108 results about "Breast ultrasonography" patented technology

Methods and related systems are described for detection of breast cancer in 3D ultrasound imaging data. Volumetric ultrasound images are obtained by an automated breast ultrasound scanning (ABUS) device. In ABUS images breast cancers appear as dark lesions. When viewed in transversal and sagittal planes, lesions and normal tissue appear similar as in traditional 2D ultrasound. However, architectural distortion and spiculation are frequently seen in the coronal views, and these are strong indicators of the presence of cancer. The described computerized detection (CAD) system combines a dark lesion detector operating in 3D with a detector for spiculation and architectural distortion operating on 2D coronal slices. In this way a sensitive detection method is obtained. Techniques are also described for correlating regions of interest in ultrasound images from different scans such in different scans of the same breast, scans of a patient's right versus left breast, and scans taken at different times. Techniques are also described for correlating regions of interest in ultrasound images and mammography images. Interactive user interfaces are also described for displaying CAD results and for displaying corresponding locations on different images.

An ultrasonic breast examination system suitable for use by unskilled users, comprising a patient positioning platform with a patient positioning image sensor, and a moveable ultrasonic transducer guidance device. The transducer guidance device can be a fluid filled container with a flexible side or bottom, and a mechanism to automatically move an ultrasonic transducer over the breast. The system will be controlled by at least one microprocessor, associated software, and an optional touch sensitive display screen. The system may use an image sensor to properly position the patient so that the transducer guidance device may be properly positioned proximate to the patient's breast. The touch sensitive display screen is designed to allow the system to be directly operated by an unskilled patient. The system will often be connected to a network, such as the Internet, so that remote operators may interpret breast ultrasound images and optionally control the system.

The present invention refers to a graphical user interface (501) and a corresponding method for rapid and consistent input, modification and display of annotation text to be linked with and displayed in at least one image visualized on a monitor screen or display (102) of a medical or other kind of imaging system (100) without needing to type this text information (e.g. by activating or deactivating softkeys on a touch screen (502) or by rotary knob selection). Additionally, said user interface (501) allows to automatically link graphical annotation information, such as e.g. body markers and graphical transducer orientation information, to the annotated text.

The invention relates to a computer-aided reference system and a computer-aided reference method for fusing multimodal mammograms. The method comprises steps: image preprocessing is performed on breast ultrasounography and molybdenum target images of a same patient; the region of interest of ultrasound image and the region of interest of molybdenum target image of breast are obtained by using thepreprocessed ultrasound and molybdenum target images as inputs; the texture and depth features of ROI in breast ultrasound images and mammography are extracted; the extracted GLCM, LBP, HIS texture features and CNN depth features are fused in series to construct texture; a deep fusion feature model is constructed; the texture-Deep fusion feature model is used as input, and a regularization limit learning machine is used to classify the breast masses into benign and malignant ones. The invention uses the regularization limit learning machine to classify the benign and malignant breast masses, realizes the judgment of the nature of the breast masses from the clinical requirements, and provides an objective reference basis for the doctor to diagnose the breast masses.

Various methods and systems are provided for ultrasonically scanning a tissue sample using a hand-held automated ultrasound system. In one example, a system for ultrasonically scanning a tissue sample includes a hand-held ultrasound probe including a housing and a transducer module comprising a transducer array of transducer elements, one or more position sensors coupled within the housing, and a controller. The controller is configured to generate one or more images based on ultrasound data acquired by the transducer module and further based on position sensor data collected by the one or more position sensors.

An ultrasound imaging breast tumor detection and diagnostic system and method is disclosed. The method uses the system to acquire a plurality of 3D breast ultrasound images, and then to cut out multiple regions from the 3D breast ultrasound images using a 3D means shift algorithm, and then to acquire the mean grayscale value (MGV) of each region, and then to classify the regions to groups subject to the mean grayscale value (MGV), and to merge each of the regions of the darkest group with adjacent regions of the similar grayscale into a respective suspicious tumor tissue full region, and then to recognize each suspicious tumor tissue full region to be a tumor tissue region or non-tumor tissue region. Thus, using region as the basic computing unit, tumor tissues are quickly recognized from the 3D breast ultrasound images.

