Systems and methods for automated ultrasound image registration based on quality scores

Abstract

A system and method for automatically recording ultrasound clips is based on quality scores of ultrasound images in a sequence of ultrasound image frames. The ultrasound imaging system includes a probe for capturing ultrasound images, an image buffer for storing the sequence of image frames, a quality buffer for storing the sequence of quality scores, and a computing subsystem for automatically recording an ultrasound clip when a quality score in the quality buffer corresponding to a set of consecutive image frames in the image buffer is greater than or equal to a first quality threshold and the set of consecutive image frames is at least a first predetermined size. Additionally, a smart capture function automatically records ultrasound clips including alternative sets of consecutive image frames having quality scores greater than or equal to a second quality threshold that are less than the first quality threshold and meet a second predetermined size.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This international application claims priority to co-pending U.S. patent application Ser. No. 17 / 885,448, entitled "SYSTEMS AND METHODS FOR AUTOMATED ULTRASOUND IMAGE RECORDING BASED ON QUALITY SCORES," filed Aug. 10, 2022, which is incorporated herein by reference in its entirety.

[0002] In the event that this application conflicts with a document incorporated by reference, this application will control. [Background technology]

[0003] FIELD OF THE INVENTION FIELD OF THE DISCLOSURE The present disclosure relates generally to ultrasound imaging systems and methods, and more particularly to systems and methods for automatically recording clips of ultrasound images in memory.

[0004] Description of Related Art Ultrasound imaging is typically performed in a clinical setting by trained ultrasound professionals. For diagnostic ultrasound imaging, certain views of organs or other tissues or body features (such as fluids, bones, joints, etc.) are clinically significant. Such views may be defined by clinical criteria as views that a sonographer should acquire, depending on the target organ, diagnostic objective, etc.

[0005] The image quality of an acquired ultrasound image varies depending on various factors, including, for example, probe positioning, imaging parameters (e.g., depth, gain, etc.), etc. For clinical use (e.g., for diagnosis), ultrasound images should generally have adequate image quality. Clinicians generally require significant training to assess the diagnostic quality of ultrasound images. Such images may be obtained in real time during image acquisition or may have been acquired previously. In both cases, clinicians must assess the level of diagnostic quality of the ultrasound image. Similarly, in training and education environments, expert ultrasound users are required to grade the diagnostic quality of images acquired by students and novice users, which can be very time-consuming for ultrasound professionals.

[0006] Furthermore, significant training is generally required for clinicians to be able to recognize anatomical structures present in ultrasound images, which is particularly difficult during real-time ultrasound image acquisition, where the ultrasound image is continually changing as the position and orientation of the probe moves relative to the organ or body feature of interest.

[0007] While conventional ultrasound imaging systems may be suitable for most patients in a hospital or similar clinical environment, such systems require significant training to operate and properly capture clinically desirable views. Such training requirements also extend to the evaluation of ultrasound images within the sequence of image frames comprising a video clip to ensure that the images within the video clip properly capture the clinically desirable views. Furthermore, conventional ultrasound imaging devices require only well-trained professionals to properly operate, thereby increasing the overall cost of ultrasound imaging and further limiting patient access to ultrasound imaging. Summary of the Invention

[0008] In various embodiments, an ultrasound imaging system is disclosed herein that includes a probe having an ultrasound transducer and a computing subsystem having a processor, non-volatile memory, and an image quality detector. The image quality detector may be executed by the processor. Using the ultrasound transducer, which may be a transducer array, the probe transmits and receives ultrasound signals, thereby acquiring ultrasound images arranged in a sequence of image frames. Each image frame includes an ultrasound image based on the ultrasound signals.

[0009] The ultrasound imaging system is configured to perform an auto-capture function that automatically identifies and saves clips of ultrasound images in a sequence of image frames that characterize clinically desirable views of an organ or other body feature imaged by the ultrasound imaging system. In this disclosure, references to an "organ" apply to the organ itself and also to other anatomical structures or body features of the patient. An image quality detector (e.g., an image evaluation process performed by a neural network in the computing subsystem) assigns a quality score, e.g., from 1 to 5, to each image frame in the sequence of image frames based on an evaluation of the quality of the ultrasound image in each image frame.

[0010] During ultrasound imaging, the ultrasound imaging system temporarily stores a sequence of image frames in an image buffer of a predetermined size (e.g., a cine memory for high-speed video recording). Depending on the size of the image buffer and the frame rate at which the ultrasound system acquires ultrasound images, the image buffer will hold the sequence of image frames acquired over a period of time. Quality scores assigned to ultrasound images in the sequence of image frames are stored in the quality buffer in a sequence of quality scores each corresponding to an image frame in the sequence of image frames.

[0011] An image quality detector in the computing subsystem determines a quality score for the ultrasound image in each image frame and stores the quality scores in a quality buffer. When the set of consecutive image frames reaches at least a first predetermined size and the quality score for each image frame in the set of consecutive image frames is equal to or greater than a first quality threshold, the computing subsystem automatically records an ultrasound clip of the image frames from the image buffer to nonvolatile memory. In some examples, the computing subsystem may automatically record an ultrasound clip including only the set of consecutive image frames from the image buffer to nonvolatile memory if the quality score for each image frame in the set of consecutive image frames is equal to or greater than the first quality threshold. In some cases, for example, the set of consecutive image frames may include 70% or more of the image frames in the image buffer, each image frame having a quality score of at least 4 on a scale of 1 to 5. In this manner, the auto-capture function automatically and retroactively saves a clip of ultrasound image frames from the image buffer to nonvolatile memory.

[0012] In some embodiments, the computing subsystem is configured to provide a smart capture function when a predetermined period of time elapses and the image buffer does not contain a set of consecutive image frames of at least a first predetermined size having a corresponding quality score equal to or greater than a first quality threshold. Activation of the smart capture function causes the computing subsystem to automatically record an alternate set of consecutive image frames from the image buffer when the quality score for each image frame in the alternate set of consecutive image frames is equal to or greater than a second quality threshold, which may be lower than the first quality threshold. In various embodiments, the alternate set of consecutive image frames may have a predetermined size that is different from the first predetermined size of the set of consecutive image frames that would otherwise be recorded by the automatic capture function if the image frame quality scores met the first quality threshold. Through these features, the systems and methods described herein provide for automatically recording clips of ultrasound images that meet a desired image quality. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram illustrating an ultrasound imaging system for automatic ultrasound image registration based on a quality score, in accordance with one or more embodiments of the present disclosure. [Figure 2] 1 is a schematic diagram of storing a sequence of image frames in an image buffer of a predetermined size and identifying a set of consecutive image frames in the image buffer that have quality scores greater than or equal to a first quality threshold, in accordance with one or more embodiments of the present disclosure. [Figure 3] FIG. 10 is a schematic diagram illustrating identifying an alternative set of consecutive image frames in an image buffer having a quality score equal to or greater than a second quality threshold and recording the alternative set of consecutive image frames in non-volatile memory, in accordance with one or more embodiments of the present disclosure. [Figure 4] 1 is a flowchart illustrating an automated process for recording ultrasound image clips based on quality scores of image frames in a set of consecutive image frames, in accordance with one or more embodiments of the present disclosure. [Figure 5] 10 is a flowchart illustrating a process for activating a smart capture function that can automatically record an alternative set of consecutive image frames based on a quality score that is equal to or greater than a second quality threshold, in accordance with one or more embodiments of the present disclosure. [Figure 6] FIG. 10 is a schematic diagram of a process for identifying a set of non-contiguous image frames in an image buffer that have a quality score equal to or greater than a quality threshold, and cumulatively splicing the set of non-contiguous image frames that are recorded in non-volatile memory, in accordance with one or more embodiments of the present disclosure. [Figure 7] 10 is a flowchart illustrating a process for activating an auto capture or smart capture function that can automatically splice together a set of non-contiguous image frames based on a quality score that meets or exceeds a quality threshold, in accordance with one or more embodiments of the present disclosure. [Figure 8] 4 is a multi-view auto-capture configuration performed by the auto-capture and smart capture functions of FIGS. 2 and 3 in accordance with one or more embodiments of the present disclosure. [Figure 9]10 is a flowchart illustrating a process for automatically registering a multi-view set of image frames based on quality scores of multiple views that are greater than or equal to a quality threshold, in accordance with one or more embodiments of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0014] Embodiments of the systems and methods described herein provide for the automated recording of clips of ultrasound images based on quality scores assigned to the ultrasound images. The systems and methods described herein may be particularly useful for ultrasound imaging performed by novice ultrasound technicians and / or utilizing handheld or mobile ultrasound imaging devices that may be deployed in non-traditional environments. Using an artificial intelligence approach, embodiments of the systems and methods described herein can automatically assign a quality score, e.g., from 1 to 5, to each acquired ultrasound image in a sequence of image frames based on an assessment of the quality of the ultrasound image in each image frame.

