19 results about "Ventricular volume" patented technology

The invention provides an echocardiogram video abstraction method based on a hierarchical space-time attention mechanism, and the method comprises the steps: obtaining an echocardiogram, inputting the echocardiogram into an anatomy and physiological basic model of a key frame recognition task, and obtaining key frame information; inputting the key frame information into a multi-dimensional spatial-temporal feature extraction model to obtain feature information and frequency domain information; fusing the feature information and the frequency domain information; performing key frame enhancement processing on the fusion information by adopting an SE channel attention mechanism to obtain key frame anatomical feature expression; processing the key frame anatomical feature expression by adopting a key frame backtracking strategy and a classification network to obtain a key frame identification classification result; performing left ventricle segmentation on the classified image by adopting an improved U-Net left ventricle region segmentation network; estimating the area and volume of the left ventricle based on a segmentation result; fusing the estimated left ventricular area and volume by adopting a hierarchical attention mechanism and a cross-layer information fusion strategy to obtain a ventricular volume result; according to the method, through abstract extraction of the key information of the ultrasonic cardiac video, the clinical practicability of intelligent screening and auxiliary diagnosis of cardiovascular diseases based on ultrasonic is improved.

The invention relates to the technical field of intracranial pressure monitoring, in particular to a multi-modal fusion-based intracranial pressure dynamic prediction method for a patient with craniocerebral injury, which comprises the following steps of: acquiring an electroencephalogram signal, extracting a peak value to a termination section path, judging whether a waveform slope is consistent with an intracranial pressure direction or not, identifying an unsynchronized response segment, and obtaining a neural migration section mark set. According to the method, the non-cooperative state between the neural activity and the pressure change is distinguished by recognizing the periodic waveform which does not cause the intracranial pressure response in the electroencephalogram signal, and the time period with the inconsistent regulation rhythm is positioned by combining the continuous expression of the ventricular volume and the arterial pressure direction deviation; the method comprises the following steps of: extracting response sequence and direction characteristics of arterial pressure, cerebral blood flow and neural signals in a continuous section, enhancing a rhythm corresponding relation among multi-modal signals, adjusting a signal alignment structure according to offset direction difference and a rhythm starting point, forming a linkage fragment sequence under a unified time reference, and obtaining a multi-modal signal sequence; and linkage and deduction of asynchronous response information on a structural level are promoted.

Methods and systems are disclosed for creating and using a neural network model to estimate a cardiac parameter of a patient, and using the estimated parameter in providing blood pump support to improve patient cardiac performance and heart health. Particular adaptations include adjusting blood pump parameters and determining whether and how to increase or decrease support, or wean the patient from the blood pump altogether. The model is created based on neural network processing of data from a first patient set and includes measured hemodynamic and pump parameters compared to a cardiac parameter measured in situ, for example the left ventricular volume measured by millar (in animals) or inca (in human) catheter. After development of a model based on the first set of patients, the model is applied to a patient in a second set to estimate the cardiac parameter without use of an additional catheter or direct measurement.

The invention provides a system and a method for drawing a ventricular contour and calculating ejection fraction by medical color Doppler ultrasound equipment. The system comprises an image acquisition and caching module for acquiring a plurality of frames of B-mode ventricular images; a background noise and ventricle distinguishing module obtains background noise and ventricle structure images; the contour boundary determination module obtains ventricle contour boundary points; the contour coordinate category distinguishing module performs category distinguishing on the ventricular contour boundary points and determines the position of the valve according to a category distinguishing result; a valve boundary point determination module determines a valve boundary point; the calculation module is used for calculating ventricular volume and ejection fraction; the display module is used for displaying the target ventricular image, the ventricular volume and the ejection fraction; according to the method, the requirement for noise can be effectively met, the ventricular contour can be accurately determined, the accuracy and efficiency of judging the cardio-pulmonary function can be guaranteed, and meanwhile the real-time state can be known more visually and efficiently through visual image superposition and numerical value display.

