56 results about "Imaging report" patented technology

The invention belongs to the technical field of medical image intelligent diagnosis, and provides a breast cancer focus benign and malignant discrimination method based on a gated multi-expert mechanism. The method comprises the following steps: firstly, carrying out standardization and semantic preprocessing on a mammary gland X-ray image, a BI-RADS imaging report and structured clinical data, embedding age, mammary gland density and focus position information into a text template in a natural language form, and realizing unified expression of multi-modal input; secondly, extracting image features by utilizing a ResNet network and a simplified CLIP model, obtaining a text semantic vector by adopting a Bio-ClinicalBERT model, and establishing two sub-paths of a lump expert and a calcification expert in a Transform structure; further, an expert weight is dynamically generated through a gating routing mechanism, and soft routing fusion is executed; and finally, outputting benign and malignant results of the breast cancer focus by the binary classification module. According to the method, deep fusion and dynamic collaboration of the mammary gland X-ray image, the BI-RADS text and the clinical information are realized, and the accuracy and interpretability of breast cancer discrimination can be remarkably improved.

The invention provides a radiographic image report generation method and system of a CT image, and the method comprises the steps: firstly transmitting an input CT image into a region division module, carrying out the whole-organ segmentation of the image, then distributing a whole-organ segmentation result according to a preset anatomical region with clinical significance, obtaining an anatomical region image with clinical significance, and carrying out the whole-organ segmentation according to a preset anatomical region with clinical significance. An anatomical area image with clinical significance is sent to a report generation module, an area image report text is generated, the report text and an original image are sent to a text-driven focus detection module, the position of a focus in the image is obtained and reported, and a doctor modifies the area image report and then sends the area image report to a report writing module to generate a final report. And meanwhile, the report modified by the doctor and the report automatically generated by the model are sent to an AI report generation quality evaluation module, so that the model carries out immediate update learning.

The invention is suitable for the technical field of medical image processing, and provides a fundus fluorescence radiography report generation method based on a visual large model, and the method comprises the steps: pre-training a fundus fluorescence radiography visual large model, extracting the image features of a fundus fluorescence radiography sequence image through the fundus fluorescence radiography visual large model, and constructing a pseudo time sequence curve; based on the pseudo time sequence curve, key point detection is carried out, a plurality of phase periods are divided, and a phase weight is set for each phase period; calculating a comprehensive score for each frame of the pseudo time sequence curve based on the phase weight in combination with time novelty; based on the comprehensive score, combining a preset target key frame number to obtain a key frame sequence; based on the key frame sequence, a multi-view sequence image report generation model is combined, a Chinese diagnosis report is generated, the time sequence information and phase characteristics of the fundus fluorescence contrast sequence can be fully mined, the diagnosis report conforming to clinical specifications is generated, and the limitation of the prior art is effectively overcome.

The invention discloses an image report process AI quality control system based on data fusion and a dynamic rule engine, and the system comprises a data layer module which is used for integrating multi-source data from an image archiving and communication system, a radiation information system and a hospital information system, and carrying out the preprocessing of the data, and forming a high-quality training data pool; furthermore, the image archiving and communication system obtains historical images in a DICOM format and corresponding report data; the radiation information system collects examination application information and doctor diagnosis information. According to the image report process AI quality control system based on data fusion and the dynamic rule engine, full-dimension and high-efficiency accurate quality control is realized: the system can perform comprehensive automatic screening of grammar, text, logic consistency and image-text relevance on an image report by integrating natural language processing, computer vision and a cross-modal matching technology; error correction response time is shortened to a second level from several minutes of each case of traditional manual auditing, and quality control efficiency is remarkably improved.

