396 results about "Imaging diagnosis" patented technology

The invention relates to the technical field of sound wave measurement, in particular to an ultrasonic image diagnosis system and method.According to the ultrasonic image diagnosis system and method, matching between reflection characteristics and tissue characteristics is made to have the self-adaptive learning ability through a deep neural network, the targeted recognition effect is improved in the aspect of signal classification, and based on the judgment result of the reflection characteristics, the diagnosis accuracy is improved. When boundary partitioning is carried out on a tissue area, the edge structure is judged by using the combination of three parameters of gray range, gradient direction and texture continuity, contour fuzziness caused by single index judgment is avoided, and the boundary partitioning accuracy is improved by extracting a gray distribution center, an amplitude change track and an edge area continuous change sequence. And the region positioning result is input into a support vector machine, accurate recognition of lesion properties is realized according to boundary classification comparison of morphological structure quantitative features and historical benign and malignant feature data, secondary verification of the structural form is performed after the image is formed, and the diagnosis integrity and the judgment confidence are effectively improved.

The invention discloses a multi-modal X-ray image diagnosis report generation method based on reinforcement learning optimization. The method comprises the following steps: acquiring a multi-source public data set and a private data set in a specific range; performing format standardization processing on the X-ray image, performing normalization processing on text information, and constructing a structured public training data pair and a private enhanced data pair; training an end-to-end multi-modal neural network by using the public training data pair in a supervised fine tuning mode; carrying out reinforcement learning on the first-stage end-to-end multi-mode neural network by adopting a strategy gradient algorithm optimization mode of GRPO and using candidate reports and weighted rewards to obtain a second-stage end-to-end multi-mode neural network; and generating a diagnosis report by using the two-stage end-to-end multi-modal neural network. The method is suitable for personalized application scenes of X-ray image diagnosis, can quickly adapt to specific clinical requirements on the basis of limited professional data, and has good practical value and popularization prospect.

The invention discloses a brain tumor image diagnosis large model pre-training method, diagnosis method and system, and the method comprises the steps: obtaining the image data of a brain tumor patient and a corresponding diagnosis text, and the image data comprises a plurality of sequences; constructing a visual unified model, carrying out complete sequence standard training and missing sequence distillation training on the visual unified model by utilizing the image data, learning unified visual representation of the image data of any sequence combination, and taking characteristics of the complete sequence image data as teacher characteristics in the missing sequence distillation training; the features of the missing sequence image data serve as student features, and distribution alignment of the student features and the teacher features in the feature space is restrained; constructing a visual language model, taking unified visual representation output by the trained visual unified model as input based on a multi-task target, and training the visual language model in combination with the diagnosis text; the trained model can effectively process the sequence missing condition, and the accuracy and robustness of diagnosis are improved.

The invention relates to the technical field of medical image processing, in particular to a neurosurgery image diagnosis method and system based on image processing, and the method comprises the following steps: obtaining gray matter edge nodes of a triaxial section, constructing a symmetric path unit, collecting an edge direction vector, and generating a direction trajectory diagram; and extracting continuous slices, constructing a rotation track sequence, identifying an abnormal region, filling gaps, combining path voxels, and dividing spatial levels to generate an image structure chart. According to the method, the grey matter edge nodes in the three-axis tangent plane are obtained, and the node paths with the symmetrical characteristics are screened out according to the space projection trend, so that the continuous region of the structure can be accurately recognized, the direction vectors in the continuous slices are extracted, the direction mutation region in the track is recognized, and the jump and fracture performance of the structure can be timely captured; and the abnormal region is accurately labeled, so that higher-dimensional expression and finer-grained recognition of the neural structure are realized, and spatial modeling and visual analysis of complex neuropathy are effectively supported.

The embodiment of the invention discloses an oral and maxillofacial surgical image recognition and diagnosis method and system based on deep learning, and the method comprises the steps: firstly obtaining an oral and maxillofacial three-dimensional image data set of a target patient, then carrying out the image feature extraction processing of the three-dimensional image data set, and obtaining a hierarchical image feature set; comprising local anatomical structure features and global spatial distribution features, and then calling a pre-trained multi-scale feature fusion network to perform multi-scale feature fusion on the hierarchical image feature set to generate a fusion feature map. And performing focus area identification processing based on the fusion characteristic spectrum, determining position information and form description information of an oral and maxillofacial abnormal area of the target patient, generating a diagnosis report according to the position information and form description information of the oral and maxillofacial abnormal area, and transmitting the diagnosis report to medical terminal equipment for display. Therefore, the accuracy and efficiency of oral and maxillofacial surgery image diagnosis are improved.