[0015] The systems and methods described herein may include an image buffer, e.g., a cine memory of a predetermined size, for storing a sequence of image frames. The quality scores assigned to each image frame of the sequence of image frames may be stored in a separate quality buffer. In various embodiments, the computing subsystem may compare the quality scores in the quality buffer to a quality threshold. The computing subsystem may initiate an auto-capture function to retroactively store a clip of ultrasound image frames from the image buffer to non-volatile memory when the quality score for a set of consecutive image frames in the image buffer is equal to or greater than the quality threshold and the set of consecutive image frames is at least a predetermined size. Alternatively, the computing subsystem may initiate a smart capture function when, after a period of time, the image buffer does not contain a set of consecutive image frames whose ultrasound images have a quality score equal to or greater than the quality threshold.

[0016] 1 is a block diagram of an example ultrasound imaging system 100 configured to automatically record ultrasound images as clips based on their quality scores. The ultrasound imaging system 100 comprises an ultrasound imaging device including a computing subsystem 104, an image buffer 110, a quality buffer 120, a non-volatile memory 130, an optional smart capture button 140, and a display 150. Each of these may be incorporated into a single ultrasound imaging device, such as a handheld or portable device, or may comprise multiple devices operably linked or linkable to each other by communication or electrical connections. As described in further detail herein, the computing subsystem 104 may include a processor 106 and at least one image quality detector 108, each of which may include programmed and / or hardwired circuitry configured to perform the functions or actions described herein.

[0017] The ultrasound imaging system 100 is operable to acquire ultrasound images of a patient and, in at least some embodiments, may be, for example, a handheld ultrasound imaging device. The ultrasound imaging system 100 may be operably coupled to or incorporated into an ultrasound probe 102.

[0018] The non-volatile memory 130 may be, for example, a non-volatile storage medium such as a flash memory, a hard disk drive, an optical storage device, a magnetic storage device, an organic storage medium, etc., and may be configured for high-speed storage of image frames.

[0019] The processor 106 may be any computer processor operable to execute instructions (possibly stored in memory 130) that cause the processor 106 to perform the functions of the ultrasound imaging device 100 as described herein.

[0020] The ultrasound probe 102 includes one or more ultrasound transducers that are driven by the ultrasound imaging system 100 to transmit ultrasound signals toward a target area on a patient and receive echo signals returning from the target area in response to the transmitted ultrasound signals. In operation, a user of the ultrasound imaging system 100 can hold the probe 102 relative to the patient's body at a position and angle to acquire a desired ultrasound image. Signals received by the probe (i.e., echo signals) are communicated to the ultrasound imaging device 100 and converted into electronic signals that can form, or be processed to form, an ultrasound image of the target area on the patient. The ultrasound image is stored in an image buffer 110. In some examples, the ultrasound imaging system 100 can include a display 150 that can display ultrasound images and / or other related information to a user.

[0021] In some embodiments, the image quality detector 108 receives ultrasound images acquired by the ultrasound probe 102, automatically determines an image quality for each of the received ultrasound images, and automatically assigns a quality score to each of the ultrasound images based on the determined image quality. For example, in some embodiments, the quality score may be a value between 1 and 5. In this manner, a quality score of 1 may represent the lowest image quality, while a quality score of 5 may represent the highest image quality.

[0022] In some examples, image quality detector 108 includes machine learning circuitry (implemented by software, hardware, or a combination thereof) that receives ultrasound images acquired by ultrasound imaging system 100 and automatically determines an image quality score for each received ultrasound image. In some embodiments, an ultrasound image grading module (which may be included as part of the machine learning circuitry) automatically determines the image quality score and stores the quality score in quality buffer 120.

[0023] In some embodiments, an ultrasound image recognition module (which may be included as part of the machine learning circuitry) may be used to automatically determine whether one or more of the acquired ultrasound images represent a clinically desirable view of an organ or other aspect, region, or feature of the patient. In some cases, the machine learning circuitry may also be configured to automatically label one or more anatomical structures within the ultrasound image. For example, the anatomical structure recognition and labeling module (which may be included as part of the machine learning circuitry) automatically recognizes anatomical structures within the ultrasound image and automatically associates a label with the recognized anatomical structure. In some embodiments, the label associated with the recognized anatomical structure is displayed (e.g., on the display 150) superimposed on or embedded in the ultrasound image in the area where the corresponding anatomical structure is displayed.

[0024] Each of the ultrasound image recognition module, the anatomical structure recognition and labeling module, and the ultrasound image grading module may be executed by a computational intelligence system employing artificial intelligence that draws from an image knowledge database to perform the functions of these modules as described herein (e.g., determining whether a received ultrasound image represents a clinically desirable view, recognizing and labeling anatomical structures within the ultrasound image, and determining an image quality score for each ultrasound image in a sequence of image frames). Suitable processes for training machine learning circuitry and generating the image knowledge database used by the ultrasound image recognition module, the anatomical structure recognition and labeling module, and the ultrasound image grading module are described, for example, in U.S. Pre-Grant Publication No. 2021 / 0077068(A1), which is assigned to the assignee of the present disclosure and expressly incorporated herein by reference.

[0025] As mentioned above, in at least some embodiments, the image quality detector 108 uses a computational intelligence system that employs artificial intelligence to determine a quality score for each acquired ultrasound image. "Artificial intelligence" is used broadly herein to describe computational intelligence systems and methods that can learn knowledge (e.g., based on training data) and use such learned knowledge to adapt an approach to solving one or more problems. Artificial intelligence machines can employ, for example, neural networks, deep learning, convolutional neural networks, and Bayesian program learning techniques to solve problems such as image recognition, anatomical structure recognition and labeling, and image quality grading. Furthermore, artificial intelligence may include any one or combination of computational techniques such as constraint programming, fuzzy logic, classification, traditional artificial intelligence, symbolic manipulation, fuzzy set theory, evolutionary computation, cybernetics, data mining, approximate reasoning, derivative-free optimization, decision trees, and / or soft computing.