Provided herein is a breast ultrasound scanning and diagnosis aid system. The scanning aid system is capable of aiding positioning and tracking the scanned breast ultrasound images and suspicious tumor ultrasound images so that the doctor can be easily informed of the scanned positions of the breast ultrasound images and the suspicious tumor ultrasound images. The diagnosis aid system selects at least one representative, breast ultrasound images with tumor characteristics, segments a suspicious tumor region, and acquires a plurality of characteristic parameters in association with tumor tissues from the suspicious tumor region so as to provide a diagnostic suggestion. With the use of the present invention, the positions of suspicious tumors can be acquired from the positioning data of the breast ultrasound images, and tumor diagnosis accuracy can be improved by selecting an ultrasound image with tumor characteristics.

The invention discloses a disease grading method and device based on machine learning, equipment and a medium, the method belongs to the field of medical artificial intelligence, and the method comprises the following steps: acquiring a breast ultrasound image; calling a multi-task network to process the mammary gland ultrasonic image to obtain a first feature map and a semantic segmentation map of the mammary gland ultrasonic image, the first feature map comprising high-level semantic features of the mammary gland ultrasonic image, and the semantic segmentation map being a map for semantic segmentation of a focus area in the mammary gland ultrasonic image; performing feature extraction of different weights on the lesion area and the non-lesion area in the first feature map according to the semantic segmentation map to obtain a second feature map; and predicting according to the second feature map to obtain the BI-RADS grading of the breast ultrasound image. High-level semantic features with classification guidance are adopted as main grading features, different feature extraction is carried out on a lesion area and a non-lesion area by introducing a lesion contour, and an accurateBI-RADS grading result is obtained through regression.

Disclosed is an automatic breast ultrasound scanning method. The method is characterized in that a used automatic breast ultrasound scanning device comprises an ultrasonic coupling device, a probe seat, an ultrasonic probe and a probe moving device; the ultrasonic coupling device comprises a supporting frame, an upper elastic membrane and a lower elastic membrane, the edges of the upper elastic membrane and the lower elastic membrane are both connected with the supporting frame, and an airtight cavity is formed between the upper elastic membrane and the lower elastic membrane and is filled with coupling liquid; when scanning is performed, an examined person lies on an examination couch on the back, then the automatic breast ultrasound scanning device is put on the breast of the examined person, and the lower elastic membrane of the ultrasonic coupling device is tightly pressed on the breast; then the probe moving device drives the probe seat and the ultrasonic probe to move together according to a predetermined route, and the ultrasonic probe clings to the upper surface of the upper elastic membrane and scans the breast. The breast can be scanned automatically; the examined person adopts a supine posture in the process of examination, thereby feeling comfortable; operation is stable, scanning speed is high, and examination accuracy is high.

The invention provides an ultrasonic imaging system, a BI-RADS grading method and a model training method. The ultrasonic imaging system comprises an ultrasonic probe, a transmitting/receiving circuit, a processor performing feature map extraction on the breast ultrasound image based on one or more first feature map extractors to obtain one or more first feature maps about breast lesion features,the breast lesion features including BI-RADS features, performing feature map extraction on the breast ultrasound image based on a second feature map extractor to obtain a second feature map about BI-RADS grading, and classifying the first feature map and the second feature map based on a first classification model to obtain a BI-RADS classification result, and an output equipment used for outputting the BI-RADS grading result. The ultrasonic imaging system can eliminate the difference between the features, and can improve the accuracy of BI-RADS classification.

