at least toward the at least one focal zone.

12. The method according to claim 10, further comprising: receiving a plurality of first ultrasound images acquired when the ultrasound transducer is moving relative to the object and first position information about the ultrasound transducer relative to the object corresponding to each of the plurality of first ultrasound images; identifying a bone in each of the plurality of first ultrasound images and detecting the presence of a hematoma adjacent to the bone; identifying a subset of the hematomas in the plurality of first ultrasound images that are adjacent to the bone; and Second position information about the ultrasound transducer relative to the object corresponding to each of the subsets in the plurality of first ultrasound images is generated.

13. A computer readable medium comprising executable instructions which, when executed, cause a computer processor to perform the method according to any one of claims 10-12.

Images