21 results about "Image diagnosis" patented technology

The application provides an ophthalmic image diagnosis method and system based on multi-modal imaging cooperation, and relates to the technical field of medical image diagnosis. First, the OCT, fundus camera and ultrasonic original image data of an ophthalmic examination object are acquired, and multi-modal dynamic correlation mapping results are obtained through dynamic correlation and trend correlation processing. Cross-modal lesion feature progressive mining and interactive verification are performed to obtain a cross-modal lesion correlation feature set. Multi-modal cooperative diagnosis reasoning and weight feedback optimization are used to generate an ophthalmic disease reasoning result containing disease types, lesion dynamic distribution and reasoning confidence. Finally, an ophthalmic diagnosis report with dynamic labeling and confidence explanation is generated. The application comprehensively utilizes the advantages of various image technologies to improve the accuracy of ophthalmic image diagnosis.

This invention discloses a cross-modal medical image segmentation method based on pathological anchoring. Using pathological priors as semantic anchors, this method integrates cross-scale collaboration and uncertainty perception in a closed-loop prompting refinement process. This achieves interpretable suppression and robust learning of cross-modal domain shifts, forming a closed-loop path from structure to semantics to noise control. Under strong cross-modal zero-sample settings such as multi-source ultrasound cross-domain segmentation and color dermoscopy, stable and consistent improvements are achieved, manifested in more precise boundaries, better calibration, and more robust generalization. This provides technical support for disease image diagnosis.

The present application relates to the technical field of heart ultrasound image diagnosis, in particular to a heart ultrasound image diagnosis report automatic generation system, comprising an OMM data preprocessing unit, used for processing data and outputting to a high-frequency time domain feature extraction unit and a multi-dimensional space-time correlation unit; the high-frequency time domain feature extraction unit is used for outputting to a coupling control and state discrimination unit; the multi-dimensional space-time correlation unit is used for evaluating a spatial decoupling index and outputting to the coupling control and state discrimination unit; the coupling control and state discrimination unit is used for receiving a time domain instability score and the spatial decoupling index; generating a control signal; reversely sending to the high-frequency time domain feature extraction unit and the multi-dimensional space-time correlation unit; a random process modeling unit is used for generating a probabilistic diagnosis feature set; a diagnosis report generation unit is used for generating a diagnosis report; the present application solves the problem that high-frequency time event fidelity and multi-dimensional space-time correlation integrity cannot be considered simultaneously, and realizes reliable modeling of non-periodic data.

This application provides a method, apparatus, device, and storage medium for self-correction of a medical visual language model based on a dynamic experience base (DEKB), relating to the field of medical visual language processing technology. The method includes: when the initial diagnostic result of the medical visual language model for a current clinical case does not match the fact label, constructing the current case as a structured experience unit and storing it in a dynamic experience knowledge base; upon receiving a query, retrieving historical experience cases related to the new query from the dynamic experience knowledge base; using the historical experience cases as contextual prompts to guide the medical visual language model in chain-like thinking, generating a corrected reasoning result. By constructing an endogenous dynamic experience knowledge base, designing a deep attribution analysis mechanism, and employing a dual-threshold retrieval algorithm, this application enables DEKB to transform the model's historical errors into structured knowledge that can be used for future reference, significantly improving the robustness and generalization ability of medical model image diagnosis reasoning.

This specification relates to the field of artificial intelligence technology, specifically to a medical image diagnosis method, a medical image diagnosis device, and a computing device. The medical image diagnosis method includes: acquiring an unenhanced scan image of the area to be diagnosed, wherein the unenhanced scan image is obtained by scanning the area to be diagnosed without injecting contrast agent; based on the unenhanced scan image, determining a global image and multiple local images of the area to be diagnosed; performing feature encoding using the encoding layer of a medical image diagnosis model to obtain a global feature vector and multiple local feature vectors; performing attention calculation using an input attention layer to obtain a local aggregated feature vector; performing feature fusion calculation using a feature fusion layer to obtain a fused feature vector; and inputting the fused feature vector into a decoding layer for decoding to obtain a medical diagnosis result for the area to be diagnosed. This reduces the diagnostic risk and cost of medical image diagnosis and improves the applicability, accuracy, and efficiency of diagnosis.