An ultrasonic breast inspection instrument comprises an inspection instrument main body, a first display, a supporting arm and an ultrasonic scanning device. The first display is mounted on the inspection instrument main body. One end of the supporting arm is mounted on the inspection instrument main body, and the ultrasonic scanning device is mounted at the other end of the supporting arm and comprises a shell, an ultrasonic probe and a probe moving device capable of moving the ultrasonic probe. The bottom of the shell is provided with a hard base plate. The ultrasonic probe and the probe moving device are both mounted in the shell. The lower end face of the ultrasonic probe makes contact with and is coupled with the upper surface of the hard bottom plate. Since the ultrasonic scanning device adopts the hard bottom plate, when the ultrasonic scanning device is pressed to the human body to-be-scanned portion, the hard bottom plate can fix a target body so that the target body cannot move and deform in the scanning process of the ultrasonic probe, and an ultrasonic scanning result is more accurate accordingly; and in addition, the ultrasonic probe does not make direct contact with the human body, and discomfort caused to the human body in the scanning process is avoided.

The invention discloses a breast ultrasound tumor recognition method based on deep learning. The breast ultrasound tumor recognition method comprises the following steps: S1, carrying out benign and malignant labeling on a breast ultrasound image of an existing case; S2, preprocessing the labeled breast ultrasound image; S3, acquiring the characteristics of the preprocessed image by adopting a convolutional neural network model; S4, taking the obtained features and the corresponding labels as training data to train different classification models respectively; S5, fusing all the trained classification models by adopting a stacking method; and S6, taking the breast ultrasound tumor to be identified as the input of the fused model, and completing the identification according to the output result. According to the method, the image recognition result can be directly obtained only by putting the breast ultrasound image to be recognized, the recognition time is short, diagnosis can be carried out through the connection server, or the breast ultrasound image can be directly deployed in a local computer, flexibility is high, an interface is simple, operation is easy, and user friendliness is achieved.

The invention belongs to the technical field of image processing, and specifically relates to a lightweight incisional hernia patch three-dimensional ultrasonic image feature extraction method. The method comprises the steps of firstly extracting related textural feature parameters of regions to be classified in three-dimensional volume of interest (VOI) of an automated three-dimensional breast ultrasound (ABUS) image in automatic quantification manner by using textural feature extraction algorithm so as to be used for differentiating a patch and a fascia; then introducing three-dimensional textural parameters and three-dimensional location parameters to improve the robustness of a lightweight patch classification and recognition algorithm in allusion to a problem that two-dimensional textural parameters are sensitive to spatial transformation such as post-operation curling and contraction of an incisional hernia patch; and finally performing feature selection by using a distance-between-class algorithm and a sequential forward selection method. The method provided by the invention is good in feature selection effect, high in efficiency, capable of effectively improving the classification accuracy of lightweight incisional hernia patch three-dimensional ultrasonic images, and convenient for automatic classification and recognition.

The invention discloses a breast ultrasound examination robot. The breast ultrasound examination robot comprises the components: an ultrasonic probe position adjustment module and an ultrasonic probeposture adjustment module, wherein the ultrasonic probe position adjustment module comprises a circular arc sliding rail, a sliding block, a feeding screw, a feeding sliding table, lifting screws, lifting sliding tables, screw motors and a lifting rod; and the ultrasonic probe posture adjustment module comprises a spherical joint sleeve, a spherical joint, an inner swing ring, an outer swing ring,drive motors, a swing rod and a tail end clamping plate. According to the breast ultrasound examination robot, automation of breast ultrasound scanning can be achieved, and an ultrasonic probe can bekept hovering at an arbitrary position within the working range and does not need to be held by a doctor all the way, so that the scanning efficiency is improved and the labor intensity of the doctoris reduced.

