Auxiliary diagnosis device and method for abnormal heart blood flow and storage medium

Abstract

The invention discloses an auxiliary diagnosis device and method for abnormal heart blood flow and a storage medium. The auxiliary diagnosis device for the abnormal heart blood flow comprises a monitoring control module used for monitoring whether a patient to be detected executes a Valsalva action or not in real time, and if yes, a pulse signal is sent; the image acquisition module is used for acquiring an image sequence of the to-be-detected patient in a right atrium pressure rise period based on the pulse signal, and performing target ROI extraction on the image sequence to obtain a target image sequence for displaying the target ROI; the light spot recognition module is used for detecting a microbubble signal in the target ROI of each frame of target image in the target image sequence in real time and counting the number of microbubbles in the target ROI of each frame of target image; the decision module is used for generating a corresponding diagnosis suggestion based on the number of microbubbles in the target ROI of each frame of target image; the full-process automation is realized, and the accuracy of heart blood flow abnormality diagnosis is improved; the diagnosis time is shortened, and the diagnosis efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical diagnosis, and in particular to an auxiliary diagnostic device, method and storage medium for cardiac blood flow abnormalities. Background Technology

[0002] Patent Foramen Ovale (PFO): The foramen ovale is a potential passageway in the atrial septum during embryonic development of the heart. It typically closes anatomically by one year of age. If it remains open after three years of age, it is called patent foramen ovale (PFO), with a prevalence of 20%-34% in the population. However, under certain conditions such as coughing or holding one's breath, the pressure in the right atrium increases. If this pressure exceeds the target pressure, a right-to-left shunt occurs, causing venous blood to enter the left ventricular system, resulting in abnormal cardiac blood flow. Small emboli are often generated and present in the venous system. These emboli may travel with the blood flow into the left ventricular system and further to various organs, causing organ embolism (e.g., stroke), known as paradoxical embolism. Patients with paradoxical embolism require aggressive treatment (e.g., closure). Therefore, determining the presence of cardiac blood flow abnormalities and the size of the shunt is crucial. Currently, an important clinical screening method is right heart echocardiography. The principle is to identify the amount of contrast agent microbubbles entering the left atrium and left ventricle. However, it requires a special breath-holding action (valsalva maneuver) to identify the frame with the largest amount of microbubbles in the left atrium and left ventricle at the moment of excitation, and to calculate the number of microbubbles in this frame to determine the size of the shunt flow.

[0003] Because capturing images of maximum microbubble volume while simultaneously monitoring the patient's breathing is difficult using conventional methods, or they may be unable to identify the section with the maximum microbubble volume, thus hindering accurate diagnosis of cardiac blood flow abnormalities. Therefore, more advanced methods are needed to improve the accuracy and efficiency of diagnosing cardiac blood flow abnormalities. Consequently, it is necessary to provide an auxiliary diagnostic device, method, and storage medium for cardiac blood flow abnormalities to address the aforementioned problems. Summary of the Invention

[0004] This invention provides an auxiliary diagnostic device, method, and storage medium for cardiac blood flow abnormalities, in order to solve the problems of low accuracy and efficiency in the diagnosis of cardiac blood flow abnormalities in the prior art.

[0005] In a first aspect, embodiments of the present invention provide an auxiliary diagnostic device for cardiac blood flow abnormalities, comprising:

[0006] The monitoring and control module is used to monitor in real time whether the patient under test performs the Valsalva maneuver. When the patient under test performs the Valsalva maneuver, a pulse signal is sent.

[0007] The image acquisition module is used to acquire image sequences of the patient under test during periods of increased right atrial pressure based on the pulse signal, and to extract the target ROI from the image sequences to obtain a target image sequence displaying the target ROI.

[0008] A spot recognition module is used to detect microbubble signals within the target ROI of each frame of the target image sequence in real time, and to count the number of microbubbles within the target ROI of each frame of the target image, wherein the target ROI includes the left atrium and the left ventricle;

[0009] The decision module is used to generate corresponding diagnostic suggestions based on the number of microbubbles within the target ROI of each frame of the target image and the anomalous embolism risk score.

[0010] Optionally, the image of the patient during a period of increased right atrial pressure is a standard cross-section of the patient's three-dimensional cardiac image. The image acquisition module includes:

[0011] The image switching unit is used to monitor the quality index of each frame in the image sequence during the period of increased right atrial pressure of the patient under test in real time. When the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard cross-section of the three-dimensional heart image of the patient under test at the same moment as the image.

[0012] Optionally, the monitoring and control module includes:

[0013] The monitoring unit is used to monitor the respiratory flow and intrathoracic pressure of the patient under test in real time through a respiratory flow sensor and a pressure sensor.

[0014] The control unit is used to compare the real-time monitored respiratory flow and intrathoracic pressure values ​​with preset respiratory flow thresholds and pressure thresholds, determine the comparison result, and determine whether the patient to be tested should perform the Valsalva maneuver based on the comparison result. If so, a pulse signal is sent.

[0015] Optionally, the image acquisition module includes:

[0016] The synchronous image acquisition unit is used to acquire an image sequence of the period when the right atrial pressure of the patient under test increases, based on the trigger time of the pulse signal and the right atrial pressure change curve.

[0017] The ROI extraction unit is used to extract the target ROI mask of each frame in the image sequence, and to overlay the image sequence with the ROI mask to obtain a target image sequence that displays the target ROI.

[0018] Optionally, the spot recognition module includes:

[0019] The microbubble detection unit is used to detect microbubble signals within the target ROI of each frame of the target image sequence by combining a dynamic threshold segmentation algorithm with a lightweight convolutional neural network algorithm, and to visualize the signal.

