58 results about "Lung segmentation" patented technology

Disclosed is a systematic way of automatically segmenting lung regions. To increase the efficiency of a lung segmentation technique, a region-based technique, such as region growing, is performed by a computer on a middle slice of the CT volume. A contour-based technique is then used for a plurality of non-middle slices of the CT volume. This allows the implementation to be multithreaded and results in an improvement in the segmentation algorithm's efficiency.

A computer aided diagnostic system and automated method diagnose lung cancer through tracking of the growth rate of detected pulmonary nodules over time. The growth rate between first and second chest scans taken at first and second times is determined by segmenting out a nodule from its surrounding lung tissue and calculating the volume of the nodule only after the image data for lung tissue (which also includes image data for a nodule) has been segmented from the chest scans and the segmented lung tissue from the chest scans has been globally and locally aligned to compensate for positional variations in the chest scans as well as variations due to heartbeat and respiration during scanning. Segmentation may be performed using a segmentation technique that utilizes both intensity (color or grayscale) and spatial information, while registration may be performed using a registration technique that registers lung tissue represented in first and second data sets using both a global registration and a local registration to account for changes in a patient's orientation due in part to positional variances and variances due to heartbeat and/or respiration.

An automated or semi-automated system and methods are disclosed that provide a rapid and repeatable method for identifying lung, lobe, and fissure voxels in CT images, and allowing for quantitative lung assessment and fissure integrity analysis. An automated or semi-automated segmentation and editing system and methods are also disclosed for lung segmentation and identification of lung fissures.

The invention discloses an X-ray chest radiography lung segmentation method so as to carry out accurate segmentation on lungs in an X-ray chest radiography. The method includes the following steps: S101: through horizontal and vertical projection, obtaining two rectangular areas which respectively surround left and right lung images in the X-ray chest radiography; S102: initializing the lungs in the two rectangular areas so as to obtain the initial shapes of the lungs; S103: according to a weighted grey local texture model, searching for optimal matching points of characteristic points in the lung images; S104: through adjustment of attitude parameters and a shape parameter b, enabling current shapes I<Xc> of the lungs to approximate I<X>+dI<X> in the largest degree; and repeating S103 and S104 until the change quantities are smaller than preset thresholds when obtained lung shapes are compared with lung shapes obtained through a previous adjustment next to a current adjustment. The method and device are capable of obtaining better initialization lung shapes and do not cause an over-segmentation phenomenon in a follow-up adjustment process and are capable of carrying out accurate segmentation on lung areas in the X-ray chest radiography under the restraint of an active shape model.

A three-dimensional medical image obtainment unit that obtains a three-dimensional medical image of a chest, a bronchial structure extraction unit that extracts a bronchial structure from the three-dimensional medical image, a divided lung region obtainment unit that divides, based on the divergence of the bronchial structure, the bronchial structure into plural bronchial structures, and obtains plural divided lung regions based on the plural divided bronchial structures, a distance image generation unit that generates, based on the plural divided lung regions, a distance image based on a distance between each voxel in an entire region excluding at least one of the plural divided lung regions and each of the plural divided lung regions, and a border non-existing region extraction unit that extracts, based on the distance image generated by the distance image generation unit, a border non-existing region, which does not include any borders of the divided lung regions, are provided.

The invention discloses a method and an apparatus for lung segmentation in a medical image. The method comprises the following steps of obtaining an average lung template according to lung contours of M chest films in a training set, wherein M is an integer greater than or equal to 2; preprocessing a lung image to be segmented to obtain a preprocessed lung image; extracting a binary image of a lung boundary of the preprocessed lung image, and performing initial locating according to generalized Hough transform to obtain a corresponding Hough locating position; aligning the average lung template to the Hough locating position to obtain an alignment result; performing segmentation by utilizing a dynamic planning algorithm, and inversely transforming a segmentation result to be in an original coordinate system, thereby finishing lung region segmentation. According to the method and the apparatus, the lung region to be segmented is located through the Hough transform, and the locating position is aligned through an average lung region shape to obtain an initial contour of the lung, so that a deviation of the initial position and the actual position of the lung is reduced to the maximum extent and the accuracy of segmentation is improved.