Versatile ultrasound scanning of a breast is described using an apparatus (1202) including a hand-manipulable compression/scanning assembly (1208). The compression/scanning assembly (1208) comprises an ultrasound transducer (1304) and a compressive member comprising an at least partially conformable membrane (1218) in a substantially taut state, the membrane (1218) having a first surface compressing the breast in a generally chestward direction and a second surface opposite the first surface. The compression/scanning assembly (1208) further comprises a transducer translation mechanism coupled to the ultrasound transducer (1304) and configured to sweep the ultrasound transducer across the second surface of the membrane to scan the breast while compressed in the generally chestward direction. Systemized and/or standardized ultrasonic scanning of a breast based on hand-manipulable scanners (1208, 1508) having substantially planar scanning surfaces is also described.

An apparatus and a method are disclosed for obtaining ultrasound images of a patient's breast that is chestwardly compressed with a template that is essentially planar and rotates relative to the breast while one or more ultrasound transducers moving with the template take 2D images of the breast through one or more respective radially oriented slots in the template, preferably through a membrane that is porous to a gel. The 2D images are processed into slice images representing breast slices of desired thicknesses and orientation that are displayed alone or with some of the 2D images, preferably pairs of orthogonally disposed 2D images.

The invention discloses an automatic marking method for lesion area morphology in a breast ultrasound contrast video, and the method comprises the steps: designing an end-to-end network model structure, just transmitting to-be-recognized data to a model, enabling the model to automatically carry out the convolution operation of each frame of image, and extracting a discrimination feature of a classification basis. The focus area range does not need to be manually drawn in the whole identification process; certain lesion morphological characteristics describe contrast change under related normal tissues and contrast change of lesion tissues, such as enhanced intensity and enhanced time sequence, a convolutional neural network is used for automatically carrying out convolution calculation ona whole radiography video frame sequence, mapping data of normal tissues and lesion areas are shown through calculated characteristic values, and comparison is carried out according to network rulesto obtain a result. In addition, for morphological characteristics such as crab foot shape and enhancement sequence, the designed network is used to automatically calculate the characteristics corresponding to the morphological dynamic change for the spatial-temporal characteristics of the continuous frames of the video.

The invention discloses a breast ultrasound image self-learning extraction method and system based on a stacked noise reduction self-encoder. The method comprises the steps of extracting manual shallow layer features from each ultrasound breast lesion area image ROI as a training sample to form a training sample set set_unlabeled = {x(1), x(2), ..., x(n)}, the i-th sample x(i) belonging to [0, 1]<d>, i = 1, 2, ..., n; based on the training sample set, training a first noise reduction self-encoder DAE1; after training the first noise reduction self-encoder, re-entering the training sample set, using the self-encoder trained in the step S4 to extract feature expressions obtained through hidden layer learning of all the samples to form a new sample {y(1), y(2), ..., y(n)}, and using the new sample as an input of a second noise reduction self-encoder DAE2 to train the second noise reduction self-encoder. The invention achieves extraction of breast ultrasound image features, thereby provides valuable reference opinions for clinic diagnosis, and improves the accuracy and efficiency of breast cancer diagnosis.

The invention discloses an intelligent breast focus analysis method and system based on a breast ultrasonic image. The method mainly comprises three parts of dynamic identification, auxiliary analysisand report/case generation. The three parts can be used independently to output corresponding results in stages; and the parts can also be combined for use in the whole process of breast ultrasonic examination. According to the method, identification and analysis is completed by using a tailored and optimized deep learning algorithm; an analysis result is high in reliability and high in timeliness, the analysis result obtained through the method is mainly used for assisting doctors in efficiently processing daily breast ultrasonic examination work, auxiliary analysis and report/case generation are completed according to user requests, and compared with a traditional breast focus analysis method, the method is more user-friendly, and the misdiagnosis rate and the missed diagnosis rate aregreatly reduced.