The application provides a thymus disease risk prediction method and system based on cross-modal feature interaction, which comprises the following steps: respectively preprocessing CT image data, MRI image data and image reports; sequentially performing word segmentation, vector conversion, feature extraction and pooling on the third data, inputting the labeled semantic feature vector into an image generation network to generate a PET-CT image; inputting the first data, the second data and the PET-CT image into the same medical mamba feature extraction network respectively to obtain the first modal feature corresponding to the first data, the second modal feature corresponding to the second data and the third modal feature corresponding to the PET-CT image; performing feature fusion interaction on the first modal feature, the second modal feature and the third modal feature to obtain a fusion feature, and obtaining the thymus disease risk according to the fusion feature. The application can greatly improve the thymus disease prediction accuracy.

The embodiment of the invention belongs to the technical field of medical image processing, and relates to a cervical cancer image report automatic generation method and related equipment, and the method comprises the steps: receiving a report generation request which is sent by a user terminal and carries a multi-sequence medical image; inputting the multi-sequence medical images into a multi-sequence three-dimensional MRI feature coding module for feature extraction operation to obtain medical image feature data; inputting the medical image feature data into a Mamb-Transform hybrid multi-mode decoding module to carry out an image report generation operation so as to obtain a medical image report; and outputting the medical image report to the user terminal. According to the method, the limitation of a traditional 2D image or a single-sequence 3D image is broken through, the unstructured, explainable and clinical-value image report is automatically generated on the basis of the three-dimensional multi-sequence MR I for the first time, and the method has wide clinical application value and important social benefits.

The invention discloses a medical image report generation method, system and device and a storage medium, and relates to the technical field of computer vision and natural language processing, and the method comprises the steps: constructing a causal graph model, taking an image as an input variable, taking a final report as an output variable, and taking a region-level pathological state in the image as an intermediary variable; the data imbalance factor is an unobservable hybrid factor; based on a causal graph model, intervening the intermediary variable by using a front door adjustment strategy, and establishing a causal path from visual evidence to report text; based on the intervened intermediary variable, executing a report generation process: a, identifying an abnormal region in the image by using a focus detection model and outputting a corresponding pathological discovery description; and b, inputting the intervened pathological discovery description and the original image into a visual language model to generate a complete clinical report by taking the intervened pathological discovery description and the original image as conditions. And the sensitivity of the model to pathological changes and the clinical reliability of report generation are improved.

The invention discloses a medical image report generation method and device and a storage medium. The method comprises the following steps: acquiring a target image examination part corresponding to a target examination object and a first image description text; and if it is detected that the first image description text has a writing error, correcting the first image description text based on a reference correction rule matched with the image description text in a correction rule library, and generating a second image description text. And identifying a plurality of medical entity words contained in the second image description text, and classifying the plurality of medical entity words to obtain classification tags corresponding to the plurality of medical entity words. According to the method, multiple pieces of structured data used for representing lesion features are generated based on the classification labels corresponding to the multiple medical entity words, and the multiple pieces of structured data are filled into the medical image report template corresponding to the target image examination part, so that the medical terms in the image description text can be more accurately filled into the template; and a high-quality image report is obtained.

The application discloses an artificial intelligence-based image analysis method and system for degenerative cervical spinal cord disease, relates to the technical field of artificial intelligence, and solves the technical problems of the existing technology, such as non-uniform image report standards, rough image analysis, increased patient anxiety, wasted medical resources, significant observer bias in DCM evaluation, and low efficiency of degenerative cervical spinal cord disease image analysis. A cervical vertebra dissection segmentation model is used to obtain a segmentation label; an intervertebral canal stenosis quantification grading model is used to obtain an intervertebral canal stenosis prediction grade of the segmentation label for the intervertebral canal; a spinal cord compression classification model is used to obtain a spinal cord compression result of the segmentation label for the cervical spinal cord; and a cervical vertebra image analysis model is used to obtain an analysis report of the cervical vertebra nuclear magnetic image. The image report is standardized, improved segmentation models are used, classification models are used in cooperation, and cervical vertebra image analysis models are used for processing, so that the accuracy of analysis is improved, and the efficiency of degenerative cervical spinal cord disease image analysis is improved.