The invention discloses a method for cardio-thoracic proportion calculation of a medical image. The method includes steps of acquiring a thorax image; performing segmentation on a lung area in the thorax image and acquiring a left lung image and a right lung image; acquiring the central axis of the left lung and the right lung; measuring a podoid right side maximal cross diameter, a podoid left side maximal cross diameter and a thorax maximal cross diameter; calculating the proportion between the sum of the podoid right side maximal cross diameter and the podoid left side maximal cross diameter and the thorax maximal cross diameter. According to the invention, on the basis of lung segmentation, the podoid maximal cross diameter and the thorax maximal cross diameter are measured automatically and the cardio-thoracic proportion is calculated.

A lung segmentation processor (40) is configured to classify magnetic resonance (MR) images based on noise characteristics. The MR segmenatation processor generates a lung region of interest (ROI) and detailed structure segmentation of the lung from the ROI. The MR segmentation processor performs an iterative normalization and region definition approach that captures the entire lung and the soft tissues within the lung accurately. Accuracy of the segmentation relies on artifact classification coming inherently from MR images. The MR segmentation processor (40) correlates segmented lung internal tissue pixels with the lung density to determine the attenuation coefficients based on the correlation. Lung densities are computed by using MR data obtained from imaging sequences that minimize echo and acquisition times. The densities differentiate healthy tissues and lesions, which an attenuation map processor (36) uses to create localized attenuation maps for the lung.

A system for detecting and diagnosing lung neoplasms on CT images is provided. The system consists of a lung parenchyma segmentation module, a candidate nodule detection module and a candidate nodulediagnosis module. In the lung segmentation module, the original CT image is morphologically denoised, then the binary segmentation is performed by the optimal threshold method, the initial boundary isextracted by the boundary tracking method, the boundary is repaired by the boundary matching repair algorithm, and finally the lung parenchyma is segmented by the flood filling algorithm. In the candidate nodule detection module, the candidate nodule detection algorithm based on ring filter and the candidate nodule detection algorithm based on threshold are combined. The detection results will include a large number of false positives. First, the rule method is used to eliminate the false positives, and then the fuzzy super-box neural network based on K-means clustering algorithm is used as asystem classifier for the diagnosis of candidate nodules. This system provides a good support for doctors to diagnose lung cancer.

The invention discloses a system and a method, and lists health degrees according to categories from the chest radiography of healthy people through a CAA (Computer-Aided Analysis) system based on cloud computing in non-diagnostic and diagnostic environments. Input images of the system comprise images which are used for diagnosis and are not used for diagnosis, and image data stored in a personal computer/ smartphone or cloud. The image data is transmitted to a server where the CAA is positioned through the Internet. The CAA based on the cloud computing firstly carries out lung segmentation on the image data to generate various image omics characteristics, then executes image omics characteristic classification and finally generates a classification result. The classification result is used for retrieving relevant clinical cases and health situations, and the relevant clinical cases and health situations and health knowledge stored in a database form a new data set. The data set is sent to a terminal user and is displayed on a query system or the personal computer/ smartphone.

The embodiment of the invention discloses a model generation method, a medical image segmentation method and device, equipment and a medium. The model generation method comprises the steps of obtaining a sample image of a sample chest, a lung mask image of a known lung in the sample image, and shape priori knowledge of the lung mask image; taking the sample image, the lung mask image and the shapepriori knowledge as a group of training samples, and training the original segmentation model based on multiple groups of training samples to generate a lung segmentation model, wherein the originalsegmentation model comprises a feature extraction network, and an image segmentation network and a priori knowledge regression network which are respectively connected with the feature extraction network. According to the technical scheme of the embodiment of the invention, the problem that the existing segmentation model does not effectively utilize the shape priori knowledge of the lung region is solved, and the priori knowledge regression network set based on the shape priori knowledge is matched with the existing image segmentation network, so that the matching degree of the lung segmentation model and lung region segmentation is improved.

