Ultrasonic imaging device and method, device and storage medium for detecting B line

Abstract

An ultrasonic imaging device and a method, device and storage medium for detecting B lines. The ultrasonic imaging device comprises an ultrasonic probe (01), a transmitting circuit (02) for stimulating the ultrasonic probe (01) to transmit an ultrasonic beam to the lungs, a receiving circuit (03) for receiving the echo of the ultrasonic beam to obtain an ultrasonic echo signal, a beam synthesis module (04), a processor (05) and a human-computer interaction device (06). The processor (05) processes the ultrasonic echo signal to obtain at least one frame of a lung ultrasonic image, identifies the image signs of the lung ultrasonic image (including at least B lines), calculates the B line coverage percentage and / or B line interval of the lung ultrasonic image whose image sign is the B line, and then sends it to the human-computer interaction device (06) for display. The device and method realize quantitative analysis of B lines and improve the intelligence level of lung ultrasonic examination.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic imaging, and in particular to an ultrasonic imaging device and a method, device and storage medium for detecting B lines. Background Art

[0002] Ultrasound imaging, a medical imaging technique used to image organs and soft tissues in the human body, plays a vital role in clinical medicine. Lung ultrasound is highly valuable in the identification and diagnosis of exudative pulmonary lesions. It demonstrates excellent sensitivity and specificity for the diagnosis of various lung diseases and can even replace chest CT scans in the diagnosis of lung diseases in emergency and critical care medicine. This can save time and costs in clinical practice, potentially saving patients' lives.

[0003] Unlike ultrasound examinations of other parts of the body, lung ultrasound imaging in most cases does not directly reflect the lung tissue, but rather a series of artifacts, which can be defined based on the display characteristics of lung ultrasound images. Ultrasound examinations of normally inflated lungs only detect the pleura. However, as the air content decreases, the echo loss effect between the lungs and surrounding tissues decreases, allowing ultrasound to reflect images of deeper areas to a certain extent, producing typical imaging signs. SUMMARY OF THE INVENTION

[0005] Technical issues

[0006] In recent years, lung ultrasound has gained increasing popularity and application in critical care and emergency medicine. By identifying typical imaging features, it helps medical staff quickly diagnose lung diseases. However, current lung ultrasound examinations rely primarily on manual measurement, which is time-consuming and labor-intensive. Results are significantly affected by the operator's experience and lack intelligence.

[0007] Solution to the problem

[0008] Technical Solutions

[0009] The present invention mainly provides an ultrasonic imaging device and a method, device and storage medium for detecting B lines, so as to improve the intelligence level of lung ultrasonic examination.

[0010] According to a first aspect, an embodiment provides an ultrasonic imaging device, comprising:

[0011] Ultrasound probe;

[0012] a transmitting circuit, configured to stimulate the ultrasound probe to transmit an ultrasound beam toward the lungs;

[0013] A receiving circuit and a beamforming module, configured to receive the echo of the ultrasonic beam and obtain an ultrasonic echo signal;

[0014] a processor configured to process the ultrasonic echo signal to obtain at least one frame of lung ultrasound image; the processor further configured to identify image features of the lung ultrasound image, the image features including at least B-lines, and calculate parameter information of the lung ultrasound image whose image features are B-lines, the parameter information including at least one of a B-line coverage percentage and a B-line interval, the B-line coverage percentage being a percentage of an area occupied by B-lines relative to a lung detection area, and the B-line interval being a distance between adjacent B-lines;

[0015] A human-computer interaction device is connected to the processor and is used to detect user input information and display detection results, wherein the detection results include the parameter information.

[0016] According to a second aspect, an embodiment provides a method for automatically detecting B-lines in a lung ultrasound image, comprising:

[0017] acquiring at least one frame of lung ultrasound image;

[0018] identifying image features of the lung ultrasound image, the image features comprising at least B-lines;

[0019] Calculating parameter information of a lung ultrasound image whose image features are B-lines, the parameter information including at least one of a B-line coverage percentage and a B-line interval, wherein the B-line coverage percentage is a percentage of an area occupied by B-lines to a lung detection area, and the B-line interval is a distance between adjacent B-lines;

[0020] Displays the parameter information.

[0021] According to a third aspect, an embodiment provides a device for automatically detecting B lines in a lung ultrasound image, comprising:

[0022] An acquisition module, configured to acquire at least one frame of lung ultrasound image;

[0023] a determination module, configured to determine a lung detection area of ​​the lung ultrasound image;

[0024] an identification module, configured to identify image signs corresponding to the lung ultrasound image within the lung detection area, wherein the image signs include at least B lines;

[0025] a calculation module, configured to calculate parameter information of a lung ultrasound image whose image features are B-lines, the parameter information including at least one of a B-line coverage percentage and a B-line interval, wherein the B-line coverage percentage is a percentage of an area occupied by B-lines to a lung detection area, and the B-line interval is a distance between adjacent B-lines;

[0026] The display module is used to display the parameter information.

[0027] According to a fourth aspect, an embodiment provides a computer-readable storage medium, which includes a program, and the program can be executed by a processor to implement the method described above.

[0028] Advantageous Effects of the Invention

[0029] Beneficial effects

[0030] According to the ultrasound imaging equipment and the method, device, and storage medium for detecting B lines in the above-mentioned embodiments, since the ultrasound imaging equipment can automatically detect image signs in lung ultrasound images, which image signs at least include B lines, and can automatically calculate the percentage of the area occupied by B lines in the lung detection area and / or the distance between B lines, quantitative analysis of B lines is achieved, thereby improving the intelligence level of lung ultrasound examination.