[0020] The statistics unit is used to count the number of microbubbles within the target ROI region of each frame of the target image.

[0021] Optionally, the decision module includes:

[0022] The microbubble number grading unit is used to determine the corresponding microbubble level based on the number of microbubbles in the target ROI of each frame of the target image, according to a preset grading rule.

[0023] The diagnostic generation unit is used to generate corresponding diagnostic recommendations based on the microbubble level and the anomalous embolism risk score.

[0024] Optionally, the ROI extraction unit includes:

[0025] The image preprocessing subunit is used to preprocess the image sequence of the patient under test during the period of increased right atrial pressure;

[0026] The mask extraction subunit is used to extract the target ROI mask of each frame in the preprocessed image sequence.

[0027] The mask post-processing subunit is used to optimize the target ROI mask for each frame of the image;

[0028] The mask overlay subunit is used to overlay the optimized ROI mask of each frame image onto the preprocessed image sequence to obtain the target image sequence displaying the target ROI.

[0029] Secondly, an auxiliary diagnostic method for cardiac blood flow abnormalities is provided, executed by the cardiac blood flow abnormality auxiliary diagnostic device described in any embodiment of the present invention, comprising:

[0030] The system monitors in real time whether the patient under test performs the Valsalva maneuver. When the patient under test performs the Valsalva maneuver, a pulse signal is sent.

[0031] Based on the pulse signal, an image sequence of the patient under test during the period of increased right atrial pressure is acquired, and the target ROI is extracted from the image sequence to obtain a target image sequence displaying the target ROI, wherein the target ROI includes the left atrium and the left ventricle;

[0032] Real-time detection of microbubble signals within the target ROI of each frame of the target image sequence, and counting of the number of microbubbles within the target ROI of each frame of the target image;

[0033] Based on the number of microbubbles within the target ROI of each frame of the target image and the anomalous embolism risk score, corresponding diagnostic suggestions are generated.

[0034] Optionally, the image of the patient under test during the period of increased right atrial pressure is a standard cross-section of the three-dimensional image of the patient's heart. After acquiring the image sequence of the patient under test during the period of increased right atrial pressure based on the pulse signal, the method further includes:

[0035] The image quality index of each frame in the image sequence during the period of increased right atrial pressure of the patient under test is monitored in real time. When the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard cross-section of the three-dimensional heart image of the patient under test at the same moment.

[0036] Thirdly, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being used to cause a processor to execute and implement the auxiliary diagnostic method for cardiac blood flow abnormalities according to any embodiment of the present invention.

[0037] This invention addresses the shortcomings of the prior art and has the following beneficial effects:

[0038] (1) This invention provides an auxiliary diagnostic device for cardiac blood flow abnormalities. Through a monitoring and control module, it monitors in real time whether the patient to be tested performs the Valsalva maneuver and sends pulse signals; through an image acquisition module, it acquires image sequences of the patient to be tested during periods of increased right atrial pressure and extracts the target ROI to obtain a target image sequence displaying the target ROI; through a spot recognition module, it detects microbubble signals within the target ROI in real time and counts the number of microbubbles; through a decision module, it generates corresponding diagnostic suggestions based on the number of microbubbles combined with an abnormal embolism risk score; it not only achieves full-process automation, reduces the impact of operator experience differences on the results, and effectively improves the accuracy of cardiac blood flow abnormality diagnosis; but also shortens the diagnosis time and improves diagnostic efficiency.

[0039] (2) Through the monitoring and control module, precise motion capture is achieved, and the Valsalva maneuver is monitored in real time by the sensor to avoid missed detection due to poor patient cooperation; through the image acquisition module, dynamic ROI tracking is achieved, and the target region is segmented in real time using deep learning (such as U-Net), wherein the target ROI includes the left atrium and left ventricle, overcoming the influence of heart pulsation and respiratory displacement; through the spot recognition module, microbubble signals in the target ROI are detected in real time, and the number of microbubbles is counted to intuitively display the severity of shunt; through the decision module, the number of microbubbles is combined with clinical parameters to directly generate corresponding diagnostic suggestions, thereby achieving precise diagnosis for different patients to be tested and providing corresponding diagnostic suggestions, which not only improves the accuracy of diagnosis but also improves the efficiency of diagnosis.

[0040] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of the structure of an auxiliary diagnostic device for cardiac blood flow abnormalities provided in Embodiment 1 of the present invention;

[0043] Figure 2 This is a schematic diagram of the specific structure of an ROI extraction unit provided in Embodiment 2 of the present invention;

[0044] Figure 3 This is a flowchart of an auxiliary diagnostic method for cardiac blood flow abnormalities provided in Embodiment 3 of the present invention. Detailed Implementation

[0045] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0046] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0047] Example 1:

[0048] Figure 1 This is a schematic diagram of a cardiac blood flow abnormality auxiliary diagnostic device provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where cardiac blood flow abnormalities are diagnosed using the cardiac blood flow abnormality auxiliary diagnostic device, such as... Figure 1 As shown, the device includes:

[0049] The monitoring and control module 10 is used to monitor in real time whether the patient to be tested performs the Valsalva action. When the patient to be tested is detected to perform the Valsalva action, a pulse signal is sent.

[0050] In this context, the patient being tested can be understood as someone with a structurally abnormal heart, specifically a patent foramen ovale (PFO). The Valsalva maneuver can refer to a physiological action that alters the pressure in the thoracic and abdominal cavities through a specific breathing pattern. Specifically, the patient takes a deep breath, closes the glottis (pinching the nostrils or closing the mouth), and forcefully exhales to increase intrathoracic pressure while maintaining abdominal muscle contraction (similar to defecation). The pulse signal can be understood as an external trigger command used to activate the ultrasound equipment to acquire images.