The invention discloses a multi-stage loss optimization OCTA (Optical Coherence Tomography Angiography) image diagnosis method based on semi-supervised learning and knowledge distillation. The method comprises the following steps: (1) constructing a teacher-student model framework in combination with semi-supervised learning; (2) data set processing: randomly removing part of sample labels of the training set as label-free data; (3) training a teacher model by using the labeled data, then generating a pseudo label of non-labeled data through the teacher model, and only retaining a high-quality pseudo label meeting a confidence threshold; and (4) training a student model by using a multi-stage loss function, wherein the student model comprises supervision loss of label data, supervision loss of pseudo-label data and knowledge distillation loss. Dependence on annotation data is reduced through semi-supervised learning, the performance of the lightweight student model is improved through knowledge distillation, and the method is suitable for scenes with limited clinical resources.

An AI image diagnosis apparatus into which three-dimensional tooth image data captured by a dental OCT device is inputted and which analyzes the inputted three-dimensional tooth image data, comprising: a model executor that sequentially inputs two-dimensional tomographic image data constituting the diagnostic target three-dimensional tooth image data into a trained model, to thereby obtain, as an execution processing result of the trained model, lesion information data, which is data related to a part identified as a characteristic part such as a lesion in the inputted tomographic image data, for each piece of tomographic image data in the three-dimensional tooth image data, and detect lesions from the inputted three-dimensional tooth image data using the obtained lesion information data, wherein the trained model is constructed by training of three-dimensional tooth image data of multiple examinees captured by a dental OCT device in the past.

The invention discloses an X-ray imaging diagnosis system based on a Timepix3 detector, and belongs to the technical field of plasma diagnosis. The system comprises a Timepix3 detector, a data server, an upper computer, a signal generator and a high-voltage power supply, the signal generator is used for performing remote control on the Timepix3 detector and providing pulses during plasma discharge to ensure that the Timepix3 detector acquires data in a trigger mode; the high-voltage power supply is used for providing bias voltage for a semiconductor sensor of the Timepix3 detector; the upper computer is used for sending an instruction to the Timepix3 detector to carry out data acquisition; and the data server is used for receiving and storing the collected data. The device can be applied to a plasma discharge experiment carried out in a Tokamak device, and has higher spatial resolution, stronger time resolution capability and capability of quickly acquiring an X-ray signal energy spectrum.

The invention provides a pathological section image analysis method and device based on a large language model, and belongs to the field of computer vision, deep learning and medical image analysis. The method comprises the following steps: constructing an image coding large model which is composed of a pre-training model and an additionally-added task head and aims at a pathological section; freezing the pre-training model, training a task head by using a preset training set, and checking by using a preset verification set to obtain a task head after preliminary training; unfreezing the pre-training model, performing fine tuning training on the large model by using the training set, and performing inspection by using the verification set so as to obtain a finally trained large model; and inputting a to-be-analyzed pathological section image into the large model, and calculating corresponding features of the image based on a result output by the large model to obtain a final analysis result. According to the method, the features in the pathological section image can be automatically extracted, and the diagnosis report is generated based on the features, so that the accuracy and efficiency of pathological section image diagnosis are improved.

The invention belongs to the technical field of brain CT image diagnosis, and discloses a diagnosis management system based on brain CT image analysis. Comprising an acquisition module used for acquiring basic information, clinical information and brain CT images of a patient; the diagnosis module is used for comparing the brain CT image of the patient with a pre-constructed normal brain model to obtain abnormal brain information, and inputting the abnormal brain information into a pre-constructed brain disease diagnosis model to obtain a diagnosis result; according to the invention, tiny structural abnormality, vasculopathy and brain sulcus change in a brain CT image can be captured, and the accuracy and efficiency of brain disease diagnosis are improved; the diagnosis result of the patient can be matched to generate a disease development path, so that a doctor and the patient are helped to know the clinical stage of an illness state, and the disease development speed of the patient can be adjusted by setting the gain factor and the inhibition factor.