The invention provides a ventricular elasticity value estimation method and device.The method comprises the steps that a first parameter and a second parameter are obtained, the first parameter comprises at least one of target ventricular pressure, target aortic pressure and target pulmonary arterial pressure of a target user, and the second parameter comprises at least one of the target ventricular pressure, the target aortic pressure and the target pulmonary arterial pressure; the second parameter comprises the pumping flow of the target ventricular assist device and the cardiac cycle of the target user; inputting the first parameter and the second parameter into a target decision model, and outputting a target ventricular volume; and estimating a target ventricular elasticity value according to the ratio of the target ventricular pressure to the target ventricular volume. The ventricular volume of the current user is estimated through the decision model, the estimated ventricular elasticity value is calculated according to the ratio of the ventricular pressure to the ventricular volume, and the ventricular elasticity value of the patient can be accurately predicted in real time without depending on ultrasonic equipment.

Systems, devices, user interfaces, and methods for managing patient conditions. A system for monitoring and / or managing a patient suffering from a brain condition may include a controller obtaining and / or storing data related to one or more monitored patient parameters contributing to the brain condition of the patient and displaying values of the one or more monitored patient parameters in trend lines over time on a user interface. The displayed trend lines may represent values of ventricular CSF volume, values of IVH volume, total ventricular volume in the brain, and / or values of other suitable parameters. The data related to the one or more monitored patient parameters may include images of the patient's brain and / or data derived from images of the patient's brain and / or suitable patient data.

The application discloses a noninvasive calculation method of coronary microcirculation resistance and resistance index, and the method comprises the following steps: firstly, based on coronary computer tomography angiography data, blood vessel segmentation is carried out by using a U-Net convolutional neural network, and then a personalized 3D coronary stenosis branch vessel model is reconstructed; secondly, the information of multiple blood vessels (physiological parameters such as age, blood pressure and left ventricular volume in the systolic phase) is integrated, and the length, cross-sectional area, volume and other geometric structure parameters of a specific part of the blood vessel are measured and calculated; thirdly, fluid dynamics simulation is utilized to obtain the invasive CMR and IMR; fourthly, correlation analysis is carried out between the characteristic parameters and the CMR and IMR; fifthly, the characteristic parameters with high correlation are selected to carry out machine learning prediction, so as to realize noninvasive calculation of the CMR and IMR. The noninvasive method for evaluating the CMR and IMR can complete measurement in a short time and has high precision, and is convenient and safe.

The invention relates to the field of biomedicine, in particular to a pressure-volume ring drawing method with real-time feedback, which comprises the following steps: S1, collecting ECG data, ultrasonic data and non-invasive blood pressure data, integrating and inputting into an integrated processor; s2, extracting time sequence data of the left ventricular area, measuring the volume of the left ventricular area, positioning the cardiac cycle and calculating a volume-pressure coordinate through an integrated processor; and S3, dynamically drawing a PV ring according to the processed data, and extracting corresponding indexes of the PV ring. The method has the advantages that the ECG data and the ultrasonic data are connected with the integrated processor at the same time for data transmission, delay possibly existing in the transmission process is avoided, automatic time alignment of the ECG data and the ultrasonic data is achieved, then the starting point and the ending point of the cardiac cycle of the aligned ultrasonic data are divided by automatically recognizing the R-wave starting point, and the accuracy of the ECG data and the ultrasonic data is improved. The accuracy of cardiac cycle division is ensured, and preconditions of dynamic PV ring drawing are met.

A method is disclosed for continuous or intermittent wireless monitoring of cardiac ventricular volume with the ability to combine with pressure measurements for optimal remote monitoring of cardiac patients. A set of wireless position sensors or reflectors is placed within a cardiac ventricle (i.e., the left ventricle), the method including means for sensing pressure within one or more cardiac ventricles, and an external device configured to interrogate the wireless sensors to determine their relative and / or absolute positions with respect to each other and / or an external device, calculate the resulting ventricular volume, and create a pressure-volume loop.