[0026] In various embodiments, the machine learning circuitry may be trained based on training images. The training images may include any ultrasound image information. For example, the training images may include image information used to train an ultrasound image grading module, such as that used by image quality detector 108. The training images may include a variety of ultrasound images of different image quality (e.g., higher quality images, lower quality images, blurry images, etc.). The quality of the training images used to train the ultrasound image grading module may first be graded by an expert, such as a physician or other clinician. The quality of the training images may be graded based on any grading system. In some embodiments, the quality of the training images may be graded based on a standard grading system, such as the American College of Emergency Physicians (ACEP) grading rubric provided in Table 1 below.

[0027] [Table 1]

[0028] Each of the training images may be assigned a particular quality score or grade (eg, 1-5) by a physician or other clinician, with the assigned quality score or grade representing the quality of the training image.

[0029] In embodiments with an ultrasound image recognition module, the training images may include image information used to train the ultrasound image recognition module, such as various ultrasound image information associated with known views of an organ such as the heart. As a further example, the training images may be clinically desirable images, such as, for example, a suprasternal view of the heart. In such cases, the training images may be ultrasound images that have been pre-determined (e.g., by a physician or other trained professional) as adequately illustrating a clinically desirable suprasternal view of the heart. Each such training image may have slightly different characteristics (e.g., higher quality images, lower quality images, blurry images, images taken at slightly different angles, etc.), yet each such training image may be pre-determined as adequately representing a clinically desirable view of the heart or other anatomical structure or body feature.

[0030] Furthermore, the training images may include not only image information associated with clinically standard or desirable views, but also image information associated with clinically non-standard or clinically undesirable views. Thus, for example, the ultrasound image recognition module may receive a view of the heart that does not represent a particular clinically desirable view (e.g., suprasternal, subcostal, short-axis and long-axis parasternal, two-chamber apex, three-chamber apex, four-chamber apex, and five-chamber apex views). In such a case, the ultrasound recognition module may nevertheless be trained to recognize the image as a view of the heart and may further recognize the image as, for example, somewhere between the two-chamber apex view and the three-chamber apex view. The clinically standard three-chamber apex view can generally be obtained, for example, by rotating the ultrasound imaging probe counterclockwise by approximately 60° relative to the two-chamber apex view. Ultrasound images obtained using a probe rotated, for example, at a counterclockwise angle of 5° to 55° relative to the two-chamber apex view may be determined not to represent a clinically desirable view of the heart.

[0031] Additionally, the ultrasound image recognition module may be trained with training images showing various known, but clinically undesirable, views of the heart (such as views somewhere between the two-chamber apical view and the three-chamber apical view) and therefore may recognize such views (e.g., the ultrasound image recognition module may recognize a view as representing a 35° counterclockwise rotation of the probe 102 relative to the two-chamber apical view). In some embodiments, upon recognizing that an ultrasound image contains a known, clinically undesirable view, guidance may be provided to the user to move the ultrasound probe in a specific manner that ultimately achieves the acquisition of a clinically desirable view.

[0032] In some embodiments, the training images may include image information used to train the anatomical structure recognition and labeling module. For example, the training images may include various ultrasound image information associated with known anatomical structures, such as a particular organ (e.g., the heart) or a particular function of an organ (e.g., the left ventricle, the right ventricle, the left atrium, the right atrium, the mitral valve, the tricuspid valve, the aortic valve, etc.). Furthermore, the training images may include image information associated with such known anatomical structures from various different views. Anatomical structures may appear very different across different views; for example, the left ventricle may appear different in ultrasound images acquired in various different views (e.g., apical-LV, parasternal-LV, parasternal-LV). Therefore, ultrasound images representing a known anatomical structure (e.g., the left ventricle) in various different views may be provided as training images that may be utilized to train the anatomical structure recognition and labeling module to recognize not only the anatomical structure but also the particular view provided by the ultrasound image.

[0033] Other training inputs may also be provided to the ultrasound image recognition module for training, including, for example, manually entered inputs to adjust or otherwise manage the image recognition model developed within the image recognition module through the training process.

[0034] Using the training images, the machine learning circuitry (including the ultrasound image grading module, the ultrasound image recognition module, and / or the anatomical structure recognition and labeling module) may perform an iterative training process. The training may be based on a wide variety of learning rules or training algorithms. For example, the learning rules may include one or more of backpropagation, real-time recurrence learning, pattern-by-pattern learning, supervised learning, interpolation, weighted sum, reinforcement learning, time difference learning, unsupervised learning, and / or record learning.

[0035] Returning now to FIG. 1 , in various embodiments, the image buffer 110 may be a temporary storage device (e.g., volatile memory) having a predetermined size. In some examples, the image buffer 110 may store a sequence of image frames acquired by the probe 102. In some embodiments, the size of the image buffer 110 may be configured to be the same as or greater than the length of the ultrasound clip to be recorded. For example, the length of the ultrasound clip may be configured as 3 seconds, 5 seconds, 10 seconds, or 20 seconds. In such cases, the image buffer 110 is sized to be able to store a sequence of image frames that provide an ultrasound clip that is 3 seconds, 5 seconds, 10 seconds, or 20 seconds in length. In some examples, the image buffer 110 may be a circular or ring buffer, and storing image frames in the image buffer 110 may include deleting the oldest image frame from the image buffer 110 and adding new image frames to the image buffer 110 on a rolling basis. In such an example, the quality buffer 120 may be a circular or ring buffer, and the quality scores for the oldest image frames may be deleted from the quality buffer 120, while the quality scores for new image frames being added would be added to the quality buffer 120 on a rolling basis.

[0036] In various embodiments, the quality score for each image frame stored in image buffer 110 is stored separately in quality buffer 120. In some examples, quality buffer 120 may be a temporary storage device (e.g., volatile memory) having a number of logical storage locations equal to the number of logical storage locations in image buffer 110 for storing ultrasound image frames. In this manner, image quality detector 108 may be configured to assign a quality score to the ultrasound image in each acquired image frame and store the assigned quality scores in quality buffer 120, such that a quality score in quality buffer 120 corresponds to a respective image frame in image buffer 110. In various embodiments, quality buffer 120 may be implemented in a different memory device than image buffer 110 or in the same memory device as image buffer 110.

[0037] According to the auto-capture functionality described herein, processor 106 may evaluate the quality score assigned to each acquired image frame by comparing the quality score to a first quality threshold. In some examples where the quality scores are 1 to 5, the first quality threshold used by the auto-capture functionality may be a score of 4 on a scale of 1 to 5. In some examples, processor 106 may begin comparing the quality scores of image frames stored in image buffer 110 to the first quality threshold when (in response to, or thereafter) a predetermined amount of quality scores have been stored in quality buffer 120 or when a predetermined period of time has passed during which the quality scores have been stored in quality buffer 120.

[0038] If processor 106 determines that each of the image frames in the set of consecutive image frames in image buffer 110 has a corresponding quality score that is greater than or equal to a first quality threshold and that the set of consecutive image frames has reached at least a first predetermined size, processor 106 activates an auto-capture function described herein in which an ultrasound clip consisting of the set of consecutive image frames in image buffer 110 is automatically recorded, for example, in non-volatile memory 130. In some examples, processor 106 may continue acquiring images of a current region of interest, direct image acquisition to a new region of interest, or stop image acquisition. The auto-capture function may cause processor 106 to retroactively record in non-volatile memory 130 an ultrasound clip that includes all of the image frames stored in image buffer 110, or in some cases, retroactively record in non-volatile memory 130 an ultrasound clip that includes some but not all of the image frames in image buffer 110, the clip covering a period of time that is shorter than the period covered by all of the image frames stored in image buffer 110.

