Multi-faceted base heart wall motion detection in medical ultrasound

Abstract

To improve the data quality in detecting cardiac wall motion.SOLUTION: Regions of abnormal wall motion are detected from a view. The scan settings are then changed to focus scanning on each region, providing improved data such as data with more speckles being present. The scan settings may include changing an orientation of a scan plane, reducing out-of-plane motion, and / or increasing speckle content. The improved data is used to more accurately determine strain.SELECTED DRAWING: Figure 1

Description

Background Art

[0001] This embodiment relates to medical ultrasonic images. In particular, cardiac wall motion analysis is performed using ultrasound.

[0002] Two-dimensional (2D) B-mode cardiac echocardiography provides important information regarding cardiac surgery such as volume, diastolic function, right ventricular function, hemodynamics, and valvular regurgitation. Furthermore, the detection of heart failure is facilitated by the evaluation of global longitudinal strain and local strain in the cardiac wall. Myocardial mechanics, ischemic heart disease, cardiomyopathy, left ventricular (LV) diastolic dysfunction, and asymptomatic myocardial dysfunction in patients can be detected.

[0003] Current clinical tools require an ultrasound examiner to acquire multiple standard views of the cardiac chambers such as A2C, A3C, A4C, and longitudinal views, while optimizing image quality, maximizing frame rate, and minimizing shortening, all of which are important for reducing measurement variability. The global and regional strain of each view is first calculated by creating an initial contour at the myocardial border, which is then tracked across all frames using speckles. The 17-segment bulls-eye model (or 16 or 18 segments) is constructed from six regional strains of each view. Specifically, a quantitative wall motion score can be assigned to each segment. Typically, a 4-grade scoring is applied based on the segment model: (1) normal or hyperkinetic, (2) hypokinetic, (3) akinetic, and (4) dyskinetic.

[0004] Unfortunately, the semi-quantitative wall motion score for each segment is not always reliable. The accuracy of regional strain can be adversely affected by speckle tracking problems due to insufficient imaging at specific locations. It can be difficult to select a set of capture parameters that result in capturing wall motion in all areas of the standard views. It can be difficult to track speckles when significant motion is out-of-plane at a specific location.

Summary of the Invention

[0005] First, the preferred embodiments described below include a method, a system, and a non - transitory computer - readable medium for detecting cardiac wall motion. To improve data quality, regions of abnormal wall motion are detected from a view. The scan settings are then changed to focus the scan on each region, providing improved data such as data with more speckles. The scan settings may include changing the direction of the scan plane, reducing out - of - plane motion, and / or increasing the speckle content. The improved data is used to more accurately determine strain.

[0006] In a first aspect, a method for detecting cardiac wall motion by a medical ultrasound scanner is provided. The medical ultrasound scanner first performs a first scan of a first plane through a patient's heart using a first value of scan settings. The first plane is scanned at various times, providing first ultrasound data representing the heart at various times. Abnormal wall motion is detected from the first ultrasound data. The medical ultrasound scanner then performs a second scan of a region of abnormal wall motion within a second plane using a second value of scan settings. The second settings are different from the first value of the scan settings. The second plane is different from the first plane, and the second scan results in second ultrasound data. Strain is determined for the abnormal wall motion from the second ultrasound data. An image represents the strain.

[0007] In one embodiment, a standardized view of the heart is scanned first. The second scan has a smaller field of view that just covers the region of abnormal wall motion. The region is smaller than the field of view in the standardized view.

[0008] In another embodiment, a machine - learning model detects abnormal wall motion.

[0009] During scanning, various scanning settings such as gain, contrast, depth, and / or field of view may be changed. For example, the field of view for the first scan is for the entire heart region (a larger field of view), and the field of view for the second scan is a smaller field of view for the region around the abnormal motion. The scanning settings can include the orientation of the plane, such as changing the scanning settings to reorient from the first plane to the second plane. In one example, the scanning settings of the second scanner configure a medical ultrasonic scanner for generating speckles.

[0010] According to one embodiment, the second plane is selected based on the level of dispersion of the heart wall and / or the amount of relative motion. The abnormal region may be scanned along various planes, and one of the planes may be selected more optimally compared to the other planes.

[0011] As one embodiment, the strain is determined for the segments of the multi-segment model. A multi-segment model including strain is displayed. As another embodiment, the strain is determined in multiple dimensions with respect to the first plane and the second plane. A multi-dimensional strain map is displayed.

[0012] The second scan and the determination of strain may be repeated for various regions corresponding to various positions of abnormal wall motion. The first scan, the detection of abnormal wall motion, the second scan, and the determination of strain may be repeated for various views of the heart (e.g., various standardized views). Then, the strain from the various standardized views is displayed.

[0013] In a second aspect, an ultrasonic system for detecting heart wall motion is provided. The ultrasonic scanner detects the position of an abnormality in the heart wall motion of a patient, scans that position with various values of one or more settings, and is configured to determine the strain of the heart wall motion at that position from the ultrasonic data from the scan using at least one of the various values of the one or more settings. The display is configured to display an image of the strain.