The invention relates to a cardiac assist system comprising: - an intraventricularly implantable cardiac pump, - a processing unit, - at least one lead intended to be placed on an external wall of the heart. The processing unit includes a cardiac pump management function, a pacing function, and / or a defibrillation function. The processing unit is configured to: - perform cardiographic impedance measurements between said at least one lead and a metallic part of the cardiac pump in order to determine the following hemodynamic parameters: - a cardiac electromechanical delay from the cardiographic impedance measurements, - a blood filling time in the heart from the cardiographic impedance measurements, - a blood ejection time from the heart from the cardiographic impedance measurements, and - a change in right ventricular and / or left ventricular volume over time.Figure for the abridged version: Fig. 1.

Methods and systems are disclosed for creating and using a neural network model to estimate a cardiac parameter of a patient, and using the estimated parameter in providing blood pump support to improve patient cardiac performance and heart health. Particular adaptations include adjusting blood pump parameters and determining whether and how to increase or decrease support, or wean the patient from the blood pump altogether. The model is created based on neural network processing of data from a first patient set and includes measured hemodynamic and pump parameters compared to a cardiac parameter measured in situ, for example the left ventricular volume measured by millar (in animals) or inca (in human) catheter. After development of a model based on the first set of patients, the model is applied to a patient in a second set to estimate the cardiac parameter without use of an additional catheter or direct measurement.

An apparatus includes a processor configured for communication with an ultrasound transducer. The processor is configured to control the ultrasound transducer to obtain a first plurality of ultrasound images of a first heart valve of a heart chamber of a patient. The first plurality of ultrasound images may be a plurality of Doppler ultrasound frames. The processor is further configured to determine a first plurality of volume measurements of the heart chamber based on the first plurality of ultrasound images, wherein each volume measurement of the first plurality of volume measurements corresponds to an ultrasound image of the first plurality of ultrasound images. The processor is also configured to generate a pressure-volume (PV) loop of the heart chamber based on the first plurality of volume measurements without using a PV loop catheter and output the PV loop.

The invention discloses an artificial intelligence platform for comprehensive radiological image analysis across CT, MRI, and X-ray modalities. The system automatically detects and quantifies cardiac, pulmonary, and oncological abnormalities through multiple AI modules including heart localization, cardiac MRI quantification, pneumonia classification, and tumor segmentation / grading. The platform provides automated tumor detection, 3D volumetric segmentation, and malignancy grading and treatment planning. The system integrates hybrid CNN-Transformer networks, attention-based 3D segmentation, and multi-modal feature fusion, offering precise quantitative metrics for tumor size, volume, shape, and malignancy probability. It also a cardiac MRI analysis module measures atrial and ventricular volumes and detects atrial fibrillation or mitral valve stenosis. It calculates the cardiothoracic ratio and detects heart displacement. DICOM / PACS integration and federated learning enable clinical deployment and cross-institutional adaptability while preserving patient data privacy and assists clinicians in accurate and rapid diagnosis while minimizing human error. This first-in-world platform facilitates early detection, staging, grading, and treatment planning with high reproducibility and explainable outputs.

The application discloses a heart ultrasound image optimization enhancement method based on image filtering and belongs to the technical field of heart ultrasound images. The heart ultrasound image optimization enhancement method based on image filtering comprises the following steps: S1, obtaining a heart ultrasound binary image; S2, obtaining a heart ultrasound image to be processed; S3, highlighting the edge features of a blood vessel wall and myocardial tissue; S4, obtaining an enhanced heart ultrasound image; and S5, obtaining an optimized and enhanced heart ultrasound image. The application solves the problem that the prior art is prone to affecting the overall consistency judgment of heart cavities, valves and other structures, can more intuitively display a heart disease area, reduces the misjudgment of medical staff caused by noise or low contrast, can effectively capture texture information at different levels, enhances the recognition ability of complex heart structures, and further more accurately quantifies myocardial strain rate, ventricular volume and other parameters, thereby providing a reliable basis for evaluating heart function.