[0039] In some embodiments, the processor 106 may continue to automatically record additional image frames from the image buffer 110 that are contiguous with a set of consecutive image frames already recorded or identified to be recorded, for example, in the non-volatile memory 130, where the additional image frames have a quality score that is greater than or equal to the first quality threshold. In yet other embodiments, the processor 106 may be configured to automatically record additional image frames from the image buffer 110 that are contiguous with an already recorded set of consecutive image frames but have one or more quality scores that are not greater than or equal to the first quality threshold. Thus, the processor 106 may stop recording image frames from the image buffer 110 to the non-volatile memory 130 when an ultrasound clip consisting of a set of consecutive image frames is recorded, and possibly when a later newly acquired image frame has a quality score that is less than the first quality threshold or when a maximum number of consecutive image frames has been recorded.

[0040] In some embodiments, the processor 106 may be configured to provide a smart capture function when, after a predetermined period of time during which ultrasound images are acquired, the processor 106 determines that the image buffer 110 does not contain a set of consecutive image frames of at least a first predetermined size having corresponding quality scores equal to or greater than a first quality threshold. In such embodiments, the smart capture function may be provided and activated. Activation of the smart capture function causes the processor 106 to automatically record ultrasound clips including alternate sets of consecutive image frames from the image buffer 110 when the quality score for each image frame in the alternate sets of consecutive image frames is equal to or greater than a second quality threshold. Typically, the second quality threshold is less than the first quality threshold. Thus, according to the auto capture feature, ultrasound imaging system 100 first attempts to automatically record ultrasound clips of higher quality images, but if the quality scores of the acquired ultrasound images do not meet a higher first quality threshold for a set of consecutive image frames of at least a first predetermined size, ultrasound imaging system 100 provides a smart capture option to record an ultrasound clip consisting of an alternative set of consecutive image frames of at least a second predetermined size (e.g., as a "best available quality" clip), where the ultrasound images in the alternative set meet a lower second quality threshold. The second predetermined size of the alternative set of consecutive image frames may be smaller than, or in some cases may be equal to, or greater than, the first predetermined size of the (initial) set of consecutive image frames.

[0041] In some examples, the smart capture function provides a user-selectable smart capture button 140 that, when selected by a user of the ultrasound imaging system 100, activates the smart capture function, which enables the processor 106 to automatically record alternative sets of consecutive image frames from the image buffer 110. In this manner, the user can select the smart capture button 140 and choose whether to automatically record ultrasound clips consisting of alternative successive portions of the sequence of image frames in the image buffer 110, e.g., as "best available" image quality clips. In some embodiments, the smart capture function may be automatically activated and automatically record ultrasound clips consisting of alternative sets of consecutive image frames into the non-volatile memory 130 without waiting for user input.

[0042] 2 is a schematic diagram of an example of an image buffer 110 ready to be processed by the auto-capture function described above. In this embodiment, the image buffer 110 has a predetermined size 116. A sequence of image frames begins at time "tT" 112 and is stored in the image buffer 110 until time "t" 114. The temporal length of the stored sequence of image frames is equal to time T, which is the difference between time 112 and time 114.

[0043] 1, the processor 106 may be configured to automatically record an ultrasound clip including a set of consecutive image frames from the image buffer 110 when the quality score for each image frame in the set of consecutive image frames is greater than or equal to a first quality threshold and the set of consecutive image frames has at least a first predetermined size. The first predetermined size of the set of consecutive image frames may be a first predetermined portion of the image buffer 110. By way of example, the set of consecutive image frames may have a first predetermined size that is a defined percentage of the size 116 of the image buffer 110.

[0044] 2, the consecutive set of image frames 118 comprises 70% of the size 116 of the image buffer 110. Thus, the set of consecutive image frames 118 represents a clip of ultrasound images that encompasses at least 70% of the period covered by the image frames stored in the image buffer 110. In some other examples, the set of consecutive image frames 118 may have a first predetermined size that is greater than or less than 70% of the size 116 of the image buffer 110.

[0045] By way of example, in an embodiment of the present disclosure in which ultrasound image quality is measured on a scale of 1 to 5, the first quality threshold may be set to a quality score of 4. However, the processor 106 may operate using different scales and different quality thresholds depending on the quality criteria implemented by the ultrasound imaging system 100.

[0046] When evaluating the quality scores in the quality buffer 120, the processor 106 can identify when the quality score of a consecutive portion of the sequence of image frames in the image buffer 110 is greater than or equal to a first quality threshold (e.g., 4) and the size of the consecutive portion of the sequence of images is at least a first predetermined size. When such conditions are met, an auto-capture function causes the processor 106 to automatically retroactively record an ultrasound clip, for example, from the image buffer 110 to the non-volatile memory 130. The length of time of the ultrasound clip can be specified by the ultrasound imaging system 100 or can be set by a user of the ultrasound imaging system 100. Depending on the frame rate of the ultrasound imaging device that acquired the ultrasound image frames, the number of image frames in an ultrasound clip can vary. Therefore, as described herein, to automatically record an ultrasound clip of a desired length of time, the ultrasound imaging system 100 can be configured to record an ultrasound clip consisting of a set of consecutive image frames including at least a predetermined number of image frames.

[0047] Figure 3 is a schematic diagram of an example in which the quality of the image frames in the image buffer 110 is not in a condition to automatically record an ultrasound clip in accordance with the automatic capture function described above with respect to Figure 2. However, in the example of Figure 3, a smart capture function may be activated that causes the ultrasound imaging system 100 to record an ultrasound clip consisting of an alternate set of consecutive image frames, for example, as previously described herein.

[0048] FIG. 3 shows a time period “tT CINE 3 illustrates a high-speed buffer (e.g., cine) memory 330 capable of storing one or more sequences of image frames ranging from time "t" 332 to time "t" 334. In FIG. 3, the image buffer 110 is illustrated as a portion of the buffer memory 330, and the image buffer 110 is smaller in size than the buffer memory 330.

[0049] In the example shown in Figure 3, the image buffer 110 does not have a set of consecutive image frames with corresponding quality scores that are greater than or equal to the first quality threshold, as described with respect to Figure 2. In this case, the processor 106 may be configured to provide a smart capture button 140 that enables recording of an alternative ultrasound clip (e.g., recording of an alternative set of consecutive image frames from the image buffer 110 that have the "best available" quality).

[0050] A user of the ultrasound imaging system 100 can select the smart capture button 140 to cause the system 100 to record alternative ultrasound clips of a predetermined size (e.g., duration T) when the alternative sets 218 of consecutive image frames have quality scores equal to or greater than a second quality threshold and the alternative sets 218 of consecutive image frames have at least a second predetermined size. The second quality threshold is generally lower than the first quality threshold. As a non-limiting example, the second predetermined size may be approximately 70% of the predetermined size 116 (or duration T) of the image buffer 110. In some examples, the predetermined size 116 (or duration T) may be approximately 3 seconds, and the alternative sets 218 of consecutive image frames may be at least 2 seconds in length and have quality scores that meet or exceed the second quality threshold to be automatically recorded in the non-volatile memory 130. In some examples, the second quality threshold may be a quality score of 3 on a scale of 1 to 5, or a quality score of 2 on a scale of 1 to 5. In various embodiments, the size and quality criteria (ie, the length and quality thresholds of the set of consecutive image frames described herein) are adjustable according to the needs or specifications of the ultrasound imaging system 100 .