[0014] In one embodiment, one or more settings are for scanning various planes such that the ultrasonic scanner is configured to scan various in-plane positions. In another embodiment, one or more settings are for scanning positions for the ultrasonic scanner to image at various levels of speckle.

[0015] In another embodiment, the ultrasonic scanner is configured to scan the patient's heart in a standardized view, detect a position from the standardized view, and scan the position in a field of view smaller than the standardized view.

[0016] In a third aspect, a non-transitory computer-readable recording medium stores data representing instructions executable by a processor programmed to detect heart wall motion. The recording medium includes instructions for scanning the patient's heart in a standardized view, detecting abnormal wall motion of the heart from the standardized view, acquiring B-mode data in a patch of the abnormal wall motion, determining strain from the B-mode data for the patch, a first segment of a strain map having the strain determined from the B-mode data for the patch, and a second segment of a map having another strain determined from the standardized view, and for displaying the strain map including the first and second segments. Distortion including instructions for displaying a strain map including the first and second segments.

[0017] According to one embodiment, instructions for acquiring B-mode data within a patch include acquiring B-mode data by a patch in a plane different from the standardized view. The instructions further include selecting different planes based on the amount of speckle or motion.

[0018] As another embodiment, instructions for acquiring B-mode data include acquiring B-mode data having a higher level of speckle in the patch than in the standardized view at the position of the abnormal wall motion.

[0019] In yet another embodiment, the instructions further include repeating scanning, detecting, acquiring, and determining for various standardized views. Instructions for displaying a distortion map include displaying a 16, 17, or 18 segment bull's-eye view.

[0020] The following claims define the invention, and nothing in this section should be construed as a limitation on those claims. Further aspects and advantages of the invention are described below in conjunction with the preferred embodiments.

Brief Description of the Drawings

[0021] The components and figures are not necessarily to scale, and emphasis has been placed on illustrating the principles of the invention. Further, with respect to the drawings, like reference numerals indicate corresponding parts throughout the various figures.

[0022]

Figure 1

[0023]

Figure 2

[0024]

Figure 3

[0025]

Figure 4

Modes for Carrying Out the Invention

[0026] Tracking acquisition assistance is provided for, e.g., the fusion of 2D B-mode echocardiography for global and / or local 2D cardiac mechanics. The user is assisted, for example, in obtaining the best possible or sufficient B-mode echocardiography data for quantifying global and / or local strain by sampling in multiple planes for any region of abnormal wall motion.

[0027] Automated solutions calculate global and local cardiac mechanics from 2B echo data. The strain analysis of these solutions strongly depends on image quality. One set of parameters or settings for the ultrasound scanner, such as gain settings, may not capture wall motion in all areas. For example, one or more gaps in the cardiac wall appear due to the lack of speckle, shadowing, and / or other artifacts.

[0028] To assist the user and / or automated solutions for better images in specific regions, patches are scanned with various values for the scanning settings. Whether due to actual abnormalities or poor scanning, regions with abnormal wall motion are rescanned with various settings to improve the data for improving the data by additional acquisition and / or for better settings for scanning their positions. In one approach, the various settings are for scanning the abnormal region along various scan planes. The field of view is shifted not only in the region (smaller) and focus (within the region), but also in the plane being scanned. This multi-faceted analysis of abnormalities in cardiac wall motion from 2D echo can provide data that can be used for more accurate tracking and corresponding strain analysis. Improved data, and even reliability measures, can be provided for tracking and image quality. The image quality is improved, especially in low-reliability regions, by changing the acquisition parameters, such as visually rotating the probe if necessary. Tracking may be re-executed only in low-reliability regions.

[0029] In one embodiment, an ultrasound examiner is assisted during acquisition of 2D B-mode cardiac echo examinations to acquire good quality data within a patch and fuse results for strain analysis. When the ultrasound examiner acquires a 2D B-mode cardiac echo examination image in a standard orientation view such as A2C, the system first detects myocardial boundaries with abnormal wall motion and adjusts input parameters and the orientation plane to acquire more data, for example, only for regions with abnormal wall motion (automatically or through user guidance). The system causes a certain amount of probe rotation, either through automation and / or guidance to the ultrasound examiner, and quantifies motion in multiple planes only in the vicinity of abnormal wall motion in real time. Tracking is performed, for example, only in these corrected regions. The resulting calculated strain is collected or fused with previously obtained results from other regions. A multi-plane system can provide more robust strain analysis and less inter- and intra-user variability. In particular, image quality is improved for images acquired by less experienced ultrasound examiners. A framework for 2D B-mode cardiac echo image acquisition based on semi-autonomous or fully autonomous robotic devices is established.