[0051] Specifically, sensors can collect signals such as respiratory flow and abdominal pressure from the patient in real time. A preset algorithm analyzes whether the signals conform to the typical pattern of the Valsalva maneuver, such as intrathoracic pressure exceeding a set threshold or respiratory flow exceeding a set threshold. When the patient is detected performing the Valsalva maneuver, a pulse signal is sent via wired / wireless (e.g., Bluetooth, WiFi) to a contrast transthoracic chocardiography (cTTE) device, synchronizing the patient's physiological movement (Valsalva) with ultrasound imaging to ensure that key images are captured at the moment when the patient's right ventricular pressure increases or a potential shunt occurs.

[0052] The image acquisition module 11 is used to acquire an image sequence of the patient under test during the period of increased right atrial pressure based on the pulse signal, and to extract the target ROI from the image sequence to obtain a target image sequence displaying the target ROI, wherein the target ROI includes the left atrium and the left ventricle;

[0053] The period of increased right atrial pressure can be understood as the stage in which, due to the Valsalva maneuver, the pressure in the right atrium is significantly higher than normal when the foramen ovale is open; the image sequence can be understood as a set of images arranged chronologically during the period of increased right atrial pressure; the target ROI can be understood as the left atrial and left ventricular regions extracted from cardiac images through image segmentation or manual annotation; and the target image sequence can refer to the image sequence displaying the left atrial and left ventricular regions.

[0054] Specifically, by simultaneously triggering the cTTE device with a pulse signal and injecting an ultrasound contrast agent (e.g., a mixture of saline, blood, and air microbubble contrast agent) into the patient, the contrast and resolution of ultrasound imaging can be enhanced, helping doctors to more clearly observe the blood flow and structure of tissues or organs. The timestamps carried by the pulse signal can be used to precisely align physiological signals (such as pressure curves) with ultrasound image frames. Using the cTTE device combined with microbubble contrast technology, image sequences of periods of increased right atrial pressure in the patient can be acquired. Using traditional image processing methods or deep learning methods, the target ROI mask for each frame of the image sequence corresponding to the increased right atrial pressure stage is determined. The corresponding transparent target ROI mask (e.g., a red outline) is then superimposed on the original ultrasound image sequence to obtain a target image sequence displaying the target ROI.

[0055] The spot recognition module 12 is used to detect microbubble signals in the target ROI of each frame of the target image sequence in real time, and count the number of microbubbles in the target ROI of each frame of the target image.

[0056] Among them, microbubbles refer to the tiny bubbles that appear in the heart chambers after the contrast agent is injected during right heart ultrasound imaging. Similar to blood flow staining, they are used to indicate whether blood flow in the right heart system has entered the left heart system through the abnormal pathway (PFO). The number of microbubbles refers to the total number of microbubbles observed in the target ROI. It is a key indicator for assessing the severity of right-to-left shunt. The higher the number, the more significant the pathological significance.

[0057] Specifically, the target ROI can be extracted directly from the target image sequence based on the target ROI mask. By combining a dynamic threshold segmentation algorithm with a lightweight convolutional neural network algorithm, microbubble signals within the target ROI in each frame of the target image can be detected. Furthermore, the number of microbubbles within the target ROI in each frame of the target image can be directly counted using a counter.

[0058] The decision module 13 is used to generate corresponding diagnostic suggestions based on the number of microbubbles in the target ROI of each frame of the target image and the anomalous embolism risk score.

[0059] The Risk of Paradoxical Embolism (RoPE) is a clinical tool used to assess the correlation between cryptogenic stroke (CSS) and patent foramen ovale (PFO), aiming to help determine whether PFO is a cause of stroke and predict the risk of recurrence.

[0060] Specifically, the number of microbubbles can be graded based on the average total number of microbubbles in the target image sequence. Then, based on the microbubble grade and the RoPE score, corresponding diagnostic suggestions are generated using pre-set diagnostic suggestion generation rules. The average total number of microbubbles in the target image sequence can be obtained by dividing the sum of the number of microbubbles within the target ROI in each frame of the target image sequence by the number of frames in the target image sequence.

[0061] This embodiment provides an auxiliary diagnostic device for cardiac blood flow abnormalities. The device includes: a monitoring and control module for real-time monitoring of whether the patient performs the Valsalva maneuver; when the Valsalva maneuver is detected, a pulse signal is sent; an image acquisition module for acquiring image sequences of periods of increased right atrial pressure in the patient based on the pulse signal, and extracting target areas of interest (ROIs) from the image sequences to obtain a target image sequence displaying the target ROIs; a spot recognition module for real-time detection of microbubble signals within the target ROIs of each frame of the target image sequence, and counting the number of microbubbles within the target ROIs of each frame; and a decision module for generating corresponding diagnostic suggestions based on the number of microbubbles within the target ROIs of each frame, combined with an abnormal embolism risk score. This device solves the problems of low accuracy and efficiency in diagnosing cardiac blood flow abnormalities in existing technologies. Through full-process automation, it reduces the impact of operator experience differences on the results, effectively improving the accuracy of cardiac blood flow abnormality diagnosis; and further shortens the diagnostic time and improves diagnostic efficiency.

[0062] Optionally, the image of the patient during the period of increased right atrial pressure is a standard cross-section of the three-dimensional cardiac image of the patient. The image acquisition module 11 includes:

[0063] The image switching unit 111 is used to monitor the quality index of each frame of the image sequence during the period of increased right atrial pressure of the patient under test in real time. When the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard cross-section of the three-dimensional heart image of the patient under test at the same moment as the image.