[0031] Brief description of the accompanying drawings BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a schematic structural diagram of an ultrasonic imaging device according to an embodiment of the present invention;

[0033] Figure 2 Flowchart of a method for automatically detecting B lines in lung ultrasound images according to an embodiment of the present invention;

[0034] Figure 3 This is a flow chart of a method for automatically detecting B lines in lung ultrasound images in a specific embodiment of the present invention;

[0035] Figure 4 A schematic diagram of the structure of a processor in a specific embodiment of the present invention;

[0036] Figure 5 A schematic diagram showing the display effect of a target image and quantitative results in a specific embodiment of the present invention;

[0037] Figure 6 Schematic diagram of the structure of a device for automatically detecting B lines in lung ultrasound images according to an embodiment of the present invention;

[0038] Figure 7 Schematic diagram of another device for automatically detecting B lines in lung ultrasound images according to an embodiment of the present invention;

[0039] Figure 8 FIG. 4 is a schematic structural diagram of another device for automatically detecting B lines in lung ultrasound images according to an embodiment of the present invention.

[0040] Invention Embodiments

[0041] Modes for Carrying Out the Invention

[0042] The present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. The features, operations, or characteristics described in this specification may be combined in any appropriate manner to form various embodiments. The terms "connection" and "coupling" used herein include both direct and indirect connections (couplings), unless otherwise specified.

[0043] When the air content in the lungs decreases, exudates, transudates, collagen, and blood increase lung density, reducing the echo dropout between the lungs and surrounding tissues. This allows ultrasound to image deeper areas to a certain extent. This phenomenon produces vertical mixed echoes known as B-lines. In lung ultrasound images, B-lines appear as discrete vertical reverberation artifacts extending from the pleural line to the bottom of the screen, without dropout, and moving synchronously with lung glide.

[0044] When quantitatively evaluating B-lines, available methods, such as manually counting them, present no significant challenges when the number of B-lines is small. However, as the number increases, the B-lines become increasingly fused and difficult to distinguish, making quantitative evaluation based on the number of B-lines difficult. Furthermore, manual measurement is time-consuming and laborious, relying on operator experience, which hinders the promotion and application of lung ultrasound in critical care settings. Intelligent automatic B-line recognition and quantitative analysis tools that support multiple parameters are expected to improve the efficiency and accuracy of B-line quantitative analysis, promoting a greater role for lung ultrasound in critical care settings.

[0045] In an embodiment of the present invention, the ultrasound imaging device identifies image signs of the acquired lung ultrasound image, where the image signs include at least B lines, and then calculates the number of B lines, the percentage of the area occupied by the B lines in the lung detection area, and / or the distance between the B lines, thereby completing automatic detection and quantitative analysis of the B lines.

[0046] Please refer to Figure 1 , Figure 1 This is a structural schematic diagram of an ultrasonic imaging device provided in an embodiment of the present invention. The ultrasonic imaging device includes an ultrasonic probe 01, a transmitting circuit 02, a receiving circuit 03, a beamforming module 04, a processor 05, and a human-computer interaction device 06. The transmitting circuit 02 and the receiving circuit 03 can be connected to the ultrasonic probe 01 via a transmitting / receiving selection switch 07.

[0047] During ultrasound imaging, the transmitting circuit 02 transmits a delayed, focused transmit pulse of a certain amplitude and polarity to the ultrasound probe 01 via the transmit / receive selector switch 07, stimulating the ultrasound probe 01 to transmit an ultrasound beam toward the target tissue (e.g., an organ, tissue, or blood vessel in a human or animal body). In this embodiment, the ultrasound beam is transmitted toward the lungs. After a certain delay, the receiving circuit 03 receives the echo of the ultrasound beam via the transmit / receive selector switch 07, obtaining an ultrasound echo signal. This echo signal is then sent to the beamforming module 04, which performs focusing, delaying, weighting, and channel summing on the ultrasound echo signal to obtain a beamformed ultrasound echo signal. This beamformed ultrasound echo signal is then sent to the processor 05 for relevant processing to produce the desired ultrasound image or video file composed of the ultrasound image.

[0048] The ultrasound probe 01 typically comprises an array of multiple elements. Each time an ultrasound wave is transmitted, all elements of the ultrasound probe 01, or a portion of all elements, participate in the transmission. Each element, or a portion of elements, participating in the transmission is excited by the transmit pulse and independently transmits an ultrasound wave. The ultrasound waves emitted by these elements superimpose during propagation, forming a composite ultrasound beam that is transmitted to the scan target. In this embodiment of the present invention, this composite ultrasound beam is the ultrasound beam transmitted to the lungs.

[0049] The human-computer interaction device 06 is connected to the processor 05. For example, the processor 05 can be connected to the human-computer interaction device 06 via an external input / output port. The human-computer interaction device 06 can detect user input information. The input information can be, for example, control instructions for the timing of ultrasonic transmission and reception, operation input instructions for editing and annotating ultrasonic images, or other types of instructions. Typically, the operation instructions obtained when the user edits, annotates, measures, or other operation inputs on the ultrasonic image are used to measure the target tissue. The human-computer interaction device 06 can include one or a combination of a keyboard, a mouse, a scroll wheel, a trackball, a mobile input device (such as a mobile device with a touch screen display, a mobile phone, etc.), a multi-function knob, etc. Therefore, the corresponding external input / output port can be a wireless communication module, a wired communication module, or a combination of the two. The external input / output port can also be implemented based on USB, bus protocols such as CAN, and / or wired network protocols.

[0050] The human-computer interaction device 06 also includes a display that can display the ultrasound image obtained by the processor 05. In addition, while displaying the ultrasound image, the display can also provide the user with a graphical interface for human-computer interaction. One or more controlled objects are set on the graphical interface, and the user can use the human-computer interaction device 06 to input operating instructions to control these controlled objects, thereby performing corresponding control operations. For example, an icon is displayed on the graphical interface, and the human-computer interaction device can be used to operate the icon to perform a specific function, such as annotating the ultrasound image. In actual applications, the display can be a touch screen display. In addition, the display in this embodiment can include a single display or multiple displays.