[0030] Figure 1 shows a method for detecting the motion of a heart wall using a medical ultrasound scanner. Strains and / or other metrics of abnormal and / or normal motion or operation of the heart are detected. Detection depends on acquiring additional data at various settings such as various fields of view, gains, and / or scanning planes for regions that appear to have abnormal heart wall motion. By collecting this additional data at various values to construct the scan, improved ultrasound data for strain or other analysis in those regions is provided.

[0031] This method is performed by the system of FIG. 4, an ultrasonic scanner, an image processor, or various systems or devices. In one embodiment, an ultrasonic imaging system or a medical ultrasonic scanner performs the operations of FIG. 1. In other embodiments, a processor performs operations 110, 130, 140, and 150. The ultrasonic scanner performs operations 100 and 120. A display can be used for operation 150.

[0032] Additional, various, or fewer operations may be performed. For example, operations for determining volumetric flow rate, ejection fraction, information regarding heart motion, or other flow analysis of the heart are provided. As another example, operations for using strain or other heart wall motion analysis for classifying the state of the heart, for segment scoring, and / or for diagnosis are provided. As another example, operation 150 is not provided.

[0033] The operations are performed in the order shown (top to bottom or numerically) or in various orders.

[0034] In operation 100, the ultrasonic scanner scans the patient's heart. The patient's heart is scanned in B-mode using a transducer scan through a rib window from outside the patient, a transesophageal transducer scan from within the patient's throat, or an intracardiac echocardiogram catheter scan from within the heart.

[0035] The scan is performed using values of scan settings. The values are for a beamformer, a transducer, an image processor, and / or post-processing. Values of the field of view (e.g., depth, width, and / or scan pattern), gain, contrast, frequency, and / or other setting or acquisition parameters for the scan of operation 100 are set by an ultrasonic examiner and / or are default values for a heart image.

[0036] The value generates a field of view that scans the entire heart region and / or the heart chambers of the heart. In one embodiment, the scan is for a standardized view of the heart, such as a longitudinal, A4C, A3C, or A2C view. The scan of operation 100 can be for other standardized views or non-standard views. Since the scan is 2D, it is a scan along a plane passing through the heart.

[0037] The scan of operation 100 is repeated to collect two or more images representing the patient's heart at various times. For example, a sequence of five or more images per cardiac cycle is acquired. The ECG can be used to specify the timing of the images relative to the cardiac cycle. Gates or triggers can be used to acquire images at desired times of the cardiac cycle, such as end-diastole (ED) and end-systole (ES).

[0038] 2D B-mode cardiac echo examination data (e.g., images) are acquired for a plane within the heart, such as in one of the standard views (e.g., A2C). FIG. 2 shows a B-mode image 200 of a plane passing through the heart.

[0039] In operation 110, the image processor detects abnormal wall motion from the ultrasonic data resulting from the scan of operation 100. Ultrasonic images from various times are used to detect abnormal wall motion. The images are used for ultrasonic data from a scan in a display format or a scan format before scan conversion. Data representing the patient that can be used or is used to generate an image from an ultrasonic scan is used as image data. The detection uses images from various times during the cardiac cycle, such as multiple images of a standardized view.

[0040] After obtaining an image of the best or acceptable quality for the 2D view, the image processor detects the endocardial and / or epicardial boundaries (e.g., myocardial boundaries). FIG. 2 shows an example where boundary 216 is detected in operation 210 and in ED image 212 and ES image 214. Any detection may use, for example, a random walker, thresholding, or pattern or model fitting. In one embodiment, a machine learning algorithm or machine learning model detects boundary 216. The detection is performed for each of the images at other times of the cardiac cycle. In another embodiment, boundary 216 is automatically detected on the ED frame or image. The detected boundary 216 is tracked within other frames, such as by using speckle or local feature tracking (e.g., optical flow, motion priors, or speckle tracking). As a result, boundary 216 is located in each of the images of the cardiac cycle.

[0041] In operation 220, any positions along boundary 216 with abnormal motion are detected (operation 110). For example, global and local strains are calculated from the tracked points along boundary 216. Examples of the tracked points in images 212, 214 of FIG. 2 are shown as white dots, and this strain indicates the extension or shortening of the boundary between points over a predetermined portion of the cardiac cycle, such as ED - ES. Strains greater than or less than a threshold may indicate abnormal motion. A machine learning model or the image processor can be used to detect abnormal motion. Segment boundaries with abnormal motion are identified. The segments or regions of abnormal motion may be larger or smaller than the standard segments used in a bull's-eye visualization, such as a standard 6 - segment division. One or more regions of abnormal wall motion may not be detected. For example, if any, a myocardial boundary region with abnormal motion is selected. The strain from the region of normal motion is calculated and may be saved for use in a strain map (e.g., bull's-eye model).