[0064] The standard section refers to a section that clearly displays the heart's structure and hemodynamic changes. The quality index refers to image sharpness, such as edge sharpness (Laplacian variance) and contrast-to-noise ratio (CNR); the preset threshold can be dynamically set according to actual conditions.

[0065] Specifically, a database of standard cross-sections of the patient's 3D cardiac images can be pre-constructed and aligned with timestamps; this database includes at least two backup standard cross-sections. Image sequences during periods of increased right atrial pressure in the patient can be monitored in real time, and the quality index of each frame in the sequence can be calculated. When the image quality index is less than a preset threshold, the image with the lower index is switched to another standard cross-section of the patient's 3D cardiac images from the same time (i.e., the same timestamp), and backup cross-sections are selected according to a predefined priority order in the database. Specifically, interpolation algorithms (such as linear blending) can be used to achieve seamless transitions between cross-sections, ensuring a smooth transition and avoiding visual jumps. If the image quality index is greater than or equal to the preset threshold, the frame is retained.

[0066] In this embodiment, by real-time monitoring of the quality index of each frame in the image sequence during the period of increased right atrial pressure in the patient under test, when the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard section of the three-dimensional heart image of the patient under test at the same moment as the image. This realizes real-time monitoring of image quality and intelligent section switching during the period of increased right atrial pressure, effectively improving the efficiency and accuracy of clinical diagnosis.

[0067] Optionally, the monitoring and control module 10 includes:

[0068] The monitoring unit 101 is used to monitor the respiratory flow and intrathoracic pressure of the patient under test in real time through a respiratory flow sensor and a pressure sensor.

[0069] The control unit 102 is used to compare the real-time monitored respiratory flow and intrathoracic pressure values ​​with preset respiratory flow thresholds and pressure thresholds, determine the comparison result, and determine whether the patient to be tested performs the Valsalva maneuver based on the comparison result. If so, a pulse signal is sent.

[0070] The respiratory flow sensor can refer to a differential pressure or thermal flow meter, installed in the breathing tubing or mask to measure expiratory / inspiratory flow in real time. The pressure sensor can refer to a thoracic pressure sensor, specifically obtained through an esophageal cuff catheter (indirect measurement) or a chest drainage tube (direct measurement) to obtain intrathoracic pressure values. The respiratory flow threshold can refer to a set lower limit for breath-hold flow, for example, 0.5 L / s; the trigger condition is a sudden drop in flow. The pressure threshold can refer to a set upper limit for intrathoracic pressure, for example, 20 cmH2O. The comparison results can be: the patient's respiratory flow is less than the respiratory flow threshold, the patient's respiratory flow is equal to the respiratory flow threshold, or the patient's respiratory flow is greater than the respiratory flow threshold; the intrathoracic pressure value is less than the pressure threshold, the intrathoracic pressure value is equal to the pressure threshold, or the intrathoracic pressure value is greater than the pressure threshold.

[0071] Specifically, respiratory flow (in L / s) and intrathoracic pressure (in cmH2O) of the patient under test can be monitored and collected in real time using respiratory flow sensors and pressure sensors. The real-time collected respiratory flow and intrathoracic pressure values ​​are compared with preset respiratory flow thresholds and pressure thresholds. If the patient's respiratory flow is less than the respiratory flow threshold or the intrathoracic pressure is greater than the pressure threshold, the Valsalva action is triggered. A high-level pulse (such as a 5V TTL signal, lasting 100ms) is output through GPIO, or a trigger command is sent through a wireless module (such as Bluetooth / WiFi).

[0072] In this embodiment, the monitoring unit in the monitoring and control module monitors the respiratory flow and intrathoracic pressure of the patient under test in real time. The control unit in the monitoring and control module compares the real-time monitored respiratory flow and intrathoracic pressure with the preset respiratory flow threshold and pressure threshold, determines the comparison result, and determines whether the patient under test performs the Valsalva maneuver based on the comparison result. If so, a pulse signal is sent, which can accurately identify the Valsalva maneuver and drive the cTTE device through the pulse signal to meet the needs of medical monitoring and automated control.

[0073] Optionally, the image acquisition module 11 includes:

[0074] The synchronous image acquisition unit 112 is used to acquire an image sequence of the period when the right atrial pressure of the patient under test increases, based on the trigger time of the pulse signal and the right atrial pressure change curve.

[0075] The ROI extraction unit 113 is used to extract the target ROI mask of each frame in the image sequence, and superimpose the image sequence with the ROI mask to obtain a target image sequence that displays the target ROI.

[0076] The trigger time of the pulse signal can refer to the timestamp of when the pulse signal is sent via GPIO or a wireless module, which can be used for subsequent alignment of the image with the pressure curve. The right atrial pressure change curve refers to the curve showing the increase in right atrial pressure over time during Valsalva maneuvers; the vertical axis of the right atrial pressure change curve can be the pressure value, and the horizontal axis can be the time value. The target ROI mask refers to a masking or labeling tool used to identify the region of interest (ROI) in an image, typically represented as a binary mask (binary image), bounding box, polygon, or other form, specifically used to extract the left atrium and left ventricle regions from the image.

