35 results about "Radiology report" patented technology

The invention discloses a radiology report generation method and system based on global dependency learning, and belongs to the technical field of natural language processing. According to the method, a global dependency learning module is designed, and the module integrates rotation position coding to enhance original Mama, so that long-sequence visual feature dependency is effectively modeled, and the model calculation efficiency is improved. According to the method, the problems that the visual long-term dependence modeling and calculation efficiency in the radiology image are difficult to effectively balance, and the heterogeneity of the characteristics of different modal information of the image and the text is difficult to effectively align and fuse are solved; the method not only can effectively capture the long-term dependency relationship on the key visual features in the radiology image, but also can improve the model calculation efficiency, and can enhance the alignment and fusion capability among the multi-modal heterogeneous information.

The present invention relates to an AI-based system and method for generating enhanced radiology reports. The system comprises a database for storing multimodal patient data, a natural language processing (NLP) module for extracting clinical information, and a machine learning module for correlating the clinical information with radiology images to identify diagnostic insights. An AI-based report generation module analyzes the images and clinical information to generate a preliminary report, which is refined based on radiologist input. The generated report is then integrated into the patient's electronic health record. The system employs techniques such as multimodal deep learning, active learning, explainable AI, and federated learning to enhance diagnostic accuracy, capture expert feedback, provide transparency, and enable multi-institutional collaboration. The invention aims to improve the accuracy, efficiency, and value of radiology reporting in patient care.

A computer-implemented method for assisting radiologists in efficiently preparing radiology reports from diagnostic images is disclosed. The method includes processing radiology images using artificial intelligence to automatically detect anatomical structures and pathologies, and generating positional and descriptive data for each detected feature. An initial radiology report, fully populated with the detected features, is automatically generated prior to user interaction and displayed through a user interface comprising synchronized image and text panels. The radiologist reviews this initial report by selectively adding, modifying, or deleting features through a user interface input that identifies each feature and an associated action. The report is updated immediately based on these inputs, ensuring continued synchronization between image annotations and their descriptive narratives. This approach reduces reporting turnaround times, decreases cognitive workload, and minimizes diagnostic errors.

The invention discloses a learnable radiology report generation method based on visual text alignment and fusion of retrieval enhancement. The method comprises the steps of collecting and respectively constructing a model training data set and an auxiliary data set, then constructing a visual text alignment and fusion model based on retrieval enhancement, and inputting the model training data set and the auxiliary data set into the visual text alignment and fusion model based on retrieval enhancement together for training. Constructing an inference model according to the trained visual text alignment and fusion model based on retrieval enhancement; and inputting the to-be-detected medical image and the auxiliary data set into the reasoning model for processing to obtain a radiology report corresponding to the to-be-detected medical image. According to the method, retrieval correlation is enhanced, meanwhile, a fine-grained vision-text alignment and fusion method is adopted to align and fuse features, and the problem that fine-grained region-sentence alignment is difficult due to weak supervision of an image report level in the medical report generation process is solved.

The invention discloses a radiology report generation method based on hierarchical interactive fusion, and relates to the technical field of medical image intelligent processing, and the method comprises the steps: collecting original image data, carrying out the intensity normalization, obtaining normalized image data, carrying out the multi-level visual feature extraction and feature dimension unification of the normalized image data, and carrying out the multi-level visual feature extraction and feature dimension unification of the normalized image data; forming a hierarchical visual feature set; coding processing is carried out on the hierarchical visual feature set, cross-layer attention relations between a shallow layer and a deep layer and between a middle layer and the deep layer are constructed in the coding process, a shallow layer two-dimensional biased field and a middle layer two-dimensional biased field are formed, and a migration smoothness index and a consistency index are obtained after nonlinear resampling; and in the decoding stage, a multi-path cross attention structure is constructed based on the coded hierarchical visual feature set, an abnormal priori graph is constructed according to the shallow two-dimensional biased field and the middle two-dimensional biased field, attention bias is formed, and a cross-hierarchical cross attention aggregation vector is generated. According to the invention, association expression of multi-level visual features is realized.

This specification provides a cardiac ultrasound coronary artery stenosis diagnostic system and method based on a visual language transfer model. The system includes: a database construction module for standardizing multi-view cardiac ultrasound video sequences and radiological reports; a cardiac region of interest (ROI) standardization and labeling module for segmenting myocardial segments, obtaining ROI data, and generating a coronary artery-specific perfusion region mask as the ROI for vascular-level stenosis prediction; a visual language transfer pre-training module for semantic alignment of video data and report text using a visual language transfer model; and a downstream task prediction module for utilizing the enhanced attention mechanism of the coronary artery-specific perfusion region mask, based on the global visual features output by the visual language transfer pre-training module, to output patient-level classification results for obstructive coronary artery disease and vascular-level stenosis prediction results for LAD, LCX, and RCA.