[0042] The abnormality can be due to actual poor heart wall motion or due to poor tracking. An example is shown in Figure 2. The abnormality detection of operation 220 detects the region 224 of abnormal wall motion. As shown in image 222 and highlighted in the zoomed-in portion 226 of image 222, there is no B-mode return in this region 224. The speckles are few. The tracking resulting from finding the boundary 216 can be inaccurate, resulting in inaccurate strain calculations and corresponding displays of abnormal wall motion.

[0043] To improve the strain calculation for region 224, in operation 230, additional target acquisition is provided. The target acquisition can obtain an image 232 with more B-mode returns such as speckles for region 224. The acquisition can be focused on region 224 and target that region. Using a smaller field of view and corresponding settings, better data for tracking and / or strain calculation within the region is obtained instead of for the entire view of the heart cavity.

[0044] In operation 120 of Figure 1, the ultrasound scanner scans region 224. The various regions 224 of detected abnormal wall motion may be scanned as separate patches. The scan results in ultrasound data or images representing region 224 over time in one or more cardiac cycles.

[0045] The ultrasonic scanner is configured by set values (e.g., acquisition parameters) for scanning the region 224 or patch. The value of operation 120 is different from the value used in operation 100. For example, a smaller field of view or patch is scanned. The depth, width, focus, and / or scan pattern are various, such as scanning only a sub-region of the region of the image 200. The field of view is set to just cover the region 224 of abnormal wall motion. This region 224 is smaller than the entire boundary 216 and the corresponding original image 200. The gain, contrast, or other settings may be the same or different. Multiple combinations of values may be tried to identify a combination that provides the desired or optimal ultrasonic data, such as increasing the speckle or landmark information used for tracking. The value of the patch or region 224 is set to generate speckles in the resulting image 232.

[0046] In one embodiment, the field of view is changed by changing the direction of the scanning plane with respect to the region 224 of the patient and / or the heart. The B-mode data is acquired using a patch or scanning plane that is a different plane from that used for the scan of operation 100. For example, the scanning plane is rotated so that the region 224 is at a non-zero angle with respect to the plane of the standardized view acquired in operation 100. An example is shown in FIG. 3. The region 224 is first scanned within the plane 300. The scanning plane is rotated to a plurality of planes 310 to scan the patch around the region 224. By scanning in multiple planes, more speckles may occur in the patch image 232 than in the region 224 of the original image 222 (see part 226). Alternatively, or additionally, more accurate motion of the heart wall can be detected because there is less out-of-plane motion at the boundary 216 within the multiple planes.

[0047] Any amount of rotation may be used. One or more rotational scan planes can be used, such as various rotations and sampling (e.g., scanning) on corresponding surfaces. The results are additional data or samples that can be used to improve accuracy and / or find an optimal or better scan plane for region 224. Alternatively, only one rotation and corresponding scan plane are used.

[0048] The transducer may be automatically rotatable. For example, a robot controlling the transducer rotates the transducer. As another example, electronic steering of a helical array or multi-dimensional (2D) array is used. In another example, mechanical rotation of an array in a transesophageal probe, etc., is used. Alternatively, an instruction to rotate is output to the sonographer, and the sonographer then manually rotates the array to scan along various surfaces and corresponding fields of view.

[0049] In one embodiment, the user is induced to obtain a higher quality image in the vicinity of such a detection region 224 by optimizing the machine or acquisition parameters and / or by rotating the imaging surface. The system recommends to the user to rotate the probe to a certain extent and quantify the motion only in this selected region 224. The probe or array is rotated by a specific angle suggested by the system, and the system analyzes the motion pattern and then estimates the next best possible position. The optimal orientation plane can be automatically determined by minimizing out-of-plane motion (e.g., finding the plane showing the maximum amount of motion) and / or by improving the quality of the speckle pattern, which results in improved tracking and motion field estimation. For example, when an abnormal motion region 224 is near the apex and the speckles move towards the apex, it is difficult to accurately track such speckles. A better orientation plane improves tracking at such a selected position or region 224, and since the image 232 includes only a small selected region 224, tracking can be performed in real time.

[0050] In another embodiment, the settings (e.g., the values of the acquisition parameters) of the ultrasonic scanner are adjusted in the selected area by the user or the image processor until a better, improved, more, or sufficient speckle pattern is visible. The speckle can be measured by a machine learning model and / or by a measure of variance. The resulting improved ultrasonic data can be used to verify that the abnormality is not due to low imaging quality in such areas. The strain measurement value of region 224 is updated by tracking only region 224 selected using the target acquisition image 232. The resulting strain is fused or collected with the previously determined strain for segments or regions without abnormal wall motion. The accuracy of the partial strain is improved due to the improved image quality. If for other reasons higher quality data cannot be obtained in such areas, having additional data as additional sampling can result in a more accurate strain determination.