[0051] In an embodiment of the present invention, processor 05 is configured to process the ultrasonic echo signals obtained by beamforming module 04 to obtain at least one frame of lung ultrasound image. Processor 05 is further configured to identify image features of the lung ultrasound image, wherein the image features include at least B-lines, and calculate parameter information of the lung ultrasound image with B-lines as image features. The parameter information includes at least one of the number of B-lines, B-line coverage percentage, and B-line spacing. The B-line coverage percentage is the percentage of the area occupied by B-lines in the lung detection area, and the B-line spacing is the distance between adjacent B-lines. Human-computer interaction device 06 displays the detection results on a display, which include the parameter information calculated by processor 05.

[0052] Based on the ultrasound imaging device of the above embodiment, the embodiment of the present invention also provides a method for automatically detecting B lines in lung ultrasound images, the flow chart of which can be found in Figure 2 , the method may include the following steps:

[0053] Step 101: Acquire at least one frame of lung ultrasound image, for example, acquire a video file including one or more frames of ultrasound image.

[0054] The processor 05 obtains at least one frame of lung ultrasound image acquired in real time by the ultrasound probe 01 , or the processor 05 may also read at least one frame of lung ultrasound image from a storage device.

[0055] Step 102: Identify image features.

[0056] After acquiring at least one lung ultrasound image frame, processor 05 identifies image features of each lung ultrasound image frame, where the image features include at least B-lines. In one embodiment, processor 05 may identify only lung ultrasound images with B-lines from the acquired lung ultrasound images based on a target detection algorithm.

[0057] Step 103: Calculate parameter information of line B.

[0058] After the processor 05 identifies the B-lines in the lung ultrasound image, it calculates parameter information of the lung ultrasound image whose image features are B-lines. The parameter information includes at least one of a B-line coverage percentage and a B-line interval, wherein the B-line coverage percentage is the percentage of the area occupied by the B-lines to the lung detection area, and the B-line interval is the distance between adjacent B-lines.

[0059] For example, when calculating the B-line coverage percentage of a frame of lung ultrasound image, the lung detection area of ​​the frame of lung ultrasound image can be determined first, and then the percentage of the area occupied by the B-line in the lung detection area can be calculated based on the identified B-line; wherein, the lung detection area can be determined according to the deep learning method, or the pleural line position of the lung ultrasound image can be identified first, and then the far-field area of ​​the pleural line position is used as the lung detection area of ​​the lung ultrasound image. The area occupied by the B-line refers to the area of ​​all B-lines in the lung ultrasound image, which can be determined during the B-line identification process. Specifically, the processor 05 can determine the identified pixel points in the area belonging to the B-line, and then determine the identified area belonging to the B-line as the area occupied by the B-line. For example, the processor 05 can determine the pixel points belonging to the B-line and use the area defined by each pixel point as the area belonging to the B-line.

[0060] For example, when calculating the B-line interval in a frame of lung ultrasound image, the position of the pleural line of the frame of lung ultrasound image can be determined first, and then the distance between adjacent B-lines at the pleural line position can be calculated based on the identified B-lines. Alternatively, the distance between adjacent B-lines at a preset distance from the pleural line position can be calculated.

[0061] Step 104: Display parameter information.

[0062] The processor 05 sends the calculated parameter information to the human-computer interaction device 06 for display.

[0063] The ultrasound imaging device and method for automatically detecting B-lines in lung ultrasound images provided by embodiments of the present invention are capable of automatically detecting image features in lung ultrasound images and calculating the B-line coverage percentage and / or B-line interval for lung ultrasound images where the image features are B-lines. This enables intelligent identification of B-lines and quantitative analysis of multiple parameters, thereby improving the intelligence level of lung ultrasound examinations. By calculating the B-line coverage percentage, the problem of being unable to quantitatively assess B-lines by number due to the difficulty of distinguishing between a large number of B-lines due to their fusion can be avoided. This overcomes the problems of low manual measurement efficiency and the susceptibility of test results to human factors, thereby improving measurement efficiency and the accuracy of test results.

[0064] In order to more clearly reflect the purpose of the present invention, further examples are given based on the above embodiments.

[0065] Please refer to Figure 3 , Figure 3A specific method for automatically detecting B lines in lung ultrasound images is provided. In this embodiment, the structure of the processor 05 can be seen in Figure 4 , which may include an acquisition unit 51, an image selection unit 52, an image analysis unit 53, a result selection unit 54 and a scoring unit 55. Specifically, the method may include the following steps:

[0066] Step 201: Acquire at least one frame of lung ultrasound image.

[0067] The processor 05 obtains at least one frame of lung ultrasound image acquired in real time by the ultrasound probe 01 through the acquisition unit 51 , or the acquisition unit 51 may also read at least one frame of lung ultrasound image from a storage device.

[0068] Step 202: Screen out images to be analyzed.

[0069] After processor 05 acquires at least one frame of lung ultrasound images, it inputs these lung ultrasound images into image selection unit 52. Image selection unit 52 then selects images to be analyzed from these lung ultrasound images. These images to be analyzed are lung ultrasound images with pathological features. This image screening can eliminate useless images, such as those without image information, images not showing lung features, and blurred images, thereby improving detection efficiency and reducing false detections.

[0070] Step 203: Identify image features of the image to be analyzed.

[0071] After the image selection unit 52 selects the images to be analyzed, it inputs these images to be analyzed into the image analysis unit 53. The image analysis unit 53 identifies the image features of each frame of the images to be analyzed. The identified image features at least include B lines.