The invention belongs to the technical field of multi-modal natural language generation, and provides a radiology report generation method and system based on feature fusion and medical guidance, and the method comprises the steps: carrying out the feature extraction of medical image data and text report data, obtaining the text features with enhanced medical knowledge, and carrying out the feature extraction of the text features; obtaining image space-time characteristics reflecting the evolution information of the focus; injecting the medical knowledge enhanced text features into image space-time features by using a cross attention mechanism to generate multi-modal space-time fusion features; and according to the multi-modal space-time fusion features and a preset decoder, generating a radiology department medical text report, and by integrating historical and current multi-temporal and multi-view image data of a patient, combining anatomical structure constraints and lesion evolution rule modeling, on the premise of guaranteeing term normalization and logic coherence, establishing a medical text report of the radiology department. Longitudinal lesion evolution information and multi-view complementary information can be fully utilized, and the professionality and accuracy of reports are improved.

In one embodiment, a method comprising accessing radiographic images of an animal, wherein one or more first radiographic images of the radiographic images depict the animal from one or more views, respectively, and wherein one or more second radiographic images of the radiographic images depict one or more body parts of the animal, respectively, determining disease classifications associated with the animal based on analyzing the radiographic images by a machine learning model, generating a diagnostic report associated with the animal based on the machine learning model, wherein the diagnostic report includes the disease classifications and a natural-language textual radiology report, and sending instructions for presenting the diagnostic report to a user device.

The invention provides a system and method for quality control in radiology reports. It collects data from reports, medical scan images, and questionnaires within a Safety, Quality, Efficiency, and Productivity (SQEP) framework. Scores for safety, quality, productivity, and efficiency are computed for radiologists and departments using weighted parameters, peer reviews, and audit outcomes. A consolidated SQEP score evaluates overall performance. The system identifies deviations in report quality through trend analysis and correlations, providing corrective feedback and training recommendations to radiologists. By addressing errors proactively, the system enhances diagnostic accuracy, operational efficiency, and compliance with quality standards. This comprehensive approach ensures improved performance and reliability in radiology departments while maintaining high-quality patient care.

The invention discloses a radiology report generation method and system based on focus guide mask and knowledge graph enhancement, and belongs to the technical field of medical artificial intelligence. The method comprises the following steps: preprocessing a chest medical image and extracting local features; constructing a lesion knowledge graph containing an organ-lesion-disease category triple; obtaining a candidate organ and a mask image thereof through a pre-trained focus detection model; retrieving related knowledge from the knowledge graph based on the candidate organs to generate focus enhanced knowledge features, and inputting the organ mask and the original image into a mask image attention layer together to obtain mask enhanced image features; using a multi-branch gating cross-modal fusion module to carry out adaptive weighted fusion on the two types of features to obtain knowledge mask enhanced fusion representation; and finally, a radiology report is generated through a Transform encoder-decoder. According to the method, image-text alignment of organ granularity can be realized, language illusion in a generated report is effectively reduced, and clinical consistency and interpretability of the report are improved.

Methods and systems described enable automatic generation of a significant portion of or all of a clinical report (e.g., radiology report), using multimodal models trained on image and language data. Methods described can transform unstructured language and image information into findings, as well as an accurate and comprehensive clinical report, in a designated style (e.g., writing style). The methods and systems described thus significantly improve performance in generation and processing of clinical reports, in relation to time saved per clinical shift, dictation effort, medical billing, and other performance factors.

The invention discloses a training method and device for a multi-modal representation learning model. The method comprises the following steps: classifying medical image data by a representation learning model to obtain a first category vector, and classifying an iconography report text to obtain a second category vector; calculating the similarity between each first category vector and all second category vectors to generate a first similarity matrix, and normalizing the first similarity matrix to obtain a first probability distribution matrix; calculating the similarity between each second category vector and all the first category vectors to generate a second similarity matrix, and normalizing the second similarity matrix to obtain a second probability distribution matrix; and performing parameter adjustment on the representation learning model according to the first probability distribution matrix and the second probability distribution matrix to obtain a multi-modal representation learning model. According to the invention, through a two-way alignment mechanism of label similarity and feature similarity, a high-precision multi-modal representation learning model is obtained through training under limited medical annotation data, and the medical assistance value of the model is improved.