The invention discloses a thyroid ultrasound image diagnosis method based on deep learning, and the method comprises the following steps: collecting a thyroid ultrasound original image set, and carrying out the preprocessing; performing focus segmentation on the standardized thyroid ultrasound image set; performing morphological constraint and boundary refinement; calculating the blood flow direction, blood flow velocity and blood flow power of each thyroid focus area and neighborhood; generating a preliminary fusion feature map based on a feature adaptive deep learning network, and fusing the preliminary fusion feature map with the thyroid focus blood flow feature vector set; obtaining a thyroid focus detection list through thyroid focus benign and malignant discrimination branches; and generating thyroid focus structured diagnosis data records based on the thyroid focus detection list, and writing the thyroid focus structured diagnosis data records into a computer-aided diagnosis system. According to the method, deep learning and multi-modal blood flow features are fused, intelligent diagnosis of the thyroid focus is achieved, and the method has the advantages of being high in precision, high in interference resistance and structured in result.

The invention discloses a medical analysis method and system based on a multi-modal large language model and chain reasoning, and belongs to the crossing field of artificial intelligence and medical image analysis. The method comprises the following steps: firstly, carrying out preprocessing and modal alignment on medical multi-modal data to obtain image visual features and text embedding vectors of the same vector spatial dimension; capturing visual evidence in a structured pathological feature form through a self-excitation mechanism; on the premise of visual evidence, a thinking chain text containing causal logic is generated through evidence anchoring chain type reasoning; and finally, performing vision and text dual consistency verification on the thinking chain text, and outputting an analysis conclusion according to a result, or triggering a negative feedback correction mechanism to regenerate the thinking chain text. According to the method, medical reasoning full-process logic visualization is realized, medical illusion is eliminated, the diagnosis accuracy of complex cases is improved, and the method is suitable for medical scenes such as medical visual questions and answers and image diagnosis report generation.

The invention discloses a multi-mode Tokamak device fracture early warning method and system, and belongs to the field of Tokamak device safety protection. The method comprises the following steps: step 1, collecting multi-modal data, and marking fracture data in the collected multi-modal data; 2, time axis alignment is carried out on the collected multi-modal data, time slices are divided, and a fracture early warning database is formed; 3, training the fracture early warning model by using the fracture early warning database, and performing real-time deployment on the optimal fracture early warning model obtained by training; and step 4, in the Tokamak experiment process, inputting 0-dimensional diagnosis data and two-dimensional image diagnosis data which are acquired in real time into the optimal fracture early warning model, and performing fracture relief and device protection according to an output fracture early warning result. According to the method, the fracture state of the Tokamak device can be predicted more accurately, the fracture processing actuator is triggered when the Tokamak device is about to be fractured, and the device is better prevented from being fractured and damaged.

Systems and methods are provided herein for developing and deploying active-prompt-tuned, domain-specific image diagnosis and categorization applications. Processes of the present disclosure may control and provide for human-confirmable diagnostics from images, based on controlled instructions provided to vision-language models. The controlled instructions may be developed by systems provided herein, which generate system prompts and example prompt sets using active prompt tuning approaches.

According to the macular oedema diagnosis system, method and device based on color fundus photography, the medium and the program product, the bright part enhancement and dark part enhancement of the CFP image are combined, the problem that diagnosis is difficult due to the fact that macular oedema features are not obvious in traditional CFP image diagnosis is solved, and the diagnosis accuracy and efficiency are greatly improved. Experimental results show that the accuracy rate of the method can reach 95% or above on a private data set, and the accuracy rate of multi-stage diagnosis of macular edema on a public macular degeneration and macular edema combined data set IDRID can reach 85%. Besides, a Mama layer in an SSM (Spatial-Sense-Modulation) framework is adopted to perform global feature modeling, so that the capability of capturing the global relationship of the image is further improved, the bottleneck problem encountered in actual clinical application due to the fact that a Transform model depends on computing resources is reduced, the requirement for high-performance computing equipment is reduced, and the method has the advantages of being high in practicability and easy to popularize. And possibility is provided for application in areas with limited resources.