The invention discloses a medical image report generation method based on a multi-modal fusion large language model. The method comprises the following steps: fusing local visual features of a medical image with a structured knowledge label through a multi-channel attention mechanism to generate a customized prompt vector; meanwhile, a context dependence matrix is introduced, a cross-graph convolution attention network is constructed, and evolution path modeling is carried out on the multi-time-point image; the prompts and contextual information are input into a large language model to automatically generate a continuous, professional image report. During training, a semantic consistency regular item loss function is combined, so that the report is kept consistent with the image in the semantic level. The system supports seamless connection and private deployment with the multi-mode PACS, and meets different medical environment requirements. According to the method, the quality and the continuity of automatic reports can be remarkably improved, and the clinical requirements on high-quality reports are met.

The invention relates to the technical field of medical image processing and artificial intelligence auxiliary diagnosis, in particular to a medical image report generation informatization planning system, which comprises a multi-modal time sequence difference feature analysis module used for acquiring current medical image data and historical image data of a patient and generating a time sequence difference residual vector; the semantic entropy flow conservation and negentropy injection control module is used for calculating a system net entropy target; the dynamic planning generation module based on residual driving is used for generating a diagnosis report text sequence; the diagnostic specificity and normativity collaborative optimization module is used for constructing a composite loss function containing a cross entropy loss term and a specificity penalty term and carrying out iterative updating on system parameters based on the composite loss function, and the specificity penalty term is used for constraining the description accuracy of the generated text on the dynamic change of the image; the problem that the dynamic change of the disease course cannot be captured only by relying on static image analysis in the prior art is effectively solved.

The invention provides a DICOM file processing method and device, equipment, a medium and a product. The method comprises the steps of obtaining a to-be-processed DICOM file, and performing anomaly detection and preprocessing on the to-be-processed DICOM file; according to the preprocessed non-abnormal DICOM file, analyzing the non-abnormal DICOM file to obtain analyzed multi-modal image data and metadata, and carrying out persistence processing on the analyzed multi-modal image data and metadata; identifying an image type of the multi-modal image data, and calling a data display template corresponding to the image type; and screening target metadata associated with the image type from the metadata after persistence processing, and filling the target metadata and the analyzed multi-modal image data into a data display template to form an image display report. The metadata of the DICOM file can be quickly obtained, and the overall efficiency of medical image data processing, image reading convenience and the standardization level of image report generation are improved.

The application discloses a kind of based on multi-feature enhancement medical image report generation method and system, method includes: using superpixel segmentation technology is segmented to medical image, obtain region of interest;Using pre-trained DenseNet is extracted to local feature to region of interest, and local feature is as graph node to construct relation graph;The relation graph is input into graph network GAT to learn graph feature, and the graph feature is input into encoder;The feature obtained by each layer of the encoder is output to the corresponding decoder layer in the form of U-type connection, the decoder predicts the next word based on the image feature obtained in the encoder and the generated text sequence feature, until all words are generated, finally obtain complete image report.Through the technical scheme of the application, the recognition of local lesions is enhanced, the interaction between text features and image features is strengthened, the semantic details and feature relationships of medical images are strengthened, and the accuracy of the generated report is improved.

The application discloses a disease evolution perception-based medical image report generation method and device based on analogy reasoning, and relates to the technical field of medical image analysis. The method comprises the following steps: based on an adaptive alignment mechanism, aligning the features of two images in a key lesion area; based on a visual evolution consistency loss function, constraining the learning of the evolution relationship between images; according to a current medical image report and a historical medical image report, extracting features by using a text encoder; based on a text evolution consistency loss function, constraining the learning of the evolution relationship between reports; and generating a medical image report based on an optimized visual encoder, an optimized adaptive region alignment module, an optimized evolution relationship learner, a text encoder, an optimized relationship mapper and an optimized large language model. The application is an accurate and efficient medical image report generation method based on cross-time disease evolution modeling.