[0077] Specifically, a hardware synchronization interface (e.g., DICOM timestamp) can be used to strictly align the pulse signal with the time axis of the right atrial pressure change curve, recording the pulse signal trigger time (T_trigger) and the sampling timestamp of the right atrial pressure change curve to establish a unified time reference. The baseline pressure (e.g., the resting average) and pressure change rate (dRAP / dt) of the right atrial pressure change curve are calculated. When the right atrial pressure exceeds the sum of the baseline pressure value and a preset pressure change amount, it is determined to be a period of right atrial pressure elevation, where the pressure change amount can be set to ΔP = 5 mmHg. The pressure change rate can also be combined to avoid false triggering due to short-term fluctuations; for example, dRAP / dt should be >0 and exceed a threshold for 1 second. Furthermore, images should be continuously acquired for N seconds (e.g., 3 seconds) starting from the pulse signal trigger time T_trigger to cover the entire pressure elevation process. The cTTE device is activated via an external pulse signal trigger interface (e.g., TTL signal), and an ultrasound contrast agent is simultaneously injected into the patient to acquire an image sequence of the right atrial pressure elevation period in the patient being tested. In addition, when acquiring image sequences of the patient under test during the period of increased right atrial pressure, considering the actual delay from pulse signal to imaging (e.g., device response time, data transmission delay), compensation time can be reserved before the trigger time T_trigger, for example, triggering 50ms in advance.

[0078] Furthermore, a pre-built deep learning segmentation model (e.g., U-Net) can be used to automatically extract the target ROI mask for each frame of an image sequence. This deep learning segmentation model can also incorporate attention mechanisms (e.g., CBAM, Transformer, etc.) to enhance sensitivity to target boundaries. Specifically, the deep learning segmentation model can be trained using medical images of the same modality, for example, using a large-scale cardiac ultrasound dataset. After obtaining the target ROI mask for each frame of the image sequence, a transparency overlay method can be used. The specific steps include: reading the image sequence and the target ROI mask for each frame; ensuring the mask is a binary image; extracting a pre-set contour color, for example, color = (0,0,255) represents red (BGR format), which can be changed to other colors (e.g., green (0,255,0)); blending the image with the target ROI mask based on a pre-set transparency, such as 0.5; and obtaining a target image sequence displaying the left atrium and left ventricle ROIs. Regarding transparency control: the alpha value is adjustable (0.3–0.7), balancing the visibility of the original image and the outline. The outline line width can be set to thickness=2, which controls the thickness of the outline lines.

[0079] In this embodiment, the synchronous image acquisition unit in the image acquisition module acquires the image sequence during the period of increased right atrial pressure in the patient under test, achieving precise synchronization between the Valsalva maneuver and image acquisition, effectively capturing the image sequence during the period of increased right atrial pressure, and providing accurate image support for the diagnosis of cardiac blood flow abnormalities. The ROI extraction unit in the image acquisition module extracts the target ROI mask of each frame in the image sequence, and superimposes the image sequence with the ROI mask to obtain a target image sequence displaying the target ROI. This efficiently generates a target image sequence with a transparent outline of the target ROI, ensuring that microbubble detection is strictly limited to the target ROI region, which not only improves the accuracy of subsequent microbubble detection and avoids false detections, but also improves the efficiency of microbubble detection.

[0080] Optionally, the spot recognition module 12 includes:

[0081] The microbubble detection unit 121 is used to detect microbubble signals within the target ROI of each frame of the target image sequence by combining a dynamic threshold segmentation algorithm with a lightweight convolutional neural network algorithm, and to visualize the signal.

[0082] The statistics unit 122 is used to count the number of microbubbles within the target ROI region of each frame of the target image.

[0083] Among them, dynamic thresholding segmentation algorithms can refer to the Otsu's algorithm, an adaptive thresholding technique based on the global image grayscale distribution. In ultrasound images, the Otsu's algorithm dynamically calculates the threshold based on the global grayscale distribution, binarizing the image (separating the foreground and background), making it suitable for scenes with uneven lighting or low contrast. Combined with the microbubble characteristics—microbubble signals typically appear as bright spots—the Otsu's algorithm can quickly locate candidate regions, reducing the input range of subsequent CNNs and improving detection efficiency. Lightweight convolutional neural network algorithms can refer to optimized object detection models for resource-constrained scenarios, aiming to significantly reduce computational load and memory consumption while maintaining high detection accuracy.

[0084] Specifically, the Otsu algorithm is used to binarize each frame of the image, extract the target ROI (Region of Interest), filter out most background noise, quickly locate the ROI, and reduce computational load. The ROIs segmented by Otsu are then cropped and scaled to a fixed size (e.g., 128×128 pixels) and input into the trained lightweight model YOLOv8-Micro. YOLOv8-Micro detects microbubbles based on their features and outputs the bounding boxes and confidence scores of microbubbles within the ROI, avoiding full-image computation. After obtaining the bounding boxes and confidence scores of microbubbles within the ROI, contours or circles are drawn to mark the microbubbles within the ROI region of each frame of the target image. These marked images are then overlaid onto the original target image sequence and visualized. A counter is used to count the number of microbubbles within the target ROI region of each frame of the target image sequence.

[0085] In this embodiment, the microbubble detection unit in the spot recognition module detects microbubble signals within the target ROI of each frame of the target image sequence using a dynamic threshold segmentation algorithm combined with a lightweight convolutional neural network algorithm, and displays them visually. The statistics unit in the spot recognition module is used to count the number of microbubbles within the target ROI region of each frame of the target image, which can efficiently and accurately realize the real-time detection and statistics of microbubble signals within the ROI region, meeting the needs of medical image analysis.

[0086] Optionally, the decision module 13 includes:

[0087] The microbubble number grading unit 131 is used to determine the corresponding microbubble level based on the number of microbubbles in the target ROI of each frame of the target image and according to a preset grading rule.

[0088] The diagnostic generation unit 132 is used to generate corresponding diagnostic suggestions based on the microbubble level and the anomalous embolism risk score.

[0089] The preset grading rules can refer to the rules for classifying levels based on the number of microbubbles. The microbubble level can be determined based on the number of microbubbles corresponding to the target image with the most microbubbles in the target image sequence, and can be divided into three levels: small amount, medium amount, and large amount.