A system and method for generating radiology reports. The systems and methods display an image of a region of interest on a display, determine image characteristics of the region of interest, determine, via a processor, a recommendation based on the image characteristics, and generate, via the processor, a report including the recommendation.

The application discloses a kind of gradient optimization method and system of multitask radiology report generation, it is related to medical information technology field, including, obtaining medical training data set containing image, label and report text, construct and include visual encoder, clinical constraint auxiliary task branch and text decoder Multi-task report generation model;Firstly, the failure mechanism of linear scalarization is studied from the perspective of gradient dynamics, and it is formalized as the "double dilemma" of drift term bias and diffusion term attenuation using the SDE framework, thereby revealing the geometric root cause of suboptimality in RRG multitask optimization, and proposing CAME-Grad, a gradient optimization algorithm designed specifically for multitask RRG. As an optimizer independent of the backbone network, it can be integrated into various model architectures in a plug-and-play manner. On the MIMIC-CXR and IU X-Ray datasets, CAME-Grad was extensively evaluated across eight representative RRG methods, and the results showed that its average clinical performance improved by 2.3% and 1.9%, respectively.

This invention discloses a method for generating radiological image reports based on explicit visual evidence. The specific implementation steps include: constructing a training set for a clinical diagnostic inference chain based on anatomical priors; fine-tuning a pre-trained multimodal large model using a ternary collaborative loss function; extracting high-resolution grid features during the inference stage, locking key lesion features through an active retrieval mechanism triggered by interleaved tokens, and embedding these features into a contextual sequence after processing by a perceptual resampling module as explicit visual evidence, driving the model to generate a radiological report supported by the image. This invention achieves a paradigm shift from static full-image injection to dynamic evidence-based inference, overcoming the problems of opacity and easy neglect of small lesions in the inference process of existing radiological image reports, while improving the diagnostic reliability and computational efficiency of the model while preserving high-resolution details.

A weakly supervised intracranial hemorrhage (ICH) detection workflow includes training a deep learning (DL) model including a coupled convolutional neural network and recurrent neural network on a large dataset of CT scans with expert-labeled slices indicating presence or absence of ICH. Transfer learning (TL) is used to further train the DL model using a second large dataset of CT scans with only scan labels extracted from radiology reports using natural language processing (NLP). The DL model weights each slice of the scan against the final ICH diagnosis using an attention-based bi-directional long-short term memory network, where the attention weights represent slice-level ICH predictions. Model-generated heatmaps highlight significant regions of the CT scans that lead to the provided ICH predictions.

The present application provides a radiology report generation method fusing clinical semantic modulation and hyperbolic prototype classification, which comprises: acquiring chest X-ray images and their associated clinical context knowledge, extracting visual features of the images using a visual encoder, extracting clinical semantic embedding using a medical language encoder, and retrieving relevant report features from a reference report database; constructing and utilizing a clinical semantic modulation module to generate modulated visual representation; constructing and utilizing a hyperbolic prototype classification module to generate diagnosis awareness prompts; inputting the modulated visual features and diagnosis awareness prompts into a decoder to complete the autoregressive generation of the radiology report. The present application can fully utilize the cross-modal fusion of clinical knowledge and visual features, and utilize the exponential expansion embedding capacity of hyperbolic space to improve the disease detection ability under the condition of class imbalance, and improve the clinical accuracy of radiology report generation.

The application discloses a longitudinal chest radiograph progress monitoring method and system based on anatomical anchor semantic difference, relates to the field of medical image processing, and generates a shared anatomical support mask by performing spatial union on candidate boxes of front and rear chest radiographs in a unified coordinate system; the mask and the original image are soft fused through a learnable gate to obtain an image enhanced in anatomical anchoring; longitudinal difference processing is performed on the front and rear enhanced images in an image-report semantic embedding space to obtain report-aware semantic difference, a visual encoder is used to extract multi-scale visual features to calculate difference features, the difference features are refined in nonlinearity, the refined difference information is asymmetrically injected into current visual features, and finally, multi-scale fusion, semantic recalibration and progress classification are completed. The application can stably, accurately and explainably monitor the disease performance changes in the front and rear chest images of the same patient without relying on the input of radiology reports in the reasoning stage.