The application discloses a kind of self-supervision low-dose CT denoising methods based on decomposition convolutional neural network.First, the low-dose computed tomography (Low Dose Computed Tomography, LDCT) imaging DICOM data is obtained;Second, the value of the pixel point outside the CT scanning boundary in LDCT imaging data is set to 0, and normalized processing is carried out in the range of [-1024, 3072], and it is stored as NPY format;Then, the low-dose CT data set is input into the self-supervision denoising network based on image decomposition for training, and the training model is obtained;Finally, the obtained training model is used to predict new LDCT, and the denoised image is obtained.The application can separate high-quality CT images from LDCT images without labeled images by decomposing LDCT images into clean images, signal-dependent noise images and signal-independent noise images.The application can directly obtain denoised images through the trained deep learning denoising model, which provides technical support for the image diagnosis work of clinicians.

This invention relates to a nuclear-targeted multimodal imaging diagnostic probe with BNCT therapeutic efficacy, its preparation, and its application, belonging to the fields of biomedicine and nuclear medicine technology. This boron drug contains... 10 Using acetylacetone difluoroborate as the parent nucleus, this invention endows the tumor with nuclear targeting ability by introducing N- or S-containing heterocyclic structures, while utilizing its conjugated structure to provide fluorescence / photoacoustic dual-modal imaging functionality. Based on this, the invention achieves three core functions: first, through fluorescence / photoacoustic dual-modal imaging, it enables the visual localization and real-time monitoring of tumor sites; second, by leveraging the binding of nuclear targeting groups to nuclear DNA / RNA, it... 10 B is precisely delivered into the nucleus, significantly improving the killing efficiency of BNCT; thirdly, thermal neutron irradiation is performed under imaging guidance, triggering... 10 The B(n,α)⁷Li nuclear fission reaction ultimately enables high-precision integrated diagnosis and treatment. This invention also discloses a method for preparing this probe, which has promising clinical application prospects.

This invention relates to the field of medical image processing technology and discloses an image enhancement method for breast imaging diagnosis assistance. The method acquires a raw dataset of breast images, comprehensively covering key information across multiple dimensions, including overall image brightness, image zone texture variability, and image floor interference intensity. Then, a dynamic feature representation system is constructed based on this raw dataset, outputting targeted image optimization and control parameters. These control parameters are used to calculate a real-time image quality factor, and a preset baseline image quality norm and the actual image contour evolution rate are fused to accurately generate an image quality deviation coefficient. A bidirectional correction operation is performed based on the quality deviation coefficient, generating brightness modulation and texture enhancement amounts, thereby effectively enhancing the image. The processed image dataset is fed back to the dynamic feature representation system to iteratively optimize the image optimization and control parameters.

The application belongs to the technical field of artificial intelligence, and discloses an adversarial sample generation method based on a double-path self-supervised learning and a random block mask strategy. A double-branch self-supervised collaborative optimization mechanism is constructed in the discriminator aspect, two parallel semantic representation paths of rotation prediction and contrast learning are introduced, multi-level feature interaction of self-supervised signals and traditional adversarial discrimination tasks is realized, and the cross-model migration and semantic robustness of adversarial samples are improved. A dynamic random block mask mechanism is introduced in the generator aspect, structured masks are applied, intermediate layer feature gradients are aggregated, the disturbance contribution of the semantic key area is highlighted, and the pertinence and concealment of the attack are enhanced. The application provides a new idea with attack effectiveness and visual concealment for the field of adversarial sample generation and the field of black box migration, and has important application value in many safety-critical fields such as automatic driving visual safety testing and medical image diagnosis system robustness evaluation.

The application provides a training method and an image reconstruction method of an image reconstruction model, belongs to the field of image processing, and is applied to improvement of spatial resolution of an inertial confinement fusion imaging diagnosis system. The training method comprises the following steps: acquiring a plurality of initial image sample pairs, each initial image sample pair comprising a first image and a second image; inputting the second image into an initial network model to obtain a predicted image; obtaining an image sample pair based on the first image and the corresponding predicted image, and dividing a plurality of image sample pairs to obtain a support set and a query set which do not overlap with each other; using the support set, performing multiple times of iterative updating on model parameters of the initial network model based on a spatial domain loss function and a frequency domain loss function to obtain a plurality of temporary network parameters; using the query set, performing meta loss calculation based on the plurality of temporary network parameters to obtain shared network parameters of the initial network model; and obtaining an image reconstruction model based on the initial network model and the shared network parameters.