[0090] Specifically, the preset grading rules can be as follows: Identify the single target image with the highest number of microbubbles within the target ROI in the target image sequence. If the number of microbubbles in this frame is 1-10, the microbubble level is determined to be small; if the number of microbubbles in this frame is 11-30, the microbubble level is determined to be medium; and if the number of microbubbles in this frame is greater than 30, the microbubble level is determined to be large. Table 1 shows the anomalous embolism risk scoring table.

[0091] Table 1

[0092]

[0093] Based on the microbubble grades and the paradoxical embolism risk score, corresponding diagnostic recommendations are generated, as detailed in Tables 2-4. Table 2 shows a small number of microbubble grades combined with the RoPE score.

[0094] Table 2

[0095]

[0096] Table 3 shows the scores for medium-volume microbubbles + RoPE, as shown in Table 3:

[0097] Table 3

[0098]

[0099] Table 4 shows the scores for a large number of microbubbles + RoPE, as shown in Table 4:

[0100] Table 4

[0101]

[0102] In this embodiment, the microbubble quantity grading unit in the decision module determines the corresponding microbubble level according to preset grading rules; the diagnosis generation unit in the decision module generates corresponding diagnostic suggestions based on the microbubble level and the abnormal embolism risk score. This not only improves the accuracy and efficiency of diagnosing abnormal cardiac blood flow, but also achieves precise diagnosis for different patients by combining microbubble quantity with clinical parameters, and provides corresponding diagnostic suggestions.

[0103] Example 2:

[0104] Figure 2This is a schematic diagram of the specific structure of an ROI extraction unit provided in Embodiment 2 of the present invention. Based on the above embodiments, preferred embodiments are provided. For example... Figure 2 As shown, the ROI extraction unit 113 includes:

[0105] Image preprocessing subunit 1131 is used to preprocess the image sequence of the patient under test during the period of increased right atrial pressure.

[0106] Preprocessing can refer to standardization processes, such as image denoising and image enhancement.

[0107] Specifically, Gaussian filtering (kernel size 3×3, σ=1) can be used to remove high-frequency noise from each frame of the image sequence. Adaptive histogram equalization (CLAHE, cropping limit = 3.0) is used to enhance the contrast of each frame, especially for low-brightness areas. Then, based on rigid registration of feature points (e.g., heart chamber contours), consecutive frames are aligned to compensate for breathing or slight displacement. The pixel values ​​of each frame in the image sequence are normalized to the range [0,1] to unify the image intensity distribution of the image sequence.

[0108] The mask extraction subunit 1132 is used to extract the target ROI mask of each frame in the preprocessed image sequence;

[0109] Specifically, the target ROI mask for each frame of the standardized image sequence is extracted using a deep learning segmentation model.

[0110] The mask post-processing subunit 1133 is used to optimize the target ROI mask of each frame image;

[0111] Among them, optimization can refer to optimizing the mask topology to eliminate noise and artifacts.

[0112] Specifically, small holes are filled and broken boundaries are connected through closing operations (structuring element = 5×5 ellipse); isolated noise points (such as vascular adhesion areas) are removed through opening operations to achieve morphological restoration of the target ROI mask. Mask edges are smoothed using two-dimensional Gaussian filtering (σ = 0.5) to reduce jagged edges; topological correction based on level sets repairs depressions or protrusions to achieve smooth boundary restoration of the target ROI mask. Kalman filtering is applied to masks in consecutive frames to predict the position of the next frame and correct deviations; frames with abrupt area changes (such as those exceeding 20% ​​difference between adjacent frames) are removed and repaired through interpolation; comprehensively, the target ROI mask of each frame in the image sequence is optimized.

[0113] The mask overlay subunit 1134 is used to overlay the optimized ROI mask of each frame image onto the preprocessed image sequence to obtain the target image sequence displaying the target ROI.

[0114] Specifically, a transparency overlay method is used to overlay the ROI mask of each optimized frame image onto the preprocessed image sequence to obtain a target image sequence that displays the target ROI.

[0115] In this embodiment, the image preprocessing subunit in the ROI extraction unit is used to perform standardized preprocessing on the image sequence during the period of elevated right atrial pressure in the patient under test, which can effectively improve the subsequent segmentation accuracy. The mask extraction subunit in the ROI extraction unit extracts the target ROI mask of each frame in the preprocessed image sequence. The mask postprocessing subunit in the ROI extraction unit optimizes the target ROI mask of each frame by performing morphological restoration, boundary smoothing and other optimizations, which not only balances noise suppression but also preserves the details of the target ROI. The mask overlay subunit in the ROI extraction unit overlays the optimized ROI mask of each frame onto the preprocessed image sequence to obtain the target image sequence displaying the target ROI.

[0116] Example 3:

[0117] Figure 3 This is a flowchart of an auxiliary diagnostic method for cardiac blood flow abnormalities provided in Embodiment 3 of the present invention. This embodiment is applicable to situations where cardiac blood flow abnormalities are diagnosed using an auxiliary diagnostic device. The method can be executed by the auxiliary diagnostic device; this device can be implemented in hardware and / or software. Figure 3 As shown, the method includes:

[0118] S310. Monitor in real time whether the patient to be tested performs the Valsalva action. When the patient to be tested is detected to perform the Valsalva action, send a pulse signal.

[0119] Specifically, the respiratory flow (in L / s) and intrathoracic pressure (in cmH2O) of the patient under test are monitored and collected in real time using respiratory flow and pressure sensors. The real-time collected respiratory flow and intrathoracic pressure values ​​are compared with preset respiratory flow and pressure thresholds. If the patient's respiratory flow is less than the respiratory flow threshold or the intrathoracic pressure is greater than the pressure threshold, the Valsalva action is triggered. A high-level pulse (such as a 5V TTL signal lasting 100ms) is output through GPIO, or a trigger command is sent through a wireless module (such as Bluetooth / WiFi).