The invention relates to the technical field of medical image intelligent diagnosis, and provides an interpretable lung CT image intelligent diagnosis method and device, and the device comprises an image calling module, an image processing module, an image analysis module and a result display module. According to the method, identity information of a patient is read to automatically verify and authorize, and an electronic CT image is called and stored in a data storage card through an encryption transmission channel. And the image processing module performs denoising, enhancement and interpretability analysis on the original image to generate a thermodynamic diagram to assist in decision making. The image analysis module completes lesion detection and classification, verifies the consistency between a model decision and doctor experience through a thermodynamic diagram, and optimizes network parameters in combination with feedback data. And the result display module displays a diagnosis result and a thermodynamic diagram in real time, and supports multi-terminal cooperation and data archiving. The device supports data sharing and remote consultation, the efficiency and convenience of diagnosis are improved, and the wide application prospect of artificial intelligence in the field of medical diagnosis is shown.

The invention relates to the field of biomedicine, particularly provides a copper-based nano-enzyme as well as a preparation method and application thereof, and aims to solve the problems that in the prior art, clinical treatment on primary sclerosing cholangitis (PSC) is difficult in diagnosis and lacks of effective treatment drugs. The copper-based nano-enzyme comprises a nano-enzyme carrier and a copper-based nano-enzyme, wherein the nano-enzyme carrier is a two-dimensional nanosheet formed by copper ions, gallic acid and ursodesoxycholic acid through coordinate bonds; the probe molecule is a cyanine dye molecule connected to the surface of the nano-enzyme carrier through a covalent bond; wherein the fluorescence intensity of the copper-based nano enzyme is enhanced after the action of the alkaline phosphatase. The three functions of alkaline phosphatase responsive diagnosis, nano-enzyme catalytic treatment and ursodesoxycholic acid hepatic targeting are innovatively and synergistically integrated into one nano platform, the treatment function can be executed while specific imaging diagnosis is performed on diseases, and a new strategy is provided for diagnosis and treatment of diseases such as PSC.

The invention discloses a brain tumor diagnosis method and system based on multi-modal supervised contrast learning, and relates to the technical field of artificial intelligence, and the method comprises the steps: inputting an MRI image of a patient into a trained brain tumor diagnosis model, so as to output a brain tumor diagnosis result; the training process of the brain tumor diagnosis model is as follows: acquiring and preprocessing MRI images of a patient to construct an original MRI image set, and fusing MRI images of different modals of the same patient in the original MRI image set to obtain a fused MRI image set; combining the original MRI image set and the fused MRI image set to obtain a combined image set, and extracting image diagnosis features from the combined image set based on an image encoder; constructing cross entropy loss and multi-modal supervised comparison loss, and performing supervised learning on an image encoder to train a brain tumor diagnosis model; according to the brain tumor diagnosis method and system, high-precision brain tumor diagnosis is realized.

The invention discloses a near-infrared light nano probe as well as a preparation method and application thereof. The preparation method of the near-infrared nano probe comprises the following steps: dissolving polylactic acid-co-acetic acid in an acetonitrile solution, and adding lecithin and 1, 2-dipalmitoyl-sn-glycerol-3-phosphatidyl glycol to obtain a solution A; dissolving indocyanine green in an ethanol solution to obtain a solution B; and mixing the solution A and the solution B, and carrying out ultrasonic treatment to obtain the near-infrared nano probe. The near-infrared light nanoprobe can be used for imaging diagnosis of acute kidney injury, and effectively solves the problems of long diagnosis time, poor stability, high in-vivo clearance rate, low cell absorption rate and the like. Compared with existing ICG nano-particles, the ICG nano-particles are high in biological tissue penetrability, high in resolution ratio, low in phototoxicity and good in fluorescence imaging effect in an acute kidney injury model and have wide application prospects in AKI clinical diagnosis.