[0072] Specifically, the image analysis unit 53 first determines the lung detection region for each frame of the image to be analyzed. In one embodiment, the image analysis unit 53 can determine the lung detection region for each frame of the image to be analyzed using a deep learning method. That is, a large number of lung regions can be pre-calibrated, and then the machine can be trained to identify these regions using an object detection algorithm, such as a Faster Regional Convolutional Neural Network (Faster RCNN) algorithm. In another embodiment, the image analysis unit 53 can also determine the lung detection region for each frame of the image to be analyzed using an image processing method. For example, the pleural line position of the image to be analyzed can be first identified, and then the far-field region of the pleural line position can be used as the lung detection region for the image to be analyzed. The pleural line position can be determined based on the identified near-field highlighted horizontal linear features, which can also be achieved using a deep learning method. In addition, a bat sign sometimes appears in lung ultrasound images, which includes the thoracic line. Therefore, the image analysis unit 53 can also use the bat sign of the image to be analyzed to determine the lung detection region. That is, when the bat sign is identified, the pleural line position can be determined from the bat sign.

[0073] After determining the lung detection region, the image analysis unit 53 identifies image features within the lung detection region. Common image features in lung ultrasound images include A-lines, B-lines, bat sign, shore sign, pleural sliding sign, and debris sign. Detected image features may include B-lines and pulmonary consolidation. In one embodiment, the image analysis unit 53 may identify image features within the lung detection region using deep learning methods. Specifically, a large number of pathologies are identified and then trained to identify them using an object detection algorithm, such as the Faster RCNN algorithm. In another embodiment, the image analysis unit 53 may also use image processing methods to identify image features within the lung detection region. Because B-lines are discrete vertical reverberation artifacts extending from the pleural line to the bottom of the image display screen without loss, this characteristic can be used to detect vertical linear features in the direction of the acoustic beam line within the lung detection region to obtain B-lines. Identification of linear features can be achieved through methods such as template matching. In practical applications, B-lines can be classified into single B-lines and diffuse B-lines based on their width.

[0074] Step 204: Determine a lung ultrasound image with B-lines.

[0075] After the image analyzing unit 53 identifies the image features of each frame of the image to be analyzed, it determines the lung ultrasound image with B-lines from the images to be analyzed according to the image features.

[0076] Step 205: Calculate parameter information of line B.

[0077] After the image analysis unit 53 determines the lung ultrasound image with B-lines, for each frame of the lung ultrasound image with B-lines, the B-line coverage percentage and / or B-line interval is calculated based on the identified B-lines to achieve quantitative analysis of the B-lines. When calculating the B-line coverage percentage, the image analysis unit 53 can first determine the area occupied by the B-lines, and then calculate the percentage of the area occupied by the B-lines in the corresponding lung detection area. The area occupied by the B-lines refers to the area occupied by all B-lines in the lung ultrasound image, which can be considered as the area defined by the position of the B-line pixel points. The B-line pixel points can be obtained in the process of identifying the B-lines. Based on this, when determining the area occupied by the B-lines, the image analysis unit 53 can first determine the identified area pixel points belonging to the B-lines within the lung detection area, and determine the area belonging to the B-lines as the area occupied by the B-lines.

[0078] When calculating the B-line interval, the image analysis unit 53 may first determine the pleural line position in the lung ultrasound image and then calculate the distance between adjacent B-lines at the pleural line position based on the identified B-lines. Alternatively, the image analysis unit 53 may calculate the distance between adjacent B-lines at a preset distance from the pleural line position. Calculating the B-line interval can assist medical personnel in assessing lung conditions.

[0079] In practical applications, the image analysis unit 53 can also calculate the number of B-lines based on the identified B-lines. When evaluating the number of B-lines, when the number of B-lines is small, each B-line can be clearly distinguished. However, as the number of B-lines increases, the B-lines merge and become difficult to distinguish. In this case, to obtain a more accurate number of B-lines, the percentage of B-lines can be calculated. Therefore, in one embodiment, after the image analysis unit 53 identifies B-lines, it can calculate the number of B-lines when each B-line can be clearly distinguished (a small number of B-lines), or it can calculate the B-line coverage percentage, or both. When each B-line cannot be clearly distinguished (a large number of B-lines merge and become difficult to distinguish), it can calculate the B-line coverage percentage, using the B-line coverage percentage to reflect the number of B-lines.

[0080] Step 206: Determine the target image.

[0081] After the image analysis unit 53 calculates the B-line parameter information, the result selection unit 54 of the processor 05 may select at least one frame from the lung ultrasound image as a target image based on a preset rule. For example, in a specific embodiment, the preset rule may be the frame with the largest number of B-lines or the frame with the largest B-line coverage percentage. The result selection unit 54 may select the frame with the largest number of B-lines from the lung ultrasound image as the target image, or the frame with the largest B-line coverage percentage as the target image. In actual applications, the selection criteria for selecting the target image may be user-configured.

[0082] Step 207: Score the target image.

[0083] After the result selection unit 54 determines the target image, the scoring unit 55 scores the selected at least one frame of the target image to obtain a scoring result for each frame of the target image, which reflects the correlation between the target image and the associated disease.

[0084] The scoring format for scoring the target image may be various forms consisting of at least one of numbers, letters, and text, and the score may be determined based on one or more of the calculated parameter information. For example, one scoring rule may be: when the target image shows no abnormality or there are two or fewer clear B-lines, the score result is determined to be N or 0; when there are three or more clear B-lines in the target image, the score result is determined to be B1 or 1; when the B-line spacing in the target image is less than a preset value (diffuse B-lines), the score result is determined to be B2 or 2; when the image signs of the target image represent pulmonary consolidation, the score result is determined to be C or 3.

[0085] In practical applications, the scoring unit 55 can not only obtain the scoring result at a single position of the lung, but also calculate the sum of the scores of the target images corresponding to each scanning position of the lung, that is, add up the scores of each scanning position to obtain the scoring result of the entire lung.