[0120] S320. Based on the pulse signal, acquire the image sequence of the period when the right atrial pressure of the patient under test increases, and extract the target ROI from the image sequence to obtain the target image sequence displaying the target ROI.

[0121] The target ROI includes the left atrium and the left ventricle.

[0122] Specifically, the pulse signal and the time axis of the right atrial pressure change curve can be strictly aligned using timestamps. The trigger time of the pulse signal (e.g., T_trigger) and the sampling timestamp of the right atrial pressure change curve are recorded to establish a unified time reference. The baseline pressure (e.g., the resting average) and the rate of change of pressure (dRAP / dt) of the right atrial pressure change curve are calculated. When the right atrial pressure is greater than the sum of the baseline pressure and a preset pressure change, it is determined to be a period of right atrial pressure elevation, where the pressure change can be set to ΔP = 5 mmHg. The cTTE device is activated via an external trigger interface (e.g., TTL signal) of the pulse signal, and ultrasound contrast agent is simultaneously injected into the patient to acquire an image sequence of the right atrial pressure elevation period. Furthermore, considering the actual delay from pulse signal to imaging, compensation time can be reserved before the trigger time T_trigger, for example, triggering 50ms in advance. The target ROI mask for each frame of the image sequence of the right atrial pressure elevation period of the patient is automatically extracted using a pre-built deep learning segmentation model (e.g., U-Net).

[0123] S330. Real-time detection of microbubble signals within the target ROI of each frame of the target image sequence, and counting the number of microbubbles within the target ROI of each frame of the target image.

[0124] Specifically, the Otsu algorithm is used to binarize each frame of the image, extract the target ROI (Region of Interest), filter out most background noise, quickly locate the ROI, and reduce computational load. The ROIs segmented by Otsu are then cropped and scaled to a fixed size (e.g., 128×128 pixels) and input into the trained lightweight model YOLOv8-Micro. YOLOv8-Micro detects microbubbles based on their features and outputs the bounding boxes and confidence scores of microbubbles within the ROI, avoiding full-image computation. After obtaining the bounding boxes and confidence scores of microbubbles within the ROI, contours or circles are drawn to mark the microbubbles within the ROI region of each frame of the target image. These marked images are then overlaid onto the original target image sequence and visualized. A counter is used to count the number of microbubbles within the target ROI region of each frame of the target image sequence.

[0125] S340. Based on the number of microbubbles within the target ROI of each frame of the target image and the anomalous embolism risk score, generate corresponding diagnostic suggestions.

[0126] Specifically, based on the number of microbubbles within the target ROI of each frame of the target image, the average total number of microbubbles in the target image sequence can be calculated. According to a preset grading rule, the microbubble level corresponding to the target image sequence is determined. By combining the microbubble level corresponding to the target image sequence with an abnormal embolism risk score, a diagnostic recommendation for the patient under test can be determined.

[0127] In this embodiment, by real-time monitoring of whether the patient performs the Valsalva maneuver, a pulse signal is sent when the patient performs the Valsalva maneuver. Based on the pulse signal, an image sequence of the patient during periods of increased right atrial pressure is acquired, and the target area of ​​interest (ROI) is extracted from the image sequence to obtain a target image sequence displaying the target ROI. The microbubble signal within the target ROI of each frame of the target image sequence is detected in real-time, and the number of microbubbles within the target ROI of each frame is counted. Based on the number of microbubbles within the target ROI of each frame and the paradoxical embolism risk score, corresponding diagnostic suggestions are generated, achieving precise diagnosis for different patients and providing corresponding diagnostic suggestions, effectively improving the accuracy of diagnosing cardiac blood flow abnormalities, shortening the diagnosis time, and improving diagnostic efficiency.

[0128] Optionally, the image of the patient under test during the period of increased right atrial pressure is a standard cross-section of the three-dimensional image of the patient's heart. After acquiring the image sequence of the patient under test during the period of increased right atrial pressure based on the pulse signal, the method further includes:

[0129] The image quality index of each frame in the image sequence during the period of increased right atrial pressure of the patient under test is monitored in real time. When the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard cross-section of the three-dimensional heart image of the patient under test at the same moment.

[0130] Specifically, the image sequence of the patient under test during the period of increased right atrial pressure is monitored in real time, and the quality index of each frame in the sequence is calculated. When the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard section of the three-dimensional heart image of the patient under test at the same time (i.e., the same timestamp) using an interpolation algorithm. The backup section is selected according to the predefined priority order in the pre-constructed standard section database of the three-dimensional heart image of the patient under test.

[0131] In this embodiment, by real-time monitoring of the quality index of each frame in the image sequence during the period of increased right atrial pressure in the patient under test, when the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard section of the three-dimensional heart image of the patient under test at the same moment as the image. This realizes real-time monitoring of image quality and intelligent section switching during the period of increased right atrial pressure, effectively improving the efficiency and accuracy of clinical diagnosis.

[0132] Example 4:

[0133] In some embodiments, the auxiliary diagnostic method for cardiac blood flow abnormalities may be implemented as a computer program tangibly contained in a computer-readable storage medium. In some embodiments, part or all of the computer program may be loaded and / or installed onto the auxiliary diagnostic device for cardiac blood flow abnormalities via ROM and / or communication lines. When the computer program is loaded into RAM and executed under the control of the auxiliary diagnostic device for cardiac blood flow abnormalities, one or more steps of the auxiliary diagnostic method for cardiac blood flow abnormalities described above may be performed. Alternatively, in other embodiments, the auxiliary diagnostic device for cardiac blood flow abnormalities may be configured to perform the auxiliary diagnostic method for cardiac blood flow abnormalities by any other suitable means (e.g., by means of firmware).