The invention discloses a static artery separation method and device based on a CT image, and the method and device improve the efficiency and precision of static artery separation compared with a conventional method and manual labeling of a doctor, achieve the full-automatic static artery separation, and do not need the manual intervention. The method mainly comprises the following steps: carrying out lung region segmentation on a chest CT image by using a preset three-dimensional lung segmentation model to obtain a three-dimensional mask of a lung region; carrying out convex hull operation on a lung mask, taking a lung region according to the lung mask subjected to convex hull operation, setting the pixel value of the extrapulmonary region to be 0, and obtaining a maximum extrapulmonarybounding box according to the lung mask; and performing static artery separation on the segmented CT image of the lung in the lung external bounding box by using a preset downsampling-free three-dimensional cavity convolutional neural network to obtain a static artery mask. As the accurate annotation data is used for training, and the three-dimensional cavity convolutional neural network is used for learning, the loss of information amount is reduced, and the accurate separation of the static artery is realized.

The invention discloses a three-dimensional lung feature extraction method based on a CT image. The method comprises the steps of performing lung segmentation on a lung CT image to obtain lung parenchyma; performing tracheal segmentation on the pulmonary parenchyma to obtain a tracheal tree; performing bifurcation point detection on the tracheal tree to obtain a bifurcation feature point set; extracting contour points through Canny edge detection, and merging bifurcation feature points. Tracheal tree bifurcation points and lung contour points are more firmly recorded by application generated by respiration, and better data are provided for feature point set registration.

The invention provides a chest radiograph segmentation and processing method, a chest radiograph segmentation and processing system and electronic equipment, and relates to the technical field of image processing. the method comprises the following steps: obtaining an original image of a target chest radiograph and a labeled image thereof, obtaining an image data set, the labeled image comprisingan image labeled with an actual lung expanding edge and a chest contour edge of a lung; Utilizing a deep learning algorithm to train the image data set to obtain an image segmentation model based on deep learning; Obtaining a to-be-segmented chest radiograph of the target patient, and processing the to-be-segmented chest radiograph to obtain a to-be-segmented image; Processing the to-be-segmentedimage through an image segmentation model to obtain a segmented lung region and a segmented thoracic region; And calculating the pulmonary-thoracic ratio according to the pulmonary region and the thoracic region. According to the method, the actual lung opening edge of the lung of the chest radiograph is segmented based on the deep learning algorithm, a very good segmentation effect is achieved, the thorax edge of the chest radiograph is segmented for the first time, then the lung-thoracic ratio is calculated, the method can serve as an important indication for measuring whether lung ventilation is abnormal or not, and the application range of the lung segmentation result is expanded.

The invention discloses a multi-center child X-ray chest radiography image lung segmentation method based on a TransUNet model. The method comprises the following steps: (1) collecting a multi-center child X-ray chest radiography image and preprocessing the multi-center child X-ray chest radiography image; (2) dividing the data into a training set, a verification set and a test set; (3) a segmentation model is constructed, a Transform layer is added to the segmentation model on the basis of UNet, and the segmentation model comprises four parts of three-time down-sampling, a linear layer, the Transform layer and up-sampling; (4) sending the training set into the constructed segmentation model for training, evaluating the performance of the segmentation model by using the verification set, adjusting the hyper-parameters of the model according to the evaluation effect, and finally obtaining the segmentation model with the performance reaching the standard through repeated training and verification; and (5) inputting a to-be-segmented multi-center child X-ray chest radiography image into the trained segmentation model so as to intelligently segment a lung region. The method provided by the invention combines the advantages of the Transformers network and the UNet network, and has relatively high segmentation precision and efficiency.

The invention discloses a position classification method, device and equipment for pulmonary nodules, and a storage medium. The method comprises the steps of obtaining a to-be-identified image; carrying out pulmonary nodule image detection on the to-be-identified image to obtain a pulmonary nodule detection image, wherein the pulmonary nodule detection image comprises one or more pulmonary noduledetection sub-images; performing left and right lung segmentation processing and lung lobe segmentation processing on the to-be-identified image to obtain left and right lung segmentation images and alung lobe segmentation image; and based on the pulmonary nodule detection image, the left and right lung segmentation images and the lung lobe segmentation image, performing pulmonary nodule positionclassification on each pulmonary nodule detection sub-image. By means of the technical scheme, lung morphological characteristics can be combined, pulmonary nodule position classification is assistedthrough left and right lung segmentation and lung lobe segmentation, the classification speed can be increased, the obtained pulmonary nodule position result is more intuitive and clearer, and follow-up screening for pulmonary nodules of different position types is facilitated.