[0086] Step 208: Display the target image and quantitative results.

[0087] After obtaining the scoring result for the target image, processor 05 may transmit the selected target image, its corresponding parameter information, and the scoring result to human-computer interaction device 06 for display. In one embodiment, human-computer interaction device 06 may also highlight at least one of the lung detection area, the B-line, and the annotation line between the B-lines in the target image.

[0088] like Figure 5 Shown is a schematic diagram of the display effect, in which the dotted line indicates the lung detection area, the vertical solid line in the area is the detected B-line, the upper right area shows the quantitative analysis result of the target image, which may include the scoring result, the number of B-lines and the B-line coverage percentage, and the table in the lower right area indicates the B-line interval.

[0089] The method provided in this embodiment for automatically detecting B-lines in lung ultrasound images first screens images with pathological features from acquired lung ultrasound images to be analyzed, thereby eliminating useless images, improving detection efficiency and reducing false positives. A lung detection region is then determined for each frame of the image to be analyzed, and image features are identified within the lung detection region, including at least B-lines. Based on the identified image features, lung ultrasound images with B-lines are identified, and at least one of the B-line coverage percentage, B-line interval, and number of B-lines corresponding to each of these lung ultrasound images is calculated, thereby achieving automatic detection and quantitative analysis of B-lines. Subsequently, at least one frame of the lung ultrasound image can be selected as a target image based on the calculated B-line parameter information, and the target image is scored. Finally, the target image, its corresponding parameter information, and the scoring result are displayed to provide a unified and intuitive quantitative analysis result, assisting medical personnel in jointly determining the patient's lung condition in conjunction with other examination results. This method overcomes the problem of low manual measurement efficiency while avoiding the influence of human factors on the examination results, improving the intelligence level of lung ultrasound examination, and also improving measurement efficiency and the accuracy of examination results.

[0090] In the method of the above embodiment, the processor 05 selects images to be analyzed that have pathological features during image screening, identifies image features of the images to be analyzed, and then determines lung ultrasound images that have B-lines based on the identified image features, thereby obtaining lung ultrasound images that have B-lines. Alternatively, in another specific embodiment, the processor 05 may select only images that have B-lines during image screening, i.e., directly identify lung ultrasound images that have B-lines from at least one acquired lung ultrasound image frame. For example, the processor 05 may identify lung ultrasound images that have B-lines from the acquired lung ultrasound image based on a target detection algorithm.

[0091] The above embodiment uses the example of determining and scoring a target image as an example. The human-computer interaction device 06 displays the target image, its corresponding parameter information, and the scoring result. In actual applications, scoring the target image is not required. After the processor calculates the parameter information and selects at least one frame of lung ultrasound images as the target image according to preset rules, the human-computer interaction device 06 directly displays the target image and its corresponding parameter information.

[0092] In the method of the above embodiment, the detection of symptoms is achieved using a fully automatic method. In practical applications, a combination of automatic and manual methods can also be used, such as automatically identifying B-lines and manually marking lung ultrasound images that do not have B-lines. Specifically, the image analysis unit 53 of the processor 05 can determine, based on the identified image features of the image to be analyzed, lung ultrasound images with image features that are not B-lines from the image to be analyzed. The human-computer interaction device 06 detects the user's operation instruction to mark the lung ultrasound images with image features that are not B-lines and sends the operation instruction to the processor 05. In this case, the processor 05 will mark the lung ultrasound images with image features that are not B-lines according to the operation instruction.

[0093] like Figure 6 As shown, a device for automatically detecting B lines in lung ultrasound images is provided, the device comprising an acquisition module 61, a determination module 62, an identification module 63, a calculation module 64, and a display module 65. The acquisition module 61 is used to acquire at least one frame of lung ultrasound image. The determination module 62 is used to determine the lung detection area of ​​the lung ultrasound image acquired by the acquisition module 61; for example, the determination module 62 may determine the lung detection area of ​​the lung ultrasound image based on a deep learning method, or the determination module 62 may first identify the pleural line position of the lung ultrasound image and then use the far-field area of ​​the pleural line position as the lung detection area of ​​the frame of lung ultrasound image. The identification module 63 is used to identify image signs corresponding to the lung ultrasound image within the lung detection area determined by the determination module 62, wherein the image signs include at least B lines. Calculation module 64 is configured to calculate parameter information for lung ultrasound images with B-line features. This parameter information includes at least one of B-line coverage percentage and B-line spacing. The B-line coverage percentage is the percentage of the area occupied by B-lines relative to the lung detection area, and the B-line spacing is the distance between adjacent B-lines. When the parameter information is B-line spacing, calculation module 64 is specifically configured to determine the pleural line position in the lung ultrasound image with B-lines as the image feature, and then calculate the distance between adjacent B-lines at the pleural line position based on the identified B-lines. When the parameter information is B-line coverage percentage, calculation module 64 is configured to calculate the percentage of the area occupied by the identified B-lines relative to the corresponding lung detection area based on the identified B-lines, thereby obtaining the B-line coverage percentage. Specifically, calculation module 64 may first determine the identified area belonging to the B-line, then define the area belonging to the B-line as the B-line occupied area, and then calculate the percentage of the area occupied by the B-line relative to the corresponding lung detection area. Display module 65 is configured to display the parameter information calculated by calculation module 64.