A digital breast tomosynthesis (DBT) method includes: disposing a breast of an examination subject between an X-ray irradiator and an X-ray detector; acquiring first image data by emitting X-rays toward the breast of the examination subject; analyzing the first image data to estimate a type of breast lesion by using a pre-trained AI image diagnosis portion; determining an angular range for performing DBT according to the estimated type of breast lesion; performing the DBT in which the X-ray irradiator rotates around the breast within the determined angular range and emits X-rays toward the breast at every predetermined angular interval, to acquire second image data; and reconstructing the second image data to generate a tomosynthesis image, wherein a first dose, which is a total X-ray dose in acquiring the first image data, is smaller than a second dose, which is a total X-ray dose in performing the DBT.

The present application addresses the technical problem of reducing false negatives and false positives in nuclear medicine image diagnosis. The diagnostic assistance device of the embodiments has a collating section and a presuming section. The collating section collates information on the accumulation site of a diagnostic drug, i.e., first site information, obtained from nuclear medicine imaging of a subject, and information on the source site of exosomes including the diagnostic drug, i.e., second site information, collected from the subject. The presuming section presumes that a tumor exists at a site where the collated results are consistent between the first site information and the second site information.

The application provides a pathological section image analysis method and device based on a large language model, and belongs to the fields of computer vision, deep learning and medical image analysis. The method comprises the following steps: constructing an image coding large model for pathological sections composed of a pre-training model and an additionally added task head; freezing the pre-training model, training the task head by using a preset training set and testing by using a preset verification set to obtain a preliminarily trained task head; unfreezing the pre-training model, fine-tuning training the large model by using the training set and testing by using the verification set to obtain a finally trained large model; inputting a pathological section image to be analyzed into the large model, calculating corresponding features of the image based on the result output by the large model to obtain a final analysis result. The application can automatically extract features in the pathological section image and generate a diagnosis report based on the features, so that the accuracy and efficiency of pathological section image diagnosis are improved.

This invention discloses a knowledge graph-based image diagnosis auxiliary decision-making method, device, and medium, relating to the field of medical artificial intelligence technology. The method includes: performing entity recognition and semantic relation extraction on a cleaned medical record text dataset and medical knowledge resources to establish a structured medical knowledge association dataset and generate knowledge graph instances; extracting features from image feature data and text semantic data to obtain image feature vectors, text semantic vectors, and graph node vectors, and mapping them to a unified semantic space using a contrastive learning method to obtain a fused trimodal feature matrix; performing knowledge reasoning on a set of candidate diagnostic paths, calculating the confidence level of disease nodes based on weighted calculations, and generating a candidate disease list and reasoning chain; by mapping the image, text, and knowledge trimodal features to a unified semantic space through contrastive learning and adaptively fusing them, the method improves the coverage and accuracy of auxiliary diagnosis in complex cases and enhances the practicality of clinical auxiliary decision-making.

This invention discloses a perylene-imide-based NIR-II fluorescent dye, nanoparticles, its preparation method, and its applications. The perylene-imide-based NIR-II fluorescent dye consists of a perylene-imide backbone and a polymethylbenzene backbone. Nanoparticles are obtained by ultrasonically mixing the dye with an anticancer drug and an amphiphilic polymer. The synthesis method of this invention is simple, the fluorescent dye has a well-defined structure, and exhibits good water solubility, biocompatibility, and stability. Both its absorption and fluorescence emission spectra are located in the near-infrared II region. The perylene-imide-based NIR-II fluorescent dye nanoparticles prepared by this invention have a fluorescence quantum yield of 1.7% under 980 nm laser irradiation, demonstrating excellent NIR-IIa fluorescence imaging capabilities, enabling high-resolution imaging diagnosis of in situ lung cancer. Under 1064 nm laser irradiation, the photothermal conversion efficiency is 40.3%, exhibiting excellent photothermal conversion performance, allowing for precise NIR-II phototherapy for in situ lung cancer.

Provided is an image diagnosis device capable of displaying cancer cell tissues and inflammatory cell tissues in a distinguishable manner at accurate positions. This image diagnosis device (1) has a GPU (1a), a storage device (1b), and a display device (1c). A CT image of a certain portion of a human body is input to the GPU (1a) from a CT device (2), and a biological resonance measurement image of the portion is input to the GPU (1a) from a biological resonance measurement instrument (3). In addition, the cancer cell tissues and the inflammatory cell tissues of the portion in the biological resonance measurement image are displayed superposed in the CT image on the basis of the locations thereof identified by the biological resonance measurement instrument (3).