The invention provides a focus identification method based on a standard template and an image report generation method. The method specifically comprises the following steps: receiving a medical image of a patient; determining a part to which the medical image of the patient belongs, and selecting a standard template corresponding to the part, the standard template comprising an anatomical marker structure; extracting a key internal structure contour in the medical image of the patient, and performing feature matching on the key internal structure contour and a corresponding anatomical marker structure in the standard template; under the condition of feature matching registration, voxel-by-voxel difference calculation is carried out on the medical image of the patient and a standard template in various transform domains, and a multi-channel residual image is generated; generating an exception enhancement graph according to the multi-channel residual image to screen an exception region; and automatically generating a focus description statement based on the abnormal region. According to the invention, the workload of initial writing of an image report by a radiologist can be obviously reduced, and the clinical working efficiency and the image interpretation consistency can be improved.

The invention provides an image report generation method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining the detection information of a user, and the detection information comprises a medical image and clinical information; obtaining abnormal sign information of each region based on the medical image and the clinical information, and obtaining description information of each focus based on the medical image; under the condition that the historical examination information of the user does not exist, screening description information of a key focus based on the description information of each focus and the clinical information; the image report is generated based on the clinical information, the abnormal sign information and the description information of the key focus, the abnormal sign information and the description information of the focus in each area are obtained by integrating the medical image and the clinical information, key details of the physical condition of a patient can be comprehensively and accurately captured, omission or misjudgment caused by a single information source is avoided, and the accuracy of the patient is improved. And the content of the image report is more complete and accurate.

Some embodiments of the present disclosure provide a framework for using unsupervised artificial intelligence to automatically abstract and align clinical facets. The approach of the present application may be shown to reduce human involvement and, accordingly, enhance privacy compliance. Aspects of the present application relate to a process of self-learning from the data available. Accordingly, aspects of the present application may be shown to be resilient to the appearance of new concepts and facets in future data. Additionally, aspects of the present application may be shown to adapt well when presented with different languages, different styles of documentation and different clinical domains. Aspects of the present application relate to processing unstructured, non-fielded data, such as clinical notes, admission and discharge summaries, surgical notes, lab reports and imaging reports. These notes may be considered to contain hidden insights in the clinical domain. Additionally, these notes may be considered to contain data that may not be captured elsewhere in a readily usable way. Aspects of the present application may be shown to support analysis of large size populations at a relatively low incremental cost.

This invention belongs to the field of medical image data processing technology, specifically disclosing a multi-agent medical image report generation method and system based on fine-grained organ perception. The method includes the following steps: constructing an organ-level medical image dataset and an organ-level report dataset; inputting the organ-level medical image dataset and organ-level report dataset into a multi-agent system based on fine-grained organ perception to extract features, performing multimodal fusion and decoding to obtain an organ-level predicted report; serializing the organ-level predicted report and then sending it to the agent semantic interaction module for unified semantic modeling to generate the final medical image report. This technical solution utilizes a fine-grained organ perception multi-agent architecture to achieve organ-level fine-grained cross-modal modeling and organ-level report prediction, and improves the consistency and completeness of multi-organ information through the agent semantic interaction module, thereby enhancing the accuracy and reliability of the generated medical image report.

The application relates to the technical field of medical data analysis, and discloses a medical data analysis method and system based on a vertical domain intelligent agent, which comprises the following steps: acquiring image report texts, wearable data streams, electronic medical records, test results and other multi-source data, extracting structured entities and time sequence observation values and preprocessing, constructing a continuous multivariate trajectory of a unified time axis, fusing images and wearable information to generate an extended illness representation sequence, calculating local statistical features and combining previous data and a case library to form an association strength vector, weighting clustering a trend sequence to extract decision clues, searching for similar case trajectories and performing distribution matching to obtain a prognosis improvement probability, further correcting the trend sequence through reverse punishment, and finally completing risk stratification, abnormality screening and outputting a comprehensive evaluation report. The method can realize time sequence alignment and association analysis of heterogeneous medical data, and improves the accuracy and timeliness of illness evolution identification and clinical decision making.