[0094] based on Figure 6 ,like Figure 7As shown in FIG, another device for automatically detecting B lines in lung ultrasound images is provided, which includes an acquisition module 61, a determination module 62, an identification module 63, a calculation module 64, a display module 65, a selection module 66 and a scoring module 67. The functions of the acquisition module 61, the determination module 62, the identification module 63 and the calculation module 64 are similar to those of the Figure 6 The one-to-one correspondence in the image is the same. The selection module 66 is configured to select at least one frame of lung ultrasound images acquired by the acquisition module 61 as a target image based on preset rules. For example, a frame of image may be selected as the target image. The calculation module 64 may also calculate the number of B-lines. The preset rule may be the number of B-lines or the maximum B-line coverage percentage. In this case, the selection module 66 may select the frame of lung ultrasound images acquired by the acquisition module 61 with the maximum number of B-lines or the maximum B-line coverage percentage as the target image based on the calculation result of the calculation module 64. The display module 65 is configured to display the target image determined by the selection module 66 and its corresponding parameter information. In one embodiment, after the selection module 66 determines the target image, the scoring module 67 may also score the selected at least one frame of target image to obtain a scoring result that can reflect the relevance of the target image to the associated condition. In this case, the display module 65 may simultaneously display the target image, its corresponding parameter information, and the scoring result. Furthermore, the display module 65 may also highlight at least one of the lung detection area, B-lines, and the annotation lines between B-lines in the target image.

[0095] based on Figure 6 ,like Figure 8 As shown, another device for automatically detecting B lines in lung ultrasound images is provided, which includes an acquisition module 61, a determination module 62, an identification module 63, a calculation module 64, a display module 65, a screening module 68, a non-B line determination module 69, and a marking module 60. The acquisition module 61 is used to acquire at least one frame of lung ultrasound image. The screening module 68 is used to screen out an image to be analyzed from the lung ultrasound image acquired by the acquisition module 61, and the image to be analyzed is a lung ultrasound image with pathological characteristics. Figure 6 The difference between the devices is that Figure 8 In the device, the determination module 62 is used to determine the lung detection area of ​​the image to be analyzed screened by the screening module 68, and the recognition module 63 is used to identify the image signs of the image to be analyzed in the lung detection area, and determine the lung ultrasound image with B lines from the image to be analyzed based on the image signs. The function of the calculation module 64 is the same as Figure 6The non-B-line determination module 69 is used to determine, from the image to be analyzed, the lung ultrasound image with the image sign being a non-B-line, based on the image sign of the image to be analyzed identified by the recognition module 63. The marking module 60 is used to receive an operation instruction to mark the lung ultrasound image with the image sign being a non-B-line, and mark the lung ultrasound image with the image sign being a non-B-line determined by the non-B-line determination module 69 according to the operation instruction. In addition, the device may also include the following: Figure 7 The selection module 66 and the scoring module 67 in the device shown are configured such that the selection module 66 is configured to select at least one frame of image as a target image from the images to be analyzed screened by the screening module 68 according to a preset rule.

[0096] This document is described with reference to various exemplary embodiments. However, those skilled in the art will recognize that changes and modifications may be made to the exemplary embodiments without departing from the scope of this document. For example, the various operational steps and components used to perform the operational steps may be implemented in different ways (e.g., one or more steps may be deleted, modified, or incorporated into other steps) depending on the specific application or considering any number of cost functions associated with the operation of the system.

[0097] In addition, as will be appreciated by those skilled in the art, the principles of this document may be embodied in a computer program product on a computer-readable storage medium pre-loaded with computer-readable program code. Any tangible, non-transitory computer-readable storage medium may be used, including magnetic storage devices (hard disks, floppy disks, etc.), optical storage devices (CD-ROMs, DVDs, Blu-Ray disks, etc.), flash memory, and / or the like. These computer program instructions may be loaded onto a general-purpose computer, a special-purpose computer, or other programmable data processing device to form a machine, such that the instructions executed on the computer or other programmable data processing device may generate a device that implements a specified function. These computer program instructions may also be stored in a computer-readable memory, which may instruct the computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory may form an article of manufacture, including an implementation device that implements the specified function. The computer program instructions may also be loaded onto a computer or other programmable data processing device, thereby causing the computer or other programmable device to execute a series of operational steps to produce a computer-implemented process, such that the instructions executed on the computer or other programmable device may provide the steps for implementing the specified function.

[0098] Although the principles of this invention have been shown in various embodiments, many modifications of structure, arrangement, proportion, elements, materials and components that are particularly suitable for specific environments and operational requirements can be used without departing from the principles and scope of this invention. The above modifications and other changes or amendments are intended to be included within the scope of this invention.

[0099] The foregoing detailed description has been described with reference to various embodiments. However, those skilled in the art will recognize that various modifications and changes can be made without departing from the scope of this disclosure. Therefore, the present disclosure will be considered in an illustrative rather than a restrictive sense, and all such modifications will be included within its scope. Similarly, the advantages, other advantages and solutions to the problems of the various embodiments have been described above. However, the benefits, advantages, solutions to the problems and any elements that can produce these, or make them more specific, should not be interpreted as critical, required or necessary. The term "comprising" and any other variants used in this article are all non-exclusive inclusions, so that a process, method, article or device that includes a list of elements includes not only these elements, but also other elements that are not explicitly listed or do not belong to the process, method, system, article or device. In addition, the term "coupled" and any other variants used in this article refer to physical connections, electrical connections, magnetic connections, optical connections, communication connections, functional connections and / or any other connections.

[0100] Those skilled in the art will recognize that many changes can be made to the details of the above embodiments without departing from the basic principles of the invention. Therefore, the scope of the invention should be determined from the following claims.

Claims

1. An ultrasonic imaging device, characterized in that: include: Ultrasound probe; A transmitting circuit, used for stimulating the ultrasound probe to transmit an ultrasound beam toward the lungs; A receiving circuit and a beamforming module, used for receiving the echo of the ultrasonic beam and obtaining an ultrasonic echo signal; A processor is used to process the ultrasonic echo signal to obtain at least one frame of lung ultrasound image; the processor is also used to identify image signs of the lung ultrasound image, the image signs at least include B lines, and calculate parameter information of the lung ultrasound image with image signs as B lines, the parameter information includes B line coverage percentage and the number of B lines, the B line coverage percentage is the percentage of the area occupied by B lines to the lung detection area; wherein the processor includes an image analysis unit, after identifying the B lines, if the number of B lines is determined to be small, the image analysis unit calculates the number of B lines, if the number of B lines is determined to be large so that the B lines are merged and difficult to distinguish, the B line coverage percentage is calculated to reflect the number of B lines; A human-computer interaction device is connected to the processor and is used to detect user input information and display detection results, wherein the detection results include the parameter information.