The application discloses a kind of photovoltaic power station power generation area inspection control system, the inspection control system includes: unmanned aerial vehicle, for collecting photovoltaic module's thermal infrared image and visible light image;Weather station, for real-time acquisition photovoltaic power station's irradiance, ambient temperature and wind speed, and send inspection request to monitoring device;Analog map, for recording the geographical location of photovoltaic module string and the correspondence of photovoltaic module string and inverter;IV detection device based on inverter, for scanning photovoltaic module string and obtaining the IV curve of corresponding photovoltaic module string;Monitoring device, including IV diagnosis module, route planning module, inspection instruction module, image diagnosis module and defect processing module.The photovoltaic power station power generation area inspection control system of the application, the advantages of inverter I-V wide range fast diagnosis and the advantages of unmanned aerial vehicle accurate positioning and identifying defect ability are combined, and the operation and maintenance workload and operation and maintenance technical difficulty are reduced by the combination and running of data.

The application discloses an AI-based intelligent diagnosis system for pancreatitis CT images and relates to the technical field of CT image diagnosis. The system comprises a data preprocessing module, a 3D feature extraction branch module, a 2D feature extraction branch module and a double-branch feature fusion module. The data preprocessing module is used for receiving CT volume data of a target object and performing preprocessing to obtain standard data. The 3D feature extraction branch module is used for extracting global spatial context features and multi-scale features of the standard data, and performing matching and weighting based on preset fixed scale anchor points to obtain 3D features. The 2D feature extraction branch module is used for performing multi-window feature extraction and fusion on the standard data, and performing edge feature enhancement and pathological feature enhancement on the fusion result to obtain 2D features. The double-branch feature fusion module is used for fusing the 3D features and the 2D features, dynamically adjusting the contribution weight of the double branch according to the lesion type, and generating a comprehensive feature map. The multi-task output module is used for outputting results according to the comprehensive feature map. The application realizes adaptive feature expression for different lesions.

The present application belongs to the technical field of medical image processing, and discloses a breast ultrasound panoramic image diagnosis system, a training method thereof, and a breast multi-modal diagnosis method. The original panoramic ultrasound image without manual region of interest (ROI) cropping is input into a hybrid visual encoder, the long-range dependence is captured by a GSA module, and the high-frequency detail features of the lesion edge are extracted by an LMA module; the global and local features are transformed to the frequency domain for deep interaction and fusion of the real part and the imaginary part by a context detail frequency fusion module; in the multi-modal alignment stage, a debiased contrast loss function is introduced, the false negative deviation in negative sample sampling is corrected by estimating the class prior probability, and robust alignment of image and text features is realized; finally, the multi-modal diagnosis result is output by fusing the dual-modal features. The present application does not need to rely on expensive manual ROI labeling, can directly realize high-precision automatic diagnosis on the full field of view image, and greatly simplifies the clinical deployment process.

An image diagnostic catheter of the present application has: a sheath for insertion into a living body; an imaging core portion located inside the sheath; and a drive shaft mounted to a proximal side of the imaging core portion and located inside the sheath, the imaging core portion including: an ultrasonic transducer having an ultrasonic wave transmitting / receiving surface and a back surface opposite the ultrasonic wave transmitting / receiving surface; a back surface material supporting the back surface of the ultrasonic transducer and absorbing a portion of ultrasonic waves transmitted from the ultrasonic transducer; and a support member supporting the back surface of the ultrasonic transducer via the back surface material, the drive shaft being movable inside the sheath along an extension direction of the sheath together with the imaging core portion and being rotatable around an axis of the sheath together with the imaging core portion, the back surface material having a distal covering portion covering a distal end surface of the ultrasonic transducer.

This invention relates to the field of radiology, specifically to a film reading device for radiological image diagnosis. The device includes a hollow housing with symmetrically arranged fixing components on its side walls. Each fixing component includes a pressure plate. A first pull rod and a second pull rod are fixedly connected to the side wall of the pressure plate near the housing, extending into the housing and slidably connected to the side wall. A protrusion is fixedly connected to the bottom of each pull rod, and a pressure block is fixedly connected to the side of the protrusion near the pressure plate, extending out of the housing. A sliding plate is slidably connected inside the housing, with the first and second pull rods fixed to its side wall. A pressure rod is fixedly connected to the other side of the sliding plate, and a linkage component for assisting in clamping the film is provided at the other end of the pressure rod. This invention, through the linkage component formed by the pressure plate, pressure block, and airbag, effectively prevents the film from sticking too tightly to the housing wall and becoming difficult to remove.