This invention relates to a medical image report generation method based on a dual-image encoder, comprising: acquiring a current chest X-ray image and historical chest X-ray images; inputting the current chest X-ray image and historical chest X-ray images into a medical image report generation model to obtain radiology report text corresponding to the current chest X-ray image; wherein, the medical image report generation model includes: a single-modal encoding module for extracting image features of the current chest X-ray image and historical chest X-ray images, as well as text features of the radiology report text; a cross-modal encoding module for performing cross-modal fusion of the image features and text features to obtain multimodal features incorporating visual difference information; and a text decoding module for generating the radiology report text corresponding to the current chest X-ray image based on the multimodal features incorporating visual difference information. This invention enables the modeling of image difference information and the generation of corresponding radiology reports.

The invention provides a radiology report generation method fusing clinical semantic modulation and hyperbolic prototype classification, and the method comprises the steps: obtaining a chest X-ray image and clinical context knowledge associated with the chest X-ray image, extracting the visual features of the image through a visual encoder, extracting clinical semantic embedding through a medical language encoder, and generating a radiology report. Related report features are retrieved from the reference report database; constructing and utilizing a clinical semantic modulation module to generate modulation visual representation; constructing and utilizing a hyperbolic prototype classification module to generate a diagnosis perception prompt; and inputting the modulated visual features and diagnosis perception prompts into a decoder to complete autoregression generation of the radiology report. According to the method, cross-modal fusion of clinical knowledge and visual features can be fully utilized, the exponential expansion embedding capacity of the hyperbolic space is utilized to improve the disease detection capability under the class imbalance condition, and the clinical accuracy of radiology report generation is improved.

A method (100) of training a machine-learned (ML) image classifier (14) to classify images (30) respective to a set of labels includes: generating image-based labels for the images from the set of labels and image-based label confidence values for the image-based labels by applying the ML image classifier to the images; generating report-based labels for the images from the set of labels and report-based label confidence values for the report-based labels by applying a report classifier (16) to corresponding radiology reports; selecting a training subset of the set of images based on the image-based labels, the report-based labels, the image-based label confidence values, and the report-based label confidence values; assigning a pseudo-label for each image of the training subset which is one of the image-based label or the report-based label for the image; and training the ML image classifier using at least the selected training subset and the assigned pseudo-labels.

The invention discloses a radiology report generation method and system based on a multi-expert model, and belongs to the field of computer vision. According to the method, a visual encoder and a text encoder are used for joint modeling, and early fusion of an image and a clinical auxiliary text is realized through a gating mechanism, so that cross-modal abstract features are better captured, and the dependency problem in long text generation is relieved; an expert network is dynamically selected by adopting a stage-view angle multi-expert module, so that style self-adaption under different doctor seeing stages and image view angles is realized; a hierarchical prefix query converter module is introduced, and prefix key value pairs are injected layer by layer in multi-layer attention of the decoder, so that semantic alignment is enhanced; through multi-task loss joint optimization including cross entropy loss, impression comparison loss and expert load balancing loss, the smoothness and clinical consistency of the generated language are improved. The method is obviously superior to the prior art in a radiology report generation task, and can also be popularized to other medical scenes needing cross-modal modeling and long text generation.

The invention belongs to the technical field of medical artificial intelligence, and particularly discloses a radiology report generation method based on focus attention and semantic knowledge fusion. According to the invention, based on the pre-trained medical image encoder, multi-focus label prediction is carried out on the radiation light image; performing coarse and fine granularity retrieval on the similar images and the corresponding similar reports; mapping the identified medical entities to a unified medical semantic system; and generating a radiology report corresponding to the radiation light image according to the screened candidate reports based on a report integration generation strategy of chain reasoning. According to the mode, the high-precision candidate reports are screened by adopting a multi-stage retrieval enhancement mechanism formed by label matching, coarse and fine granularity retrieval and knowledge screening, retrieval errors are remarkably reduced, a report integration generation strategy based on chain reasoning controls a large language model to generate texts only in a reference range, medical illusion is avoided, and the accuracy of retrieval is improved. Therefore, the accuracy and interpretability of generating the radiology report can be effectively improved.

Embodiments disclosed herein include software processes executed by a computer for ingesting and analyzing MRI image data and generating MRI-related reports. The report generation software executed by a computer generates MRI reports using medical data and MRI imagery data. The computer creates a text summary. Additionally, the computer generates the MRI report to include medical illustrations that overlay and merge with the MRI imagery data to visually explain the MRI results. Generative machine-learning models integrate colorized MRI images with the illustrations. The computer selects key MRI slices, colorizes them, matches them with relevant illustrations, and recursively refines the combined images to improve clarity and informational content. This process enhances the interpretability of radiology reports for both medical professionals and patients.