2. The ultrasonic imaging device according to claim 1, characterized in that: The parameter information includes a B-line interval, and the processor is used to calculate the B-line interval: Determine the location of the pleural line; The distance between adjacent B lines at the pleural line position is calculated based on the identified B lines.

3. The ultrasonic imaging device according to claim 1, characterized in that: The processor is used to filter out an image to be analyzed from the at least one frame of lung ultrasound image, where the image to be analyzed is a lung ultrasound image with pathological characteristics, identify image signs of the image to be analyzed, and determine a lung ultrasound image with a B line from the image to be analyzed based on the image signs.

4. The ultrasonic imaging device according to claim 3, characterized in that: The processor also determines, from the image to be analyzed, a lung ultrasound image whose image sign is a non-B-line according to the image sign of the image to be analyzed identified; the human-computer interaction device is also used to detect an operation instruction of a user to mark the lung ultrasound image whose image sign is a non-B-line, and send the operation instruction to the processor; The processor is further configured to mark the lung ultrasound image whose image sign is non-B line according to the operation instruction.

5. The ultrasonic imaging device according to claim 1, characterized in that: The processor is configured to identify a lung ultrasound image having a B line from the at least one frame of lung ultrasound image.

6. The ultrasonic imaging device according to any one of claims 1 to 5, characterized in that: The processor is configured to: determining a lung detection region of a lung ultrasound image; Image features are identified within the lung detection region.

7. The ultrasonic imaging device according to claim 6, characterized in that: The processor determines a lung detection region of a lung ultrasound image according to a deep learning method.

8. The ultrasonic imaging device according to claim 6, characterized in that: When determining the lung detection area of ​​the lung ultrasound image, the processor is used to: identify the pleural line position of the lung ultrasound image, and use the far-field area of ​​the pleural line position as the lung detection area of ​​the lung ultrasound image.

9. The ultrasonic imaging device according to claim 6, characterized in that: The parameter information includes a B-line coverage percentage, and the processor is used, when calculating the B-line coverage percentage, to calculate, based on the identified B-line, a percentage of an area occupied by the B-line to a corresponding lung detection area.

10. The ultrasonic imaging device according to claim 9, characterized in that: The processor is used to determine the identified area belonging to the B line as the area occupied by the B line.

11. The ultrasonic imaging device according to claim 6, characterized in that: The processor identifies image signs in the lung detection area based on a deep learning method.

12. The lung ultrasound imaging device according to claim 6, characterized in that: The image sign is a B line, and the processor is used to detect vertical linear features of the acoustic beam line direction in the lung detection area to obtain corresponding image signs.

13. The ultrasonic imaging device according to claim 1, characterized in that: After calculating the parameter information, the processor is further used to select at least one frame of image from the lung ultrasound image as a target image according to a preset rule; The human-computer interaction device is used to display the target image and its corresponding parameter information.

14. The ultrasonic imaging device according to claim 1, characterized in that: After calculating the parameter information, the processor is also used to select at least one frame of image from the lung ultrasound image as a target image according to preset rules, and score the selected at least one frame of target image to obtain a scoring result, which reflects the correlation between the target image and the associated disease.

15. The ultrasonic imaging device according to claim 13 or 14, characterized in that: The parameter information also includes the number of B lines, the target image is a frame of image, and the preset rules include: the largest number of B lines or the largest B line coverage percentage.

16. The ultrasonic imaging device according to claim 14, characterized in that: The processor is also used to calculate the sum of the scores of the target images corresponding to each scanning position of the lung to obtain a score result for the entire lung.

17. The ultrasonic imaging device according to claim 14, characterized in that: The human-computer interaction device is used to display the target image and its corresponding parameter information and scoring results.

18. The ultrasonic imaging device according to claim 13 or 14, characterized in that: The human-computer interaction device is also used to highlight at least one of the lung detection area, the B line and the annotation line of the B line interval in the target image.

19. The ultrasonic imaging device according to claim 1, characterized in that: The processor is also used to read at least one frame of lung ultrasound image from a storage device.

20. A method for automatically detecting B lines in lung ultrasound images, characterized in that: include: Acquiring at least one frame of lung ultrasound image; identifying image signs of the lung ultrasound image, wherein the image signs at least include B lines; Calculating parameter information of a lung ultrasound image whose image sign is a B-line, the parameter information including a B-line coverage percentage and the number of B-lines, the B-line coverage percentage being the percentage of an area occupied by B-lines to a lung detection area; wherein, after identifying the B-lines, if it is determined that the number of B-lines is small, the number of B-lines is calculated, and if it is determined that the number of B-lines is large and the B-lines are merged and difficult to distinguish, the B-line coverage percentage is calculated to reflect the number of B-lines; Displays the parameter information.

21. The method of claim 20, wherein: The parameter information includes the B-line interval, and the parameter information of the lung ultrasound image whose calculated image sign is the B-line includes: Determine the location of the pleural line on lung ultrasound images with the imaging sign of B-lines; The distance between adjacent B lines at the pleural line position is calculated based on the identified B lines.

22. The method of claim 20, wherein: The identifying image signs of the lung ultrasound image comprises: Screening out an image to be analyzed from the lung ultrasound images, wherein the image to be analyzed is a lung ultrasound image having pathological characteristics; An image sign of the image to be analyzed is identified, and a lung ultrasound image with a B line is determined from the image to be analyzed according to the image sign.