[0134] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems-on-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0135] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0136] In the context of this invention, a computer-readable storage medium stores computer instructions that, when executed by a processor, implement the auxiliary diagnostic method for cardiac blood flow abnormalities provided by this invention. The computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0137] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD)) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0138] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0139] A computing system can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is established by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0140] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0141] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An auxiliary diagnostic device for cardiac blood flow abnormalities, characterized in that, include: The monitoring and control module is used to monitor in real time whether the patient under test performs the Valsalva maneuver. When the patient under test performs the Valsalva maneuver, a pulse signal is sent. The image acquisition module is used to acquire an image sequence of the patient under test during the period of right atrial pressure elevation based on the pulse signal, and to extract the target ROI from the image sequence to obtain a target image sequence displaying the target ROI, wherein the target ROI includes the left atrium and the left ventricle; The spot recognition module is used to detect microbubble signals within the target ROI of each frame of the target image sequence in real time, and to count the number of microbubbles within the target ROI of each frame of the target image. The decision module is used to generate corresponding diagnostic suggestions based on the number of microbubbles within the target ROI of each frame of the target image and the anomalous embolism risk score.

2. The auxiliary diagnostic device for cardiac blood flow abnormalities according to claim 1, characterized in that, The image of the patient during the period of increased right atrial pressure is a standard cross-section of the patient's three-dimensional cardiac image. The image acquisition module includes: The image switching unit is used to monitor the quality index of each frame in the image sequence during the period of increased right atrial pressure of the patient under test in real time. When the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard cross-section of the three-dimensional heart image of the patient under test at the same moment as the image.

3. The auxiliary diagnostic device for cardiac blood flow abnormalities according to claim 1, characterized in that, The monitoring and control module includes: The monitoring unit is used to monitor the respiratory flow and intrathoracic pressure of the patient under test in real time through a respiratory flow sensor and a pressure sensor. The control unit is used to compare the real-time monitored respiratory flow and intrathoracic pressure values ​​with preset respiratory flow thresholds and pressure thresholds, determine the comparison result, and determine whether the patient to be tested should perform the Valsalva maneuver based on the comparison result. If so, a pulse signal is sent.

4. The auxiliary diagnostic device for cardiac blood flow abnormalities according to claim 1, characterized in that, The image acquisition module includes: The synchronous image acquisition unit is used to acquire an image sequence of the period when the right atrial pressure of the patient under test increases, based on the trigger time of the pulse signal and the right atrial pressure change curve. The ROI extraction unit is used to extract the target ROI mask of each frame in the image sequence, and to overlay the image sequence with the ROI mask to obtain a target image sequence that displays the target ROI.

5. The auxiliary diagnostic device for cardiac blood flow abnormalities according to claim 1, characterized in that, The light spot recognition module includes: The microbubble detection unit is used to detect microbubble signals within the target ROI of each frame of the target image sequence by combining a dynamic threshold segmentation algorithm with a lightweight convolutional neural network algorithm, and to visualize the signal. The statistics unit is used to count the number of microbubbles within the target ROI region of each frame of the target image.

6. The auxiliary diagnostic device for cardiac blood flow abnormalities according to claim 1, characterized in that, The decision-making module includes: The microbubble number grading unit is used to determine the corresponding microbubble level based on the number of microbubbles in the target ROI of each frame of the target image, according to a preset grading rule. The diagnostic generation unit is used to generate corresponding diagnostic recommendations based on the microbubble level and the anomalous embolism risk score.

7. The auxiliary diagnostic device for cardiac blood flow abnormalities according to claim 4, characterized in that, The ROI extraction unit includes: The image preprocessing subunit is used to preprocess the image sequence of the patient under test during the period of increased right atrial pressure; The mask extraction subunit is used to extract the target ROI mask of each frame in the preprocessed image sequence. The mask post-processing subunit is used to optimize the target ROI mask for each frame of the image; The mask overlay subunit is used to overlay the optimized ROI mask of each frame image onto the preprocessed image sequence to obtain the target image sequence displaying the target ROI.

8. An auxiliary diagnostic method for cardiac blood flow abnormalities, characterized in that, Performed by the cardiac blood flow abnormality auxiliary diagnostic device according to any one of claims 1-7; the method includes: The system monitors in real time whether the patient under test performs the Valsalva maneuver. When the patient under test performs the Valsalva maneuver, a pulse signal is sent. Based on the pulse signal, an image sequence of the period when the right atrial pressure of the patient under test is elevated is acquired, and the target ROI is extracted from the image sequence to obtain a target image sequence displaying the target ROI; The microbubble signal within the target ROI of each frame of the target image sequence is detected in real time, and the number of microbubbles within the target ROI of each frame of the target image is counted. The target ROI includes the left atrium and the left ventricle. Based on the number of microbubbles within the target ROI of each frame of the target image and the anomalous embolism risk score, corresponding diagnostic suggestions are generated.

9. The auxiliary diagnostic method for cardiac blood flow abnormalities according to claim 8, characterized in that, The image of the patient under test during the period of increased right atrial pressure is a standard cross-section of the three-dimensional cardiac image of the patient under test. After acquiring the image sequence of the patient under test during the period of increased right atrial pressure based on the pulse signal, the method further includes: The image quality index of each frame in the image sequence during the period of increased right atrial pressure of the patient under test is monitored in real time. When the image quality index is less than a preset threshold, the image with the image quality index less than the preset threshold is switched to another standard cross-section of the three-dimensional heart image of the patient under test at the same moment.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the auxiliary diagnostic method for cardiac blood flow abnormalities as described in any one of claims 8-9.

Images