Displaying breast ultrasound information on an interactive user interface is described, the user interface being useful in adjunctive ultrasound mammography environments and/or ultrasound-only mammography environments. Bilateral comparison is facilitated by a pairwise display of thick-slice images corresponding to analogous slab-like subvolumes in the left and right breasts. Coronal thick-slice imaging and convenient navigation on and among coronal thick-slice images is described. In one preferred embodiment, a nipple marker is displayed the coronal thick-slice image representing a projection of a nipple location thereupon. A convenient breast icon is also displayed including a cursor position indicator variably disposed thereon in a manner that reflects a relative position between the cursor and the nipple marker. Preferably, the breast icon is configured to at least roughly resemble a clock face, the center of the clock face representing the nipple marker location. Bookmark-centric and CAD-marker-centric navigation within and among thick-slice images is also described.

The present application relates to the field of ultrasound scanning, and discloses a three-dimensional breast ultrasound scanning method and an ultrasound scanning system, being able to accurately andautomatically determine the position of the nipple with respect to the lateral ultrasound image. In the method, putting a probe at a central position of the breast for ultrasonic scanning to obtain afirst ultrasonic image while a camera captures a first video image including a marker ring, wherein the first ultrasonic image covers a nipple position; determining a nipple position according to thefirst ultrasound image; placing a probe at a side position of the breast for ultrasonic scanning to obtain a second ultrasonic image, and simultaneously a camera captures a second video image including a marker ring; and performing image analysis on the first and second video images to determine a relative position of the second ultrasound image with respect to the nipple position based on the position of the marker ring in the first and second video images.

The invention discloses a breast ultrasound image tumor segmentation method. The method comprises the steps of breast ultrasound image data preprocessing, deep neural network model construction, loss function definition, model training and result generation. In data preprocessing, a mode of firstly filling a mirror image and then cutting is used, so that the form of a breast tumor is not changed, and a breast ultrasonic image meeting the size requirement can be obtained. In the step of constructing the deep neural network model, the design mode of the UNet model is followed on the whole. According to the method, ResNet18 is used as an encoder of the whole network, so that the method has higher feature extraction capability, and higher precision can be obtained; meanwhile, deep supervision technology is used in a model decoder part to supervise learning of each layer, and a channel-by-channel weighting module of SENet is added; thus, the problems of wrong segmentation and discontinuous segmentation boundaries can be eliminated, and the tumor boundaries can be accurately captured.

The invention discloses an ultrasonic breast tumor automatic segmentation method based on an attention-enhanced U-shaped network. The method comprises the following steps: (A) constructing an attention-enhanced U-shaped network model for ultrasonic breast tumor automatic segmentation; step (B), establishing a mixed attention loss function of the attention enhancement U-shaped network model; and step (C), according to the attention-enhanced U-shaped network model and the mixed attention loss function, training the attention-enhanced U-shaped network model through a coarse and fine combination strategy to achieve segmentation of the breast ultrasound image lesion area. The method can be used for extracting the focus area of the mammary gland ultrasonic image, can effectively improve the accuracy of mammary gland tumor segmentation, is used for assisting a doctor in quickly and accurately positioning the focus area, reducing the workload of the doctor and relieving the defects of insufficient clinical experience and the like of young doctors, and has very important research value and application prospect for modern medicine.

The invention relates to a domain ontology-based breast ultrasonic examination report structuring method, which comprises the following steps of: preprocessing a breast ultrasonic report to obtain a text description block; obtaining a branch subtree path for the obtained text description block based on a domain ontology semantic tree; generating a mammary gland ultrasonic semantic subtree in a top-down breadth-first mode; and converting the generated breast ultrasonic semantic subtree into structured data stored in a table structure. According to the method, the word segmentation accuracy is improved, and services can be provided for larger-scale medical data research.

Various methods and systems are provided for ultrasonically scanning a tissue sample using a modular transducer system. In one example, a system for ultrasonically scanning a tissue sample comprises: an adjustable arm; a scanning assembly attached to the adjustable arm, the scanning assembly including a housing configured to remain stationary during scanning and a module receiver that is configured to translate with respect to the housing during scanning; and a transducer module comprising a transducer array of transducer elements, wherein the transducer module is configured to be removably coupled with the module receiver in order to establish both a mechanical connection and an electrical connection between the module receiver and the transducer module.