23. The method of claim 22, wherein: Also includes: According to the identified image signs of the image to be analyzed, determining a lung ultrasound image whose image sign is a non-B-line from the image to be analyzed; An operation instruction for marking a lung ultrasound image whose image sign is a non-B line is received, and the lung ultrasound image whose image sign is a non-B line is marked according to the operation instruction.

24. The method of claim 20, wherein: The identifying image signs of the lung ultrasound image comprises: A lung ultrasound image having a B line is identified from the lung ultrasound image based on a target detection algorithm.

25. The method according to any one of claims 20 to 24, characterized in that Signs that can be identified on lung ultrasound images include: determining a lung detection region of a lung ultrasound image; Image features are identified within the lung detection region.

26. The method of claim 25, wherein: The step of determining a lung detection area of ​​a lung ultrasound image comprises: determining a lung detection region of a lung ultrasound image based on a deep learning method; or, The pleural line position of the lung ultrasound image is identified, and the far-field area of ​​the pleural line position is used as the lung detection area of ​​the frame of the lung ultrasound image.

27. The method of claim 25, wherein: The parameter information includes the B-line coverage percentage, and the parameter information of the lung ultrasound image whose image sign is the B-line includes: The percentage of the area occupied by the B line to the corresponding lung detection area is calculated based on the identified B line.

28. The method of claim 27, wherein: The calculating, based on the identified B-line, the percentage of the area occupied by the B-line to the corresponding lung detection area includes: Determine the identified area belonging to the B line and determine it as the area occupied by the B line; Calculate the percentage of the area occupied by the B line to the corresponding lung detection area.

29. The method of claim 20, wherein: After calculating the parameter information of the lung ultrasound image whose image sign is a B line, the method further includes: Selecting at least one frame of image from the lung ultrasound image as a target image according to a preset rule; The target image and its corresponding parameter information are displayed.

30. The method of claim 20, wherein: After calculating the parameter information of the lung ultrasound image whose image sign is a B line, the method further includes: The selected at least one frame of the target image is scored to obtain a scoring result, wherein the target image is selected from the lung ultrasound image according to a preset rule, and the scoring result reflects the correlation between the target image and the associated disease.

31. The method according to claim 29 or 30, characterized in that The parameter information also includes the number of B lines, the target image is a frame of image, and the preset rules include: the largest number of B lines or the largest B line coverage percentage.

32. The method of claim 30, wherein: Also includes: The sum of the scores of the target images corresponding to each scanning position of the lung is calculated to obtain the score result of the entire lung.

33. The method of claim 30, wherein: When displaying the parameter information, the method further includes: The target image and its corresponding scoring result are displayed synchronously.

34. The method according to claim 29 or 30, characterized in that Also includes: At least one of a lung detection area, a B line, and a label line of a B line interval is highlighted in the target image.

35. An ultrasonic imaging device, characterized in that: include: Ultrasound probe; A transmitting circuit, used for stimulating the ultrasonic probe to transmit an ultrasonic beam to the lungs; A receiving circuit and a beamforming module, used for receiving the echo of the ultrasonic beam and obtaining an ultrasonic echo signal; A processor, configured to process the ultrasonic echo signal to obtain a video file including multiple frames of lung ultrasound images; the processor is further configured to determine a lung detection area of ​​the lung ultrasound image, and to identify image signs of the lung ultrasound image in the lung detection area, wherein the image signs include at least B lines; the processor is further configured to determine quantitative parameter information of the lung ultrasound image whose image signs are B lines, wherein the quantitative parameter information includes a B-line coverage percentage and the number of B lines; wherein the processor includes an image analysis unit, and after identifying the B lines, the image analysis unit calculates the number of B lines if it is determined that the number of B lines is small, and calculates the B-line coverage percentage to reflect the number of B lines if it is determined that the number of B lines is large and the B lines are difficult to distinguish due to mutual fusion; A human-computer interaction device is connected to the processor and is used to detect user input information and display detection results, wherein the detection results include the parameter information.

36. The ultrasonic imaging device according to claim 35, characterized in that The processor is used to filter out an image to be analyzed from multiple frames of lung ultrasound images in the video file, where the image to be analyzed is a lung ultrasound image with pathological characteristics, identify image signs of the image to be analyzed, and determine a lung ultrasound image with a B line from the image to be analyzed based on the image signs.

37. The ultrasonic imaging device according to claim 35, characterized in that When determining the lung detection area of ​​the lung ultrasound image, the processor is used to: identify the pleural line position of the lung ultrasound image, and use the far-field area of ​​the pleural line position as the lung detection area of ​​the lung ultrasound image.

38. The ultrasonic imaging device according to claim 35, characterized in that The parameter information includes a B-line coverage percentage, and the processor is used, when determining the B-line coverage percentage, to calculate, based on the identified B-line, a percentage of an area occupied by the B-line to a corresponding lung detection area.

39. The ultrasonic imaging device according to claim 38, characterized in that The processor is used to determine the identified area belonging to the B line and determine it as the area occupied by the B line.

40. The ultrasonic imaging device according to claim 35, characterized in that The parameter information includes a B-line interval, and the processor is used to: Determine the location of the pleural line; The distance between adjacent B lines at the pleural line position is calculated based on the identified B lines, or the distance between adjacent B lines at a preset position from the pleural line is calculated based on the identified B lines.

41. The ultrasonic imaging device according to claim 35, characterized in that: The parameter information includes the number of B lines, and the processor is used to: count the number of B lines identified in the lung detection area of ​​the lung ultrasound image when determining the number of B lines.

42. A computer-readable storage medium, characterized in that: A program is included, which can be executed by a processor to implement the method as claimed in any one of claims 20 to 34.

Images