The application provides a multi-modal pre-training method for CT image representation and image report generation, relates to the field of natural language processing, and comprises the following steps: obtaining a multi-modal data combination; performing random data enhancement on a CT image, inputting the enhanced data into an image encoder to encode the data, and determining image features; inputting an image report into a text encoder to encode the image report, determining text features, and inputting the image features into a text decoder to determine image descriptions; determining a hybrid loss function according to the similarity of the first image features and the second image features, the similarity of the text features and the image features in a feature space, and the accuracy of the image descriptions and the image report; and performing model training on the hybrid loss function by using a gradient descent algorithm, and updating the parameters of the image encoder, the text encoder and the text decoder. The application can optimize model parameters, improve the representation capability of the model for CT images, and improve the use efficiency of data.

The invention relates to the technical field of image reports, and particularly discloses a chest image report online reasoning method and system integrating image quality evaluation. According to the method, the chest image data is denoised and enhanced, and the image quality is evaluated, so that whether the chest image data has reasoning value or not is judged; extracting standard feature data; lesion reasoning is carried out; generating a basic image report; and carrying out lesion trend analysis, and optimizing the basic image report. The method comprises the following steps of: denoising and enhancing chest image data, evaluating image quality, judging whether the chest image data has a reasoning value or not, performing region identification and image feature extraction if the chest image data has the reasoning value, performing lesion reasoning, generating a basic image report, performing lesion trend analysis, and optimizing the basic image report. The structured and intelligent level of image report reasoning is effectively improved, the situation that the image quality is uneven is reduced through image quality evaluation and judgment screening, and the diagnosis reliability is improved.

The invention discloses a medical image report generation device and method, electronic equipment and a storage medium. The medical image report generation device comprises a feature alignment module which is used for carrying out cross-modal alignment on image features of a medical image and preset text features; the label classification module is used for determining at least one label of a patient corresponding to the medical image, and the label represents a clinical symptom of the patient; and the report output module is used for inputting the at least one label and the image features after cross-modal alignment into a large language model to obtain an image report of the medical image output by the large language model.

The invention discloses an image report quality control method and device based on a large model and a storage medium. Wherein the dynamically updated medical knowledge base and the dynamically updated medical question and answer knowledge base are constructed in advance, and the medical knowledge base is continuously dynamically updated, so that the medical question and answer knowledge base can also be continuously dynamically updated, and the medical question and answer knowledge base can contain question and answer data related to latest knowledge in the medical field. Therefore, when quality control needs to be carried out on the medical image report, the target question and answer data matched with the report text of the corresponding medical image report can be searched from the medical question and answer knowledge base, and the quality control is carried out on the medical image report through the large model in combination with the corresponding target question and answer data. And determining whether a certain problem exists in the medical image report or not. Therefore, by means of the method, the large model can be combined with the latest knowledge in the medical field to conduct quality control on the medical image report, and then the accuracy of quality control on the medical image report is improved.

The invention discloses a medical image report automatic generating and editing method and a computer program product. The method solves the technical problem that an existing medical image report lacks intelligent editing and key information recognition capabilities after being generated. The technical scheme comprises: acquiring a medical image; inputting the medical image into a medical image report generation large model, and generating an initial medical image report containing text description; inputting the initial medical image report and the cue word into a large language model, and extracting a text description corresponding to the cue word from the initial medical image report; and rendering or identifying the extracted text description. According to the invention, the key information in the medical image report can be automatically identified and labeled, and the readability and editing efficiency of the report are improved.