[0051] In operation 130, the image processor selects the value of the setting that provides the most accurate tracking, strain, and / or analysis. If a plurality of various combinations of values are sampled, the one that provides sufficient or better ultrasonic data is selected. The level of speckle and / or out-of-plane motion is measured in one approach. In one example, the various scan planes in the various corresponding orientations with respect to region 224 can result in various levels of speckle and / or out-of-plane motion. By maximizing the speckle content and / or minimizing the out-of-plane motion (maximizing the in-plane motion), improved ultrasonic data for tracking and / or strain determination is provided. The best available or sufficient (above / below threshold) setting value is selected. For example, one in-plane field of view is selected on another plane based on the level of scatter and / or the amount of relative motion of the heart wall. The selection of the value is the selection of the ultrasonic data to be used for tracking or strain determination and vice versa.

[0052] Alternatively, no selection is made. Ultrasonic data from only one of the various value sets tried is used, or data from any sampling is used together (e.g., calculating the strain or wall position from each and averaging the results).

[0053] In operation 140, the image processor determines the strain of abnormal wall motion. The ultrasonic data from the target acquisition in operations 120, 230 is used to determine the strain. Patches of B-mode data of region 224 from various times in the cardiac cycle are used to track boundary 216, from which the strain is calculated. Since improved data is used, the resulting tracking and / or strain can be more accurate. A combination of data and / or set values from the selected surface is used for tracking and strain determination. Data from other combinations of set values may not be used or may additionally be used (e.g., average results).

[0054] The strain is determined for region 224. This strain can be interpolated, extrapolated, and / or used for segments of a multi-segment model. Region 224 may be at various positions and / or of various sizes relative to the segments of the multi-segment model, and thus the corresponding strain of the region is used to represent the segment or is modified to be used.

[0055] In another embodiment, sampling on one or more surfaces in addition to the original surface is used. Multidimensional strain is determined. For example, the strain as a vector in three dimensions is determined using data from various surfaces. The motion is tracked along multiple surfaces. As a result, three-dimensional strain is determined.

[0056] 1 illustrates the feedback from operation 140 to operation 120. If the image processor detects multiple regions 224 in a given view (e.g., image 200 with corresponding images 212 and 214) over different times, the target scanning of operation 120, the selection of operation 130, and the distortion determination of operation 140 are repeated for each of the regions 224. The scanning of the patch and the distortion determination for that patch are repeated to provide an improved distortion for each region 224. The repetition is performed until the distortion is determined for each of the regions of the anomaly. The same process is followed for all the anomaly segments.

[0057] 1 also illustrates feedback from operation 140 to operation 100. In the case of a multi-segment bull's-eye model, the different segments may be from different views. Operations 100, 110, 120, 130, and 140 are repeated for each of the views (e.g., A4C, A3C, and A2C). For each standard or non-standard view, the scanning of operation 100, the detection of regions 224 of abnormal wall motion in operation 110, the target scanning of a patch or patches around each region of operation 120, the selection of operation 130, and the determination of distortion 140 are repeated.

[0058] This tracking-based guidance continues until the full extent of the abnormal wall motion region is covered. The bull-eye visualization is updated from additional strain calculations. Alternatively, a 3D motion field is constructed by fusing tracked motion or strain from various directions. 3D surface tracking can be performed and 3D strain measurements can be calculated.

[0059] The image processor outputs the distortion in operation 150. The output may be an image output to a display, a computer network transmission, or a memory.

[0060] The image represents distortion. The distortion can be represented as text, graphics, emphasis, annotation, or other presentation. In one embodiment, the image is a multi-segment model that includes distortion. For example, a 16-, 17-, or 18-segment bull's-eye view or model is used. The distortion of each segment and the overall distortion are represented in color and / or alphanumerically. The distortion can be used for scoring so that the output score reflects the distortion. Other distortion maps may be used. The distortion in a given distortion map can include one or more distortions determined from a standard view and one or more distortions determined from re-scanning or target scanning of patches using various values for scanning settings. For example, a 17-segment model has a distortion map having a distortion determined from the B-mode data of a patch (e.g., from scan 120) for one segment and a second segment of the distortion map having another distortion determined from a normalized view (e.g., from scan 100). The distortions determined from various data can be fused together (e.g., presented together) to better quantify the extent of the abnormal wall motion region. Instead of values resulting from fusing only two adjacent views, high-density distortion values are now available within each segment of the map, so target scanning can improve the distortion map (e.g., of a 17-segment model). Further, a more reliable four-level scoring is constructed for rapid assessment of the medical condition and motion abnormalities.

[0061] In another embodiment, the image processor generates a multidimensional distortion map for display. Volume, surface, or other 3D rendering is used. The distortion may be from various views. Some of the distortion can be from scans along various surfaces. These distortions from various surfaces provide a 3D map or vector of the distortion. 3D surface tracking can be applied to any multi-plane acquisition to generate a 3D distortion map only in selected regions. Higher accuracy can be provided for these regions. Full 3D B-mode cardiac echocardiography has lower temporal and spatial resolution than 2D, requires a longer analysis time, and thus is used to provide 3D distortion or tracking for specific subsets or regions. Alternatively, a distortion vector or a graph of another image showing distortions of various dimensions is displayed.