The invention relates to a hybrid lung segmentation system based on deep learning and image processing. The system comprises an obtaining module which is used for obtaining a DICOM file of a lung CT image, a preprocessing module which is used for preprocessing the DICOM file, a first segmentation module which is used for carrying out bronchial segmentation, blood vessel segmentation and pulmonarynodule detection on the preprocessed DICOM file, a second segmentation module which is used for taking the DICOM file, the bronchial segmentation result and the blood vessel segmentation result as input of a deep learning model to perform lung lobe segmentation and extraction and arteriovenous segmentation, and a third segmentation module which is used for performing lung segment segmentation by taking the bronchial segmentation result, the pulmonary lobe segmentation result and the arteriovenous segmentation result as input of a segmentation model. According to the invention, the segmentationof the whole lung is realized, and the anatomical structure reference of preoperative planning can be provided for a lung wedge-shaped operation or a puncture ablation operation.

The invention provides an image processing device and method and a storage medium, and the device comprises a first obtaining module which is used for obtaining a to-be-detected chest image; the firstsegmentation module is used for carrying out lung segmentation processing on the to-be-detected chest image to obtain a plurality of lung segmentation regions corresponding to the to-be-detected chest image; the first detection module is used for carrying out bronchiectasis detection processing on each lung segmentation region to obtain a bronchiectasis detection result corresponding to each lungsegmentation region; and the first determination module is used for determining a target bronchiectasis detection result corresponding to the chest image to be detected based on the bronchiectasis detection result corresponding to each lung segmentation area. According to the invention, automatic detection and grading evaluation of branch expansion can be realized.

The invention discloses a lung lobe segmentation optimization method and system based on lung segmentation, and relates to the field of medical images. The method comprises the steps that 1, a first output result of a first lung automatic segmentation algorithm model and a second output result of a second lung lobe automatic segmentation algorithm model are obtained, and the precision of the first lung automatic segmentation algorithm model is larger than that of the second lung lobe automatic segmentation algorithm model; the method also includes: step 2, carrying out point multiplication on the first output result and the second output result to obtain an independent mask of each lung lobe; step 3, respectively acquiring the maximum connected domain matrix of the independent mask of each lung lobe; step 4, performing superposition processing on each maximum connected domain matrix to obtain first optimization data; and step 5, performing calculation and elimination processing on the first optimization data to obtain a final optimization result. The invention can achieve the effect of improving the prediction precision.

The invention discloses an image report pushing method and device based on RPA and AI and a computing device, and the method comprises the steps: carrying out the segmentation of a to-be-detected lung medical image sent by an RPA robot through a preset lung segmentation model to obtain a target medical image only containing a lung region, detecting the target medical image by using a preset pulmonary nodule detection model to obtain attribute information of each suspected pulmonary nodule, and segmenting an area where each suspected pulmonary nodule is located by using a preset pulmonary nodule segmentation model to obtain three-dimensional contour information of each suspected pulmonary nodule; and generating a suspected lung nodule image report according to the attribute information and the three-dimensional contour information of each suspected lung nodule, and sending the suspected lung nodule image report to a hospital platform through the RPA robot. The lung nodules are detected through the AI image analysis technology to obtain the suspected lung nodule image report, and the suspected lung nodule image report is sent to the hospital platform through the RPA robot, so that the time for a doctor to distinguish the lung nodules is shortened, and the efficiency is improved.

The invention discloses a rapid nodule matching method based on nodule characteristics. The method comprises the steps that a first lung CT image and a second lung CT image which meet preset conditions are acquired; lung segmentation and lung frame scaling are performed on the first lung CT image and the second lung CT image, so that lung regions of the first lung CT image and the second lung CT image are aligned, and a transfer function is obtained; adjacent nodules of the two CT images are found out according to the transfer function, and a normalized cross correlation coefficient of the adjacent nodules is calculated; and whether the adjacent nodules are matched nodule pairs or not is judged according to the relationship between the normalized cross correlation coefficient and the threshold. According to the invention, a lung alignment method of lung segmentation and lung frame scaling is adopted, so that the time required for lung position alignment caused by an iterative algorithmis greatly reduced.