A computer-implemented method for determining a fetus ventricular volume from DWI images is provided. The computer-implemented method involves acquiring DWI images, with a pre-set voxel height, of a fetus ventricle, the acquisition being made with a single b-value selected in a range between 200 and 1000 s / mm2, selecting a ROI on each of the DWI images around the fetus ventricle, automatically clusterizing pixels in the ROI, obtaining clusterized DWI images, and calculating the fetus ventricular volume based on the number of pixels in the ROI for each of the DWI images and the pre-set voxel height. An NMR assessment system and an NMR assessment method implementing the method for determining the fetus ventricular volume are also provided.

The present application relates to the biomedical field, specifically relates to a kind of pressure-volume loop drawing method with real-time feedback, comprising the following steps: S1: ECG data, ultrasonic data and non-invasive blood pressure data are collected and integrated into integrated processor input;S2: by integrated processor, left ventricular area time series data is extracted, left ventricular volume is measured, cardiac cycle is positioned and the processing of volume-pressure coordinate is calculated;S3: according to the data dynamically drawn PV loop after processing, and the corresponding index of PV loop is extracted.The beneficial effects of the present application are: by simultaneously connecting ECG data and ultrasonic data integrated processor for data transmission, avoid the delay that may exist in transmission process, realize the automatic phase alignment of ECG data and ultrasonic data, then divide the start and end point of the cardiac cycle of the aligned ultrasonic data by automatically identifying R wave starting point, ensure the accuracy of cardiac cycle division, meet the precondition of dynamic PV loop drawing.

The present invention discloses a cardiac magnetic resonance cardiac function assessment method based on multi-task learning, which belongs to the technical field of medical image processing. The present invention labels, enhances and preprocesses cardiac magnetic resonance image data; then uses the pre-processed data to train a multi-task learning model, introduces regional geometric consistency loss to supervise the multi-task training process, and improves the accuracy and generalization performance of the model; then inputs the cardiac magnetic resonance image to be evaluated into the trained multi-task learning model, obtains the left and right ventricular endocardial segmentation results in the cardiac anatomical structure, the left ventricular myocardial segmentation results, and the right ventricular insertion point and left ventricular center point in the cardiac key point position, further calculates the ventricular volume to obtain the left and right ventricular ejection fraction; calculates the thickness of each segment of the left ventricular myocardium; and then outputs the cardiac function assessment result. The present invention improves the accuracy and generalization performance of the model, and solves the problems of low accuracy and long analysis time of traditional methods.

The invention discloses a full-automatic cardiac ultrasonic ejection fraction analysis method, system and device and a medium, relates to the technical field of image processing and deep learning, and solves the technical problem that EDV and ESV cannot be accurately predicted due to the fact that a dynamic heart cannot be accurately analyzed. According to the technical scheme, the method is characterized in that noise in a cardiac ultrasound image is segmented and resisted through an image segmentation model, the actual ventricular volume is calculated, whether the blood supply function of a patient is impaired or not is judged, the boundary of ventricular wall movement is obtained in an image registration model movement tracking mode, and therefore the segmentation model is assisted in determining the correct boundary; the tiny movement of valve closure can be captured through movement tracking to help judge whether the tiny movement is a special frame in a movement period or not; and finally, introducing regression prediction of ejection fraction EF, and naturally constraining the mask predicted in the motion period from the ventricular volume ratio, so that the densely predicted mask is smoother and more accurately attached to the chamber area.

This invention discloses a method, electronic device, and storage medium for detecting imaging features of hydrocephalus based on neural networks. The method includes collecting and labeling brain imaging data of patients with normal pressure hydrocephalus, and performing preprocessing; using a neural network to segment the ventricular region in thick-slice brain imaging data to generate a binary image sequence; training a neural network model for interpolation using thin-slice brain imaging data, continuously adjusting parameters until optimal; inputting the binary segmentation result of the thick-slice image along with the original thick-slice image into the trained interpolation network, outputting an interpolated binary segmented image; counting the number of voxels in the ventricular portion of the interpolated binary segmented image, and calculating the actual volume of the ventricular cavity based on the image scanning interval information. This invention reduces the error in calculating ventricular volume by segmenting images when the scanning interval is too large, obtaining accurate ventricular volume data, thereby optimizing the computer-aided diagnosis process for hydrocephalus.