[0062] FIG. 5 shows one embodiment of a system for detecting heart wall motion. By reacquiring ultrasonic data of a suspicious patch with various acquisition settings, the motion of the heart wall can be determined more accurately.

[0063] The system performs the operations of FIG. 1. For example, the ultrasonic scanner 400 uses the beamformer 408 and the transducer 410 to execute the scans of operations 100 and 120. The image processor 402 performs operations 110, 130, and 140 with the output of the distortion in operation 150 being at the display 420. Alternatively, the system performs various operations, additional operations, or fewer operations.

[0064] The system includes a medical diagnostic ultrasound scanner 400, a detachable transducer 410, a display 420, and an ECG device 430. Additional, various, or fewer components may be provided. For example, an external processor of the ultrasound scanner 400 for analyzing ultrasound data is provided. As another example, the ECG device 430 is integrated within the ultrasound scanner 400. In yet another example, since the ultrasound scanner 400 uses ultrasound data to determine cardiac cycle characteristics (e.g., intervals), the ECG device 430 is not provided.

[0065] The ECG device 430 is a processor, circuitry, and / or electrodes. Any currently known or later developed ECG device may be used. By placing the electrodes on the patient, a cardiac trace of the patient's cardiac cycle is generated. The ECG device 430 detects the R wave, other phases, intervals, and / or heart rate of the cardiac cycle. Alternatively, the ultrasound scanner 400 detects the phase of the cardiac cycle from the trace signal received from the ECG device 430.

[0066] The ultrasound scanner 400 is a medical diagnostic ultrasound imaging system. In other embodiments, the ultrasound scanner 400 is a treatment system that can also image. Any currently known or later developed system for cardiac echocardiography can be used. The ultrasound scanner 400 is configured by settings for performing a cardiac echocardiogram of the patient. During the imaging session, the transducer array 410 is positioned to scan the patient's heart, and images are generated over one or more cardiac cycles.

[0067] The ultrasound scanner 400 includes a transducer array 410, a beamformer 408, an image processor 402, a display 420, an input 406 , meIt includes a memory 404. Additional, various, or fewer components may be provided. For example, the transducer array 410 and / or the display 420 are separate from the scanner 400. As another example, the memory 404 is remote from or not part of the ultrasonic scanner 400. As yet another example, a scan converter, a time filter, a spatial filter, or another ultrasonic imaging component is provided.

[0068] The transducer array 410 is an array of transducer elements within a housing. The housing is adapted or shaped for handheld use outside the patient. Alternatively, the housing is shaped as a catheter, an intraoperative probe, an interventional probe, a transesophageal probe, or other currently known or later developed transducer probe. The array is a linear, multi-dimensional, circular, or other currently known or later developed array of piezoelectric or microelectromechanical elements.

[0069] The transducer array 410 generates acoustic energy in response to an electrical signal from the beamformer 408. For imaging, the acoustic echoes received by the transducer array 410 are converted into electrical signals, and the transducer array 410 provides the electrical signals to the beamformer 408.

[0070] The beamformer 408 is a transmit beamformer, a receive beamformer, or both a transmit beamformer and a receive beamformer. As a transmit beamformer, the beamformer 408 includes a waveform generator or pulse, a delay, a phase rotator, a timing generator, an amplifier, combinations thereof, or other currently known or later developed transmit beamformer components in a plurality of channels. For transmission, the beamformer 408 generates relatively delayed and apodized waveforms for each of the plurality of channels for the corresponding plurality of transducer elements. The transducer array 410 forms one or more acoustic beams in response to the waveforms.

[0071] As a receiving beamformer, beamformer 408 includes a channel with a delay, a phase rotator, an amplifier, or a combination thereof, and includes a plurality of adders or one adder for adding together the signals from each channel. For reception, beamformer 408 generates samples representing various spatial positions.

[0072] The received beamformed samples are provided to image processor 402 for generating an image. Image processor 402 is a detector, a filter, a scan converter, a three-dimensional processor, a combination thereof, or other currently known or later developed image generator. The detection is in B-mode (intensity). The samples are detected, scan-converted, and provided to display 420. Other image processors may be applied.

[0073] Ultrasonic scanner 400, including beamformer 408, transducer array 410, and / or image processor 402, operates according to set values. Various acquisition parameters are set to establish a field of view, contrast, gain, and / or other characteristics of the image. Ultrasonic scanner 400 is configured to scan or acquire ultrasonic data based on the set values.

[0074] Ultrasonic scanner 400 is configured to scan a patient's heart. For example, using manual guidance and / or automatic detection, the heart is scanned to identify a standardized view. A series of images over one or more cardiac cycles of the view are acquired by the scan. The set values define the scan and the image processor for acquiring the ultrasonic data.