The application discloses a cross-modal gate fusion radiology report generation method combined with clinical semantic enhancement, and relates to the technical field of medical report generation. In the application, a semantic enhancement decoding unit is used for deeply fusing image sequence representation, historical text embedding and clinical semantic representation, and obtaining a final enhanced decoding representation of multi-modal semantic enhancement; and the final enhanced decoding representation output by the semantic enhancement decoding unit is subjected to report generation, so that a predicted radiology report is obtained. The predicted radiology report generated by the radiology report generation method disclosed in the application is closer to a radiology report sample text in terms of sentence structure and coherence.

A system (100) can receive a radiology report including radiologist observations about a region of interest of a subject and use a first AI model (310) to generate structured data including a predetermined set of fields and a corresponding set of values extracted from the radiology report. The system (1000) can use the first AI model (310) to determine, using the structured data, a medical image dataset of the subject stored in a medical image database (150) corresponding to the radiology report. The system (100) can use a second AI model (320) to segment and label a set of regions of each medical image of the medical image dataset. The system (100) can use the first AI model (310) to determine, using the structured data, a medical image of the medical image dataset that depicts the region of interest of the subject and perform an action based on determining the medical image that depicts the region of interest.

A system may receive a radiology report including a finding of a radiologist with respect to a region of interest of a subject, and generate, using a first AI model, structured data including a set of predetermined fields and a set of corresponding values extracted from the radiology report. The system may determine, using the first AI model, a medical image dataset of the subject corresponding to the radiology report stored in a medical image database using the structured data. The system may segment and label, using a second AI model, a set of regions of each medical image of the medical image dataset. The system may determine, using the first AI model, a medical image of the medical image dataset that depicts the region of interest of the subject using the structured data, and perform an action based on determining the medical image that depicts the region of interest.

The invention discloses an intelligent identification method and system for benign and malignant pulmonary nodules based on multi-modal fusion, relates to the technical field of pulmonary nodule diagnosis, and aims to solve the problems of insufficient information utilization and low identification accuracy in the prior art. The specific operation process comprises the steps that lung image data, clinical information and a radiology report of a patient are collected, and the image data are preprocessed; performing lung nodule segmentation on the preprocessed lung image data; image omics features, depth features and text features are extracted from the segmented regions; performing three-dimensional fusion on the radiomics features, the depth features and the text features to obtain a comprehensive feature vector; and performing classification model construction and training on the comprehensive feature vector. According to the method, the patient information is fully utilized, the identification accuracy and generalization ability are improved, automatic and accurate identification can be realized, reliable reference is provided for doctors, and great help is provided for improving the diagnosis efficiency.

The application provides a chest radiograph report generation method and system based on cross-modal alignment and salient semantic regions, and is applied to the technical field of data processing. The chest X-ray chest radiograph image information of a target patient and corresponding initial radiology report text information are processed based on a network model for identifying salient regions of medical semantic information to generate salient regions of medical semantic information and a saliency map; the chest X-ray chest radiograph image of the target patient and the salient region image of medical semantic information are processed based on a mask image modeling module guided by the salient region to generate image block features after mask image reconstruction; the saliency map and the image block features after mask image reconstruction are processed based on a language generation model guided by the saliency map to generate target radiology report information; and the target radiology report information is processed to generate an evaluation result of chest radiograph report generation.

The invention relates to the technical field of artificial intelligence and medical image processing, and discloses a lung cancer diagnosis and prognosis system based on a visual language basic model, which comprises a database construction module, a visual language comparison pre-training module, a radiology report generation module and a downstream task prediction module. The database construction module is used for carrying out standardization processing on the 3D chest CT and the report; the pre-training module realizes semantic alignment of image and text features through joint optimization comparison loss and auxiliary classification loss; the report generation module compresses visual features by using a 3D space pooling perceptron, and inputs a large language model to generate a diagnosis report; and the downstream task prediction module multiplexes a visual encoder and outputs malignant classification, typing, transfer, survival prognosis and other multi-dimensional evaluation results. Through cross-modal alignment and multi-task joint learning, the problems of 3D image feature processing bottleneck and modal splitting are solved, and automatic evaluation and auxiliary diagnosis of the whole process of lung cancer are achieved.