[0051] In some embodiments, when a user selects the smart capture button 140, the processor 106 determines whether the image buffer 110 contains an alternative set 218 of consecutive image frames of at least a second predetermined size that meets the quality criteria specified for the smart capture function. If the size and quality criteria are met, the processor 106 records an ultrasound clip of image frames from the image buffer 110 in the non-volatile memory 130. However, if the processor 106 fails to identify an alternative set 218 of consecutive image frames that meets the size and quality criteria, the processor 106 may send an error message to the display 150 to inform the user that the clip was not recorded in the non-volatile memory 130, at which point the user may wish to try again to acquire a higher quality ultrasound image. In some examples, the processor 106 may automatically provide the smart capture function without user input or remove the smart capture function. The smart capture button 140 may be removed by the processor 106 when the quality scores of the image frames in the image buffer 110 meet a first quality threshold used by the (initial) automatic capture function.

[0052] 4 is a flowchart 400 illustrating a process for automatic ultrasound image clip recording based on quality scores of a sequence of ultrasound image frames, for example, as described in FIGS. 1 and 2. In step 410, ultrasound images are captured by an ultrasound probe, such as the probe 102 described in FIG. 1. In operation, a user of the ultrasound imaging system 100 may hold the probe 102 relative to a patient's body at a position and angle to acquire a desired ultrasound image. Signals received by the probe 102 (i.e., echo signals) are communicated to the ultrasound imaging system 100 and may form, or be processed to form, a sequence of ultrasound images of a target area of the patient. Additionally, the ultrasound images may be provided to a display 150, which may display the ultrasound images and / or any other relevant information to a user.

[0053] In step 420, a sequence of image frames comprising ultrasound images is stored in an image buffer, such as image buffer 110 described in Figures 1-3. In some examples, as described above, storing image frames in image buffer 110 may include deleting the oldest image frame from image buffer 110 and adding a new image frame to the sequence of image frames stored in image buffer 110.

[0054] In step 430, a quality score for each image frame in the sequence of image frames is determined by an image quality detector (e.g., image quality detector 108) and stored in a quality buffer (e.g., quality buffer 120) illustrated in FIG. 1 . In some cases, steps 420 and 430 may be performed simultaneously, i.e., quality scores are determined and stored simultaneously with capturing each ultrasound image. In step 440, a processor performing an auto-capture function may evaluate the quality scores in the quality buffer to determine whether the quality scores of a consecutive portion of image frames in the image buffer are greater than or equal to a first quality threshold and whether the consecutive portion of image frames is at least a first predetermined size. If so, the auto-capture function causes processor 106 to retroactively record a clip of ultrasound image frames from the image buffer to non-volatile memory. The recorded clip includes at least the consecutive portion of image frames in the image buffer that meet the quality and size criteria.

[0055] Figure 5 is a flowchart 500 illustrating a process that includes recording an ultrasound clip using an automatic capture feature, or alternatively, a smart capture feature, based on quality scores of image frames stored in an image buffer, for example, as described in Figures 1-3. At step 510, ultrasound images are captured by a probe, such as probe 102 described in Figure 1. At 520, the ultrasound images are stored in a sequence of image frames in an image buffer, such as image buffer 110 described in Figures 1-3.

[0056] At 530, a quality score for each image frame stored in the image buffer is determined by an image quality detector and stored in a quality buffer (e.g., quality buffer 120 described in FIG. 1 ). At 540, a processor of ultrasound imaging system 100 evaluates the quality scores in the quality buffer and determines whether a set of consecutive image frames in the image buffer has a quality score that is equal to or greater than a quality threshold, such as a first quality threshold described herein. If the quality scores in the quality buffer indicate that the image buffer contains a set of consecutive image frames of a first predetermined size having a quality score that is equal to or greater than the first quality threshold, the processor proceeds to step 550, where the processor automatically records a clip consisting of at least that set of consecutive ultrasound image frames from the image buffer to non-volatile memory. On the other hand, if at step 540, the processor determines that after a period of time has elapsed, the image buffer does not contain a set of consecutive image frames of at least the first predetermined size having a quality score that is equal to or greater than the first quality threshold, the processor proceeds to step 560 and provides a smart capture function, e.g., smart capture button 140 as described with respect to FIGS. 1 and 3 . A user of the ultrasound imaging system 100 can select a smart capture button 140 that enables the system 100 to automatically record an alternative set of consecutive image frames of a second predetermined size that have quality scores equal to or greater than a second quality threshold, which can be a lower quality threshold than the first quality threshold.

[0057] The alternative sets of consecutive image frames may constitute an ultrasound clip that includes consecutive image frames from the image buffer having the "best available" quality. As described in FIG. 3, the alternative sets of consecutive image frames may be stored in non-volatile memory 130. If ultrasound imaging system 100 is unable to identify an alternative set of consecutive image frames that meets the second predetermined size and quality criteria, the processor of ultrasound imaging system 100 may send an error message to the display to indicate that the clip was not recorded in non-volatile memory, at which point ultrasound imaging system 100 may return to step 510 to allow the user to capture a higher quality image.

[0058] FIG. 6 shows a block diagram of a time domain having a predetermined size 336 and a time “tT CINE 1 illustrates a buffer (e.g., cine) memory 330 that stores one or more sequences of image frames ranging from time "t" 332 to time "t" 334. An image buffer 110 is shown as forming a portion of the buffer memory 330, the image buffer 110 having a predetermined size 116 that is smaller than the predetermined size 336 of the buffer memory 330.

[0059] 6, none of the image frames stored in the image buffer 110 have corresponding quality scores greater than or equal to the first quality threshold described in FIG. 2. However, the corresponding quality scores of one or more sets of non-contiguous image frames may be equal to or greater than a quality threshold, such as the first quality threshold. In this case, the processor 106 may be configured to splice together the non-contiguous image frames having quality scores greater than or equal to the first quality threshold, thereby generating a set of cumulative image frames 618 that are automatically recorded in the non-volatile memory 130. As a non-limiting example, in some embodiments, the set of cumulative image frames 618 may be required to have a size (e.g., length of time or number of image frames) that is approximately 70% of the predetermined size 116 of the image buffer 110 before the set of cumulative image frames 618 is automatically recorded in the non-volatile memory 130. Thus, if processor 106 determines that at least 70% of the image frames in image buffer 110 have a quality score equal to or greater than the first quality threshold, even collectively for a set of non-contiguous image frames, processor 106 may automatically record an ultrasound clip (e.g., consisting of spliced image frames 618) to non-volatile memory 130. In some cases, each of the sets of non-contiguous image frames that are spliced together must meet a predetermined size criterion before the set of non-contiguous image frames can be included in the set of accumulated image frames 618.

[0060] Alternatively or additionally, the processor 106 may provide a smart capture button 140 to enable alternate ultrasound clip recording when non-consecutive image frames have a quality score equal to or greater than the second quality threshold, but not the first quality threshold. In this manner, as illustrated in FIG. 3 , a user of the ultrasound imaging system 100 may select the smart capture button 140 and enable the system 100 to automatically record an alternate set of cumulative image frames 618.