[0075] Ultrasonic scanner 400 is configured to scan a target patch. One or more positions within the initial view are scanned in a more targeted manner. The set values are changed to rescan a local area, such as an area indicating abnormal heart movement. The field of view, gain, contrast, or other settings are changed to scan a patch around the area.

[0076] The input 406 is a user input device such as a keyboard, touch screen, trackball, mouse, button, rotatable knob, and / or slider. The input enables ultrasonic examiner input of set values for configuring the ultrasonic scanner 400. Other user input information may be input for detecting abnormal regions and / or indicating or selecting an improved image.

[0077] The image processor 402 is a general-purpose processor, digital signal processor, controller, artificial intelligence processor, application-specific integrated circuit, field programmable gate array, analog circuit, digital circuit, combinations thereof, or another currently known or later developed device for tracking, determining distortion, configuring the ultrasonic scanner 400, and / or generating a distorted image. The image processor 402 is configured by hardware, software, and / or firmware to detect anomalies, select settings, determine distortion, and / or output an image.

[0078] The image processor 402 is configured to detect the location of anomalies in the patient's heart wall motion. Ultrasonic data from scans such as scans of standardized views of the heart is used to detect regions with abnormal heart wall motion, either due to actual anomalies or poor data quality.

[0079] The image processor 402 is configured to determine the distortion of the heart wall motion at various locations along the heart wall. The image processor 402 detects the wall, such as by using artificial intelligence. The image processor 402 tracks the position along the wall through a series of images and provides the motion of the heart wall. The image processor 402 calculates the distortion from the tracking or wall motion. The distortion can be determined for various regions or segments in the original view.

[0080] When abnormal wall motion is detected, the image processor 402 is configured to cause the ultrasonic scanner 400 to scan a target patch around the region of abnormal wall motion. Using various values of one or more settings, improved ultrasonic data of the patch or region is obtained. The ultrasonic data is obtained from the scan using at least one of the various values of one or more settings as compared to the value of the setting for the original view.

[0081] The values may define various scan planes to use for scanning the region. The image processor 402 can control the ultrasonic scanner 400 to scan various planes of the patch and / or can instruct the ultrasonic examiner to change the direction of the transducer array 410. The region is scanned in two or more different planes based on the control by the image processor 402. Other settings, such as gain and / or contrast, may differ instead of or in addition to the field of view (depth, azimuth range, and / or scan plane) for rescan of the patch. The use of the settings of various values can result in various levels of speckle and / or out-of-plane motion. Ultrasonic data having greater speckle and / or less out-of-plane motion is used by the image processor 402 to determine the distortion of the region.

[0082] The display 420 is a monitor, a liquid crystal display, a plasma display, a light emitting diode display, a printer, a projector, or other device for outputting one or more images for viewing. Any currently known or later developed display device may be used. The display 420 can be part of the ultrasonic scanner 400, or can be a remote device such as a wall-mounted or remote workstation display.

[0083] The display 420 shows one or more images of the patient's heart. The images are generated as part of an echocardiogram. During the echocardiogram, the ultrasonic scanner 400 can also output guidance such as the rotation of the transducer array 410 on the display 420.

[0084] Display 420 shows one or more images having distortion information. The distorted images are generated by image processor 402 and output by display 420. The distorted images may be bull's-eye views from a multi-segment model. Other distortion maps showing distortion as a function of position along the heart wall may be output. 3D surface distortion may be output by surface or volume rendering or the like.

[0085] Memory 404 is a non-transitory computer-readable recording medium such as a cache, buffer, RAM, removable media, hard drive, or other computer-readable recording medium. Computer-readable recording media include various types of volatile and non-volatile recording media. Memory 404 stores ECG data, tracking, distortion, detected boundaries, abnormal regions, artificial intelligence (e.g., machine learning models), distorted images, or other information for guiding echocardiograms. Inputs, outputs, and / or information being processed are stored.

[0086] Memory 404 or another memory stores instructions for image processor 402 and / or other processors. Data representing instructions executable by a processor programmed for a stress echocardiogram guide is stored in the memory. The instructions are for processes, methods, and / or ways discussed herein. The functions, operations, or tasks shown in the figures or described herein are performed in response to one or more sets of instructions stored on a computer-readable recording medium or on a computer-readable recording medium. The functions, operations, or tasks are independent of a particular type of instruction set, recording medium, processor, or processing strategy and may be performed by software, hardware, integrated circuits, firmware, microcode, etc., operating alone or in combination. Similarly, the processing strategy can include multiprocessing, multitasking, parallel processing, etc. In one embodiment, the instructions are stored on a removable media device for reading by a local or remote system. In other embodiments, the instructions are stored at a remote location for transfer via a computer network or via a telephone line. In yet other embodiments, the instructions are stored within a given computer, CPU, GPU, or system.

[0087] Although the present invention has been described above with reference to various embodiments, it should be understood that many changes and modifications can be made without departing from the scope of the present invention. Accordingly, the above detailed description is to be construed as illustrative and not as limiting the present invention, and it is to be understood that the spirit and scope of the present invention are defined by the claims, which include all equivalents.