The invention discloses a structured information extraction method and device of an image report and a storage medium. The method comprises the following steps: analyzing a structured template to obtain fields which are defined in the structured template and need to be extracted, a nesting relationship between the fields and format constraint information of the fields to obtain corresponding analysis information, and then, generating prompt information according to the analysis information, so that a large language model can generate prompt information according to the prompt information. According to the method, the to-be-processed text is subjected to structured extraction in sequence according to the nesting relation and the format constraint, target structured information without omitting detailed information is obtained, the entities with the preset logic relation in the to-be-processed text are determined through entity recognition and relation extraction, the target structured information is further verified, and the text processing efficiency is improved. Therefore, the structured extraction is carried out through the method, not only can detail information in the report text of the image report be not missed, but also the accuracy of the structured extraction can be ensured.

The invention relates to a medical image report generation method and a program product. The method comprises the following steps: acquiring a medical image of a target object and a text cue word corresponding to the medical image; inputting the medical image and the text cue word into a medical report generation model for processing, and generating an initial medical report; inputting the medical image into a target disease classification model for target disease detection to obtain a detection result; and fusing the detection result into the initial medical report to form a medical image report of the target object. By adopting the method, the medical image report can be automatically generated, and the medical image report does not need to be written manually in the process, so that the generation efficiency of the medical image report can be improved. In addition, the accurate detection result and the initial medical report are fused in the medical image report, so that the accuracy of the medical image report can be improved.

Provided are a method for breast tumor breast imaging reporting and data system (BI-RADS) classification based on ultrasound and a computer device. The method includes: processing an ultrasound radio frequency (RF) signal of a breast tumor to obtain a lesion location RF signal; processing the lesion location RF signal based on a feature extraction model to obtain a feature pixel matrix; and identifying and classifying the feature pixel matrix based on a preset strategy to obtain a class of the breast tumor. The method for breast tumor BI-RADS classification based on ultrasound provided herein can identify the ultrasound RF signal standardly using more efficient deep learning and intelligent classification and identification algorithms, retain more effective information, and finally obtain an identification result by imitating a method of scoring by a clinician based on signal features during identification and constraining each scoring result using a joint loss function.

This invention relates to the field of healthcare informatics technology, specifically to a dynamic prediction system for the risk of delayed chest closure after surgery in children with congenital heart disease. It collects high-frequency vital signs signals during surgery, imaging report text, and clinical structured data; extracts the maximum Lyapunov exponent, correlation dimension, and approximate entropy as chaotic features; quantifies the ambiguous semantics in the imaging report into membership values; uses Dempster-Shafer evidence theory to fuse multimodal features and encode uncertainty; constructs a dynamic ensemble learning model, detects concept drift using ADWIN and updates it online, dynamically weights it based on Shapley values; uses multi-objective reinforcement learning to dynamically optimize the risk threshold; and generates counterfactual explanations to provide individualized intervention suggestions. This invention achieves accurate and dynamic early warning of the need for delayed chest closure.

The invention provides a method for automatically generating an oral image report, which comprises the following steps of: receiving and processing oral medical image data; classifying the processed oral medical image data, and performing matching based on the classification result and pre-stored oral illness state text data; wherein the oral disease condition text data comprises a disease description text and a diagnosis suggestion text; correlating corresponding image information and text information based on the matching result, and generating a standardized diagnosis report corresponding to the oral medical image; detecting and correcting potential errors in the standardized diagnosis report according to expert knowledge base data, and generating a target report; wherein the expert knowledge base data comprises morphological characteristics of an anatomical structure, disease characteristics and diagnosis suggestions. According to the invention, the standardization level of oral medical diagnosis is improved, the report is more intuitive, and the workload of doctors is reduced.

The disclosure provides a method, device and computer equipment for evaluating radiology image reports, and relates to the technical field of medical image processing and artificial intelligence. The disclosure rewrites the first radiology image report for reference and the second radiology image report generated by the artificial intelligence model into atomic discovery sentences, reduces the matching interference caused by the difference in report writing style, improves the stability and comparability of the evaluation, performs clinical semantic matching on the atomic discovery sentences of the two reports by contradiction restraint, evaluates the second radiology image report according to the atomic level matching result, avoids that the contradictory content gets an unreasonable high score, and improves the accuracy of the evaluation.