[0061] In some cases, selecting the smart capture button 140 causes an ultrasound clip of a predetermined size 116 to be recorded in non-volatile memory 130 without splicing together non-contiguous image frames that are equal to or greater than a second quality threshold. In some examples, the second quality threshold may be a quality score of 3 on a scale of 1 to 5, or a quality score of 2 on a scale of 1 to 5. The size and quality criteria (i.e., the length and quality thresholds described herein) may be adjusted as needed or desired.

[0062] Figure 7 is a flowchart 700 illustrating a process that includes recording an ultrasound clip using an automatic capture feature based on quality scores of image frames stored in an image buffer, for example, as described in Figures 1-3. At step 710, ultrasound images are captured by a probe, such as ultrasound probe 102 described in Figure 1. At 720, the ultrasound images are stored in a sequence of image frames in an image buffer, such as image buffer 110 described above.

[0063] At 730, a quality score for each of the image frames stored in the image buffer is determined by an image quality detector and stored in a quality buffer (e.g., quality buffer 120 described in FIG. 1 ). At 740, a processor of ultrasound imaging system 100 evaluates the quality scores in the quality buffer and determines whether a set of consecutive image frames in the image buffer has a quality score that is greater than or equal to a quality threshold, such as a first quality threshold described herein, and whether the set of consecutive image frames is at least a first predetermined size. If the quality scores in the quality buffer indicate that the image buffer contains a set of consecutive image frames of the first predetermined size that have a quality score that is greater than or equal to the first quality threshold, the processor proceeds to step 750, where the processor automatically records an ultrasound clip consisting of at least the set of consecutive ultrasound image frames from the image buffer to non-volatile memory. On the other hand, if, at step 740, the processor determines that after a period of time has elapsed, the image buffer 110 does not contain a set of contiguous image frames of a first predetermined size having a quality score greater than or equal to the first quality threshold, the processor proceeds to step 760 to determine whether the image buffer contains a set of non-contiguous image frames having a quality score greater than or equal to the first quality threshold, which, when combined together, can create a cumulative set of non-contiguous image frames having the first predetermined size. If a positive determination is made, the system 100 can automatically splice (or combine) the sets of non-contiguous image frames that meet the first quality threshold together as a cumulative set of non-contiguous image frames, and at 770, the system 100 can automatically record a clip consisting of the cumulative set of non-contiguous image frames from the image buffer to non-volatile memory. While the foregoing example illustrates a process that includes determining whether an image frame has a quality score greater than or equal to the first quality threshold, in other examples, the process may include determining whether the image frame has a quality score greater than or equal to a second quality threshold, which may be lower than the first quality threshold.

[0064] FIG. 8 illustrates a multi-view automatic acquisition configuration 800 in which the ultrasound imaging system 100 performs the automatic acquisition and smart acquisition functions of FIGS. 2 and 3 for multi-view applications. During the course of evaluating a patient, a user of the ultrasound imaging system 100 may scan multiple views of the patient's organs. As previously mentioned, references to "organs" apply to the organs themselves and also to other anatomical structures or physical features of the patient. The ultrasound imaging system 100 may be configured to automatically evaluate ultrasound images captured by an ultrasound probe and detect which views of an organ, such as the heart, are captured within the ultrasound image. For example, the ultrasound imaging system 100 may implement an ultrasound image recognition module as described herein, which may be configured to detect known views of an organ (e.g., suprasternal view, subcostal view, short-axis and long-axis parasternal views, two-chamber apex view, three-chamber apex view, four-chamber apex view, and five-chamber apex view) and output an indication of the detected view. However, not all embodiments of the ultrasound imaging system 100 require the implementation of an ultrasound image recognition module.

[0065] As with the previous embodiment, ultrasound images captured by the ultrasound probe are stored in image frames in an image buffer, such as image buffer 110. An auto-capture function, such as that described in Figure 2, can cause a set of consecutive image frames to be recorded from the image buffer to a non-volatile memory, such as memory 130.

[0066] In some examples, as a user moves the probe to different positions relative to the patient to scan different views of an organ (e.g., the heart), image frames for each view are stored in an image buffer, while quality scores for the image frames are stored in one or more quality buffers. The multi-view auto-capture function can detect quality scores for multiple sets of consecutive image frames of a first predetermined size that are equal to or greater than a first quality threshold, each set including a particular view, and in response, can record multiple ultrasound clips based on the multiple views in non-volatile memory 130.

[0067] In some embodiments, the multi-view automatic capture configuration 800 includes multiple quality buffers, such as quality buffers 810, 820, and 830 as shown. The number of quality buffers may vary depending on the number of image quality detectors implemented by the system 100 to evaluate the quality of the ultrasound images being captured by the ultrasound probe. Each image quality detector may be configured to use a computational intelligence system employing artificial intelligence, as described previously herein, to determine a respective quality score for the ultrasound images captured by the ultrasound probe. Each of the image quality detectors may be trained using ultrasound images of known quality for each particular view of the organ. In this manner, each image quality detector is tuned to evaluate the image quality of the ultrasound images according to each known view of the organ, assign a quality score to each ultrasound image, and store the quality score in a respective quality buffer corresponding to the particular view the image quality detector is tuned to evaluate. In some examples, the number of quality buffers may be automatically selected or configured by the processor 106 based on the identification of an organ in the ultrasound images and the number of known views of that organ that have a corresponding image quality detector in the system 100. In FIG. 8, for simplicity, three quality buffers 810, 820, and 830 are illustrated.

[0068] The image quality detectors corresponding to the quality buffers 810, 820, and 830 may operate simultaneously or sequentially to evaluate ultrasound images captured by the ultrasound probe, with each image quality detector evaluating the ultrasound image according to a respective view of the organ for which the image quality detector is trained. Thus, while or after a sequence of ultrasound images is acquired by the ultrasound probe, the ultrasound images may be simultaneously or sequentially evaluated for image quality by the three image quality detectors corresponding to the quality buffers 810, 820, and 830, respectively. Depending on the particular view within the ultrasound image being captured, two of the three image quality detectors in this example may determine a relatively low quality score for the ultrasound image because the two image quality detectors are configured to evaluate the image quality of a view different from the particular view being captured. Meanwhile, the image quality detector configured to evaluate the image quality of the particular view being captured may determine a relatively high quality score for the ultrasound image and record the high quality score in the respective quality buffer corresponding to the image quality detector.

[0069] Embodiments of ultrasound imaging system 100 that do not specifically implement an ultrasound image recognition module for detecting the particular view being captured may use this discrepancy in quality scores between different image quality detectors to determine the view currently being captured by the ultrasound probe. When, for a particular sequence of ultrasound images, all but one of the quality buffers contain relatively low quality scores and one of the quality buffers contains a relatively high quality score, ultrasound imaging system 100 can determine which image quality detector corresponds to the one quality buffer with the high quality score and determine that the ultrasound probe is capturing the view that the determined image quality detector was trained to evaluate.

[0070] In some examples, the ultrasound image recognition module may detect a particular view being captured before the image quality detector determines a quality score for the captured ultrasound image. In such examples, the processor 106 of the ultrasound imaging system 100 may activate one particular image quality detector corresponding to the particular detected view being captured at that time. This feature may reduce the power and time consumption of the ultrasound imaging system 100 compared to a system 100 in which all of the image quality detectors are active. When the user moves the ultrasound probe to a different position relative to the patient and begins capturing ultrasound images of a different view, the ultrasound image recognition module may detect the change in view in the ultrasound image and cause the ultrasound imaging system 100 to deactivate the previously active image quality detector and activate a different image quality detector trained to evaluate the new view being captured by the ultrasound probe.