Claims

1. 1. A method for detecting cardiac wall motion with a medical ultrasound scanner, comprising: a first scan of a first plane through a patient's heart with the medical ultrasound scanner using first values ​​of scan settings, the first plane scanned at different times providing first ultrasound data representative of the heart at the different times; detecting abnormal wall motion from the first ultrasound data; a second scan of the region of abnormal wall motion in a second plane with the medical ultrasound scanner using second values ​​of the scan settings, the second values ​​being different from the first values ​​of the scan settings, the second plane being different from the first plane, the second scan yielding second ultrasound data; determining the abnormal wall motion distortion from the second ultrasound data; and displaying an image representative of said distortion; Including, The method, wherein the determining is performed by tracking speckles present in the second ultrasound data, and the determined strain is indicative of a change in distance between the speckles over the cardiac cycle.

2. the first scan includes a first scan of a standardized view of the heart; the second scan includes a second scan with a smaller field of view that just covers the region of abnormal wall motion; the region is smaller than a field of view in the standardized view; The method of claim 1.

3. The detection includes detection by a machine learning model. The method of claim 1.

4. The scan settings include gain, contrast, depth, and / or field of view. The method of claim 1.

5. the scan setup includes the field of view; the first value is for a larger field of view than the second value; the first value is for the first surface; the second value is for the second surface; The method according to claim 4.

6. selecting the second surface based on a level of dispersion and / or an amount of relative motion at the heart wall; The method of claim 1 further comprising:

7. a third scan of the area in a third plane different from the first plane and the second plane; The method of claim 1 further comprising:

8. The second scan includes a second scan with the second value that configures the medical ultrasound scanner to generate speckle. The method of claim 1.

9. determining the distortion includes determining the distortion for a segment of a multi-segment model; the representation includes a representation of the multi-segment model including the distortion; The method of claim 1.

10. determining the distortion includes determining the distortion in two or three dimensions of the first surface and the second surface; The display includes displaying a two-dimensional or three-dimensional distortion map. The method of claim 1.

11. For various regions of abnormal wall motion, the second scan; and said determination; Repeatedly, The method of claim 1 further comprising:

12. for various standardized views of the heart; the first scan; The detection. the second scan; and said determination; Repeatedly, Further comprising: the display includes a representation of the distortion from the various standardized views. The method of claim 1.

13. An ultrasound system for detecting cardiac wall motion, the cardiac wall comprising: locating a cardiac wall motion abnormality in a patient; scanning the location with various values ​​at one or more settings; determining a strain of motion of the heart wall at the location from ultrasound data from the scan using at least one of the different values ​​of the one or more settings; An ultrasound scanner configured as follows: a display configured to display an image of the distortion; and Equipped with The determining is performed by tracking speckles present in the ultrasound data, and the determined strain is indicative of a change in distance between the speckles over the cardiac cycle.

14. The one or more settings include: wherein the ultrasound scanner is configured to scan the locations in the various planes; For scanning various surfaces, 14. The ultrasound system of claim 13.

15. The one or more settings include: wherein the ultrasound scanner is configured to scan the location for imaging at different levels of speckle; gain, contrast, and / or depth; 14. The ultrasound system of claim 13.

16. The ultrasonic scanner includes: scanning the patient's heart with standardized views; Detecting the location from the standardized view; scanning the location with a field of view smaller than the standardized view; It is configured as follows:

14. The ultrasound system of claim 13.

17. 1. A non-transitory computer readable medium having stored thereon data representing instructions executable by a programmed processor for detecting cardiac wall motion, the non-transitory computer readable medium comprising: Scanning the patient's heart in standardized views; detecting abnormal wall motion of the heart from the standardized views; acquiring B-mode data in the region of abnormal wall motion; determining distortion from the B-mode data of the region; and displaying a distortion map including a first segment of the distortion map having a distortion determined from the B-mode data of the region and a second segment of the distortion map having a different distortion determined from the standardized view; Includes instructions for the determining is performed by tracking speckles present in the B-mode data, the determined strain being indicative of a change in distance between the speckles over a cardiac cycle. A non-transitory computer-readable recording medium.

18. the instructions to acquire the B-mode data in the region include acquiring the B-mode data with the region in a different plane than the standardized view; The instructions further include selecting the different surface based on an amount of speckle or motion.

20. The non-transitory computer readable storage medium of claim 17.

19. and the instructions to acquire the B-mode data include acquiring the B-mode data having a greater level of speckle in the region than the standardized view at the location of the abnormal wall motion.

20. The non-transitory computer readable storage medium of claim 17.

20. The instructions include instructions for various standardized views: The scanning; The detection. The acquisition; said determination; Further including repetition of the instructions for displaying the distortion map include displaying a 16, 17, or 18 segment bull's eye view.

20. The non-transitory computer readable storage medium of claim 17.

Images