[0071] The automatic capture function described in FIG. 2 may be used to identify and record a set of consecutive image frames in an image buffer having a predetermined size and a quality score that is equal to or greater than a quality threshold. In some examples, the predetermined size of the set of consecutive image frames and the quality threshold to be met may be the same for each of multiple different views evaluated by different image quality detectors. Alternatively, for different views being captured, ultrasound imaging system 100 may require different predetermined sizes and / or different quality thresholds for the set of consecutive image frames to automatically record ultrasound clips based on different requirements for recording ultrasound clips of particular views of an organ.

[0072] 8, the auto-capture function may record an ultrasound clip consisting of a set of consecutive image frames from an image buffer corresponding to portion 812 of first quality buffer 810 when the quality score in portion 812 is greater than or equal to a first quality threshold and corresponds to a set of consecutive image frames of at least a predetermined size. The ultrasound clip is recorded from the image buffer to non-volatile memory, preferably in association with a particular view that the image quality detector associated with first quality buffer 810 is trained to evaluate.

[0073] A user of the ultrasound imaging system 100 may move the ultrasound probe 120 relative to the patient to acquire an ultrasound image of a second, different view, for example, corresponding to the view that the image quality detector associated with the second quality buffer 820 was trained to evaluate. While the ultrasound probe is imaging this second view, the quality score assigned and stored in the second quality buffer 820 may be relatively higher than the quality scores assigned and stored in the first and third quality buffers 810, 830 because the image quality detectors corresponding to the first and third quality buffers 810, 830 were trained to evaluate image quality of a view different from the particular view currently being captured. The auto-capture function may record an ultrasound clip consisting of a set of consecutive image frames from the image buffer corresponding to portion 822 of the second quality buffer 820 when the quality score in portion 822 is greater than or equal to a first quality threshold and corresponds to a set of consecutive image frames of at least a predetermined size. The ultrasound clip is recorded from the image buffer to non-volatile memory, preferably in association with the particular view that the image quality detector associated with the second quality buffer 820 was trained to evaluate.

[0074] In the same or similar manner, the user may again move the ultrasound probe relative to the patient to acquire an ultrasound image of yet another, different view, e.g., a third view corresponding to the view that the image quality detector associated with the third quality buffer 830 was trained to evaluate. While the ultrasound probe is capturing this third view, the image quality detectors corresponding to the first and second quality buffers 810, 820 are configured to evaluate image quality of a view different from the particular view currently being captured, so the quality score assigned and stored in the third quality buffer 830 may be relatively higher than the quality scores assigned and stored in the first and second quality buffers 810, 820. The auto-capture function may record an ultrasound clip consisting of a set of consecutive image frames from the image buffer corresponding to portion 832 of the third quality buffer 830 when the quality score in portion 832 is greater than or equal to the first quality threshold and corresponds to a set of consecutive image frames of at least a predetermined size. In various embodiments, the predetermined size of the set of consecutive image frames recorded from the image buffer may be the same as that described above with respect to the embodiment of FIG. 2, for example. However, in other embodiments, each of the predetermined sizes of the sets of consecutive image frames for different views may have different values based on the desired clip length for that particular view.

[0075] The smart capture function may be activated for the multiview auto-capture configuration 800 when the quality buffers 810, 820, and 830 do not have quality scores greater than or equal to a first quality threshold. If the quality buffers 810, 820, and 830 have quality scores greater than or equal to a second quality threshold and corresponding image frames in the image buffers form a set of image frames of at least a predetermined size, the set of image frames may be recorded in non-volatile memory as a “best available” clip. In some embodiments, the second quality threshold may be different for each of the quality buffers 810, 820, and 830, for example, based on the trained recognition system of each image quality detector for the particular view being evaluated by the image quality detector. In some examples, the smart capture function may cause the system 100 to automatically splice (or combine) non-contiguous image frames that meet the second quality threshold together as a set of alternative non-contiguous image frames and record a clip of cumulatively spliced image frames having quality scores greater than or equal to the quality threshold from the image buffers to non-volatile memory, for example, as illustrated in FIG. 6 .

[0076] FIG. 9 is a flowchart 900 illustrating a process including recording an ultrasound clip using an automatic capture function, or alternatively, a smart capture function, based on quality scores of image frames stored in an image buffer, for example, as described in FIG. 8. At 910, an ultrasound image is captured by an ultrasound probe, such as the ultrasound probe 102 described in FIG. 1. During operation, a user of the ultrasound imaging system 100 may hold the probe 102 relative to a patient's body at a position and angle to obtain a desired ultrasound image of a particular view of an organ, such as the heart. The user may then move the probe to different positions and angles to capture different views of the same or different organs. In some examples, an automated system such as those described herein may use a trained image recognition module to detect the particular view of the organ being imaged at a given time.

[0077] At 920, the ultrasound images for each of the views acquired by the ultrasound probe are stored in one or more sequences of image frames in an image buffer, as illustrated in FIG.

[0078] At 930, a quality score for each image frame stored in the image buffer is determined by a different image quality detector, each tuned to evaluate a particular view. The quality scores determined by the different image quality detectors are stored in different quality buffers (e.g., quality buffers 810, 820, 830 depicted in FIG. 8 ) corresponding to the respective image quality detector that generated the quality score.

[0079] At 940, the processor of the ultrasound imaging system 100 evaluates the quality scores in each of the quality buffers and determines whether a set of consecutive image frames in the image buffer has a quality score equal to or greater than a quality threshold, such as a first quality threshold described herein, and whether the set of consecutive image frames is at least a predetermined size. If the quality scores in each quality buffer indicate that the image buffer contains a set of consecutive image frames of a predetermined size having a quality score equal to or greater than the first quality threshold, the processor proceeds to step 950, where the processor automatically records an ultrasound clip from the image buffer to non-volatile memory, the ultrasound clip consisting of a set of consecutive ultrasound image frames corresponding to a particular view that meets the quality and size requirements. On the other hand, if at step 940, after a period of time has elapsed, the processor determines that the image buffer does not contain a set of consecutive image frames of at least a predetermined size having a quality score equal to or greater than the first quality threshold, for any of the views reflected in the quality buffer, the processor proceeds to step 960 and provides a smart capture function, such as the smart capture button 140 described with reference to FIGS. 1 and 3. A user of the ultrasound imaging system 100 may select a smart capture button, which enables the system 100 to automatically record alternative sets of consecutive image frames of a predetermined size having quality scores equal to or greater than a second quality threshold, which may be a lower quality threshold than the first quality threshold. The alternative sets of consecutive image frames may constitute an ultrasound clip that includes consecutive image frames from the image buffer having the "best available" quality for one or more particular views. As described in FIG. 3, the alternative sets of consecutive image frames may be recorded from the image buffer to non-volatile memory 130.

[0080] It should be understood that the various embodiments described above can be combined in different ways to provide further embodiments. These and other changes can be made to the embodiments in light of the above Detailed Description. In general, in the following claims, the terms used should not be construed to limit the claims to the specific embodiments disclosed in the specification and claims, but should be construed to include all possible embodiments along with the full range of equivalents to which such claims are entitled. Accordingly, the claims are not limited by the disclosure herein.

Claims

1. 1. A system for ultrasound imaging, comprising: an ultrasound imaging device configured to capture ultrasound images, the ultrasound imaging device comprising: an image buffer configured to store a sequence of image frames comprising acquired ultrasound images; a quality buffer configured to store quality scores in a sequence of quality scores each corresponding to an image frame in the sequence of image frames; a computing subsystem, and a computing subsystem configured to automatically record an ultrasound clip including a set of consecutive image frames from the image buffer when the quality score for each image frame in the set of consecutive image frames is greater than or equal to a first quality threshold and the set of consecutive image frames has at least a first predetermined size.

2. the computing subsystem: further configured to provide a smart capture function when, after a predetermined period of time, the image buffer does not have a set of consecutive image frames of the first predetermined size having corresponding quality scores greater than or equal to the first quality threshold; 2. The system of claim 1, wherein activation of the smart capture function causes the computing subsystem to automatically record an ultrasound clip including the alternative set of consecutive image frames from the image buffer when the quality score for each image frame in the alternative set of consecutive image frames is greater than or equal to a second quality threshold and the alternative set of consecutive image frames has at least a second predetermined size.

3. The system of claim 2 , wherein the first quality threshold is greater than the second quality threshold.

4. The system of claim 2 , wherein the first predetermined size of the set of consecutive image frames is different from the second predetermined size of the alternate sets of consecutive image frames.

5. 3. The system of claim 2, wherein the smart capture function includes a user-selectable smart capture button that, when selected by a user, activates the smart capture function and enables the computing subsystem to automatically record the ultrasound clip including an alternate set of the consecutive image frames from the image buffer.

6. 3. The system of claim 2, wherein activation of the smart capture function causes the computing subsystem to automatically record additional image frames from the image buffer that are contiguous with the alternative set of consecutive image frames but have one or more quality scores that are not greater than or equal to the second quality threshold.

7. 3. The system of claim 2, wherein the ultrasound clip including the set of consecutive image frames has a first predetermined length that is longer than a second predetermined length of an ultrasound clip including an alternate set of consecutive image frames.

8. 2. The system of claim 1, wherein the first predetermined size of the set of consecutive image frames corresponds to a first predetermined portion of the image buffer.

9. The system of claim 1 , wherein the image buffer is volatile memory and the ultrasound clips are automatically recorded in non-volatile memory.

10. 2. The system of claim 1, wherein the computing subsystem includes an image quality detector configured to determine the quality score for each image frame according to a particular view that the image quality detector is trained to evaluate.

11. 2. The system of claim 1, wherein the computing subsystem is configured to determine a quality score for each image frame of the sequence of image frames and store a predetermined amount of quality scores in the quality buffer before evaluating the quality scores in the quality buffer to identify a set of consecutive image frames in the image buffer having corresponding quality scores that are equal to or greater than the first quality threshold.

12. the computing subsystem: further configured to provide a cumulative recording function when, after a predetermined period of time, the image buffer does not have a set of consecutive image frames of the first predetermined size having corresponding quality scores equal to or greater than the first quality threshold; 2. The system of claim 1, wherein the cumulative recording function causes the computing subsystem to automatically splice together multiple sets of non-contiguous image frames from the image buffer when the quality score for each image frame in the multiple sets of non-contiguous image frames is equal to or greater than the first quality threshold, and to record an ultrasound clip including the spliced set of non-contiguous image frames from the image buffer.

13. the ultrasound imaging device further comprising a plurality of quality buffers, each quality buffer of the plurality of quality buffers configured to store a quality score for a sequence of image frames comprising captured ultrasound images of a respective view of an anatomical structure; 2. The system of claim 1, wherein the computing subsystem is further configured to automatically record an ultrasound clip including a set of consecutive image frames from the image buffer when the quality score for each image frame in the set is greater than or equal to a first quality threshold.

14. The system of claim 13 , wherein a different image quality detector is associated with each quality buffer corresponding to a particular view of the anatomy.

15. 1. A method for ultrasound imaging, comprising: capturing an ultrasound image with an ultrasound imaging device; storing a sequence of image frames comprising the ultrasound image in an image buffer; determining a quality score for each image frame of the sequence of image frames; storing the quality score for each image frame of the sequence of image frames in a quality buffer, the quality scores stored in the quality buffer corresponding respectively to the image frames in the sequence of image frames; and automatically recording an ultrasound clip including the set of consecutive image frames from the image buffer when the quality score for each image frame in the set of consecutive image frames is greater than or equal to a first quality threshold and the set of consecutive image frames has at least a first predetermined size.

16. The method of claim 15 , wherein determining the quality score for each image frame comprises using artificial intelligence to evaluate the ultrasound image in each image frame.

17. 16. The method of claim 15, further comprising: providing a smart capture function when, after a predetermined period of time, the image buffer does not have a set of consecutive image frames of the first predetermined size having a corresponding quality score greater than or equal to the first quality threshold; and automatically recording an ultrasound clip including the alternative set of consecutive image frames from the image buffer when, when the smart capture function is activated, the quality score for each image frame in the alternative set of consecutive image frames is greater than or equal to a second quality threshold and the alternative set of consecutive image frames has at least a second predetermined size.

18. 18. The method of claim 17, wherein providing the smart capture function includes providing a user-selectable smart capture button that, when selected by a user, enables the ultrasound clip including an alternate set of the consecutive image frames to be recorded from the image buffer.

19. 18. The method of claim 17, further comprising displaying an error message when the image buffer does not have a replacement set of consecutive image frames of the second predetermined size having corresponding quality scores greater than or equal to the second quality threshold.

20. 1. An ultrasound imaging system comprising: an ultrasound probe coupled to an ultrasound transducer configured to transmit and receive ultrasound signals, the ultrasound probe configured to capture a sequence of image frames, each image frame including an ultrasound image based on the ultrasound signals; an image quality detector configured to evaluate each image frame and assign a quality score to each image frame of the sequence of image frames based on a determined quality of the ultrasound image in each image frame; an image buffer configured to store the sequence of image frames; a quality buffer configured to store quality scores in a sequence of quality scores each corresponding to an image frame in the sequence of image frames; a computing subsystem, determining the quality score for each image frame of the sequence of image frames and storing the quality scores in the quality buffer; and a computing subsystem configured to automatically record an ultrasound clip including a set of consecutive image frames from the image buffer to non-volatile memory when the quality score for each image frame in the set of consecutive image frames is greater than or equal to a first quality threshold and the set of consecutive image frames has at least a first predetermined size.

21. 21. The ultrasound imaging system of claim 20, wherein the computing subsystem is further configured to provide a smart capture function when, after a predetermined period of time, the quality score for one or more image frames of the set of consecutive image frames of the first predetermined size is less than the first quality threshold, and when the quality score for each image frame in an alternate set of consecutive image frames is greater than or equal to a second quality threshold and the alternate set of consecutive image frames has at least a second predetermined size, activation of the smart capture function causes the computing subsystem to automatically record an ultrasound clip including the alternate set of consecutive image frames from the image buffer.

22. 22. The ultrasound imaging system of claim 21, wherein the smart capture feature includes a user-selectable smart capture button that allows the user to selectively activate the smart capture feature.

23. 22. The ultrasound imaging system of claim 21, wherein the computing subsystem is further configured to display an error message when the quality score of one or more image frames of the alternative set of consecutive image frames is less than the second quality threshold.

Images