The invention provides a comprehensive solution for a medical ultrasonic image diagnosis task, which comprises two innovative parts, namely a CAPT parameter efficient fine tuning method based on a convolution adapter and embedded prompt tuning and a model lightweight method based on evolution pruning. Aiming at double challenges of scarcity of data and difficulty in model adaptation in the field of medical ultrasonic images, on one hand, the CAPT method introduces a convolution module into a traditional adapter to enhance the fine-grained feature extraction capability by fusing adapter tuning and prompt tuning mechanisms, and dynamically generates a prompt vector embedded with attention calculation by using adapter learning context information; the adaptability and the small sample generalization ability of the Transform model in an ultrasonic image classification task are obviously improved; on the other hand, aiming at a model edge deployment requirement, designing an evolution pruning algorithm, taking DenseNet-121 as a basic framework, inheriting parent strong neuron connection by adopting an asexual reproduction mechanism, and through a channel importance evaluation method for coupling a BN layer scaling factor and a weight L1 norm and in combination with a sparse evolution strategy guided by environmental constraint, constructing a model edge deployment model. And the classification precision is maintained, and meanwhile, the model parameters and the calculated amount are greatly compressed. The two technologies respectively break through the technical bottlenecks of medical image diagnosis from two dimensions of parameter efficient fine tuning and model structured pruning, the former improves the adaptability of the model field through feature extraction optimization, and the latter balances network lightweight and performance stability by means of an evolutionary algorithm. And an innovative technical path is provided for efficient deployment of an ultrasonic image auxiliary diagnosis system in a data limited scene and an edge computing environment.

The invention relates to a medical image report automatic generation method and system. The method comprises the following steps: extracting local image features and global image features related to a focus in a medical image; fusing the local image features and the global image features to obtain fused image features, and inputting the fused image features into an image diagnosis classifier to obtain a medical image related disease diagnosis classification result; wherein the image diagnosis classifier is obtained by pre-training an original classifier based on a multi-dimensional image fusion feature, and the multi-dimensional image fusion feature is obtained by fusing a local image feature and a global image feature which are extracted from a sample medical image and related to a focus; querying a pre-constructed medical field knowledge graph based on the disease diagnosis classification result to obtain a corresponding first medical field knowledge code; and fusing the local image features, the global image features and the first medical domain knowledge codes to obtain target text features, and inputting the target text features into a medical big language model to automatically generate an image diagnosis report.

The invention belongs to the technical field of medical diagnosis, and discloses an MRI lumbar sagittal view disease image diagnosis method based on a YOLOv8s algorithm, and the method comprises the following steps: S1, screening collected MRI lumbar sagittal view image data, self-building a VOC2007 data set, and marking the data set; s2, performing data enhancement and other processing on the existing data set; s3, establishing a YOLOv8s deep learning neural network detection model; dividing the data set into a training set, a verification set and a test set according to a proportion of 8: 1: 1, training the YOLOv8s network until convergence, and obtaining a lumbar disease image detection model based on the YOLOv8s network and a weight file; and S4, testing the randomly extracted MRI lumbar vertebra sagittal view image of the patient by using the trained YOLOv8s detection model, outputting a detection result, and evaluating the detection result. According to the invention, the diagnosis of the MRI lumbar sagittal view disease can be realized, the detection precision and efficiency are greatly improved, and the diagnosis accuracy is improved.

The invention belongs to the technical field of medical instruments, and particularly relates to test equipment for diagnosis and analysis of biliary tract disease images, which comprises a bottom plate, a frame body fixedly mounted at the top of the bottom plate, a support frame fixedly mounted in the frame body, a lens cone fixedly mounted on the support frame, a converter mounted on the lens cone, and an objective table fixedly mounted on the support frame, an adjusting assembly is arranged in the strip-shaped groove and connected with a supporting plate, a strip-shaped plate is fixedly installed on the supporting plate, and a scraping plate is fixedly installed on the strip-shaped plate. According to the testing equipment for biliary tract disease image diagnosis and analysis, the surface of the objective table can be conveniently cleaned through the arranged scraper, when the objective table is cleaned, a double-output-shaft motor works to drive a threaded lead screw to rotate, when the threaded lead screw rotates, a sliding block moves, synchronous movement is driven through a supporting plate, the scraper makes contact with the top of the objective table, and thus the objective table is cleaned. And when the scraping plate moves, impurities above the objective table can be removed, so that the objective table is cleaned.

An image diagnosis device according to the present disclosure can be connected to a diagnostic imaging catheter provided with an imaging unit including an ultrasonic transmission / reception unit, and can generate an intraluminal image. The image diagnosis device is provided with: a drive motor that can rotationally drive the imaging unit; a slip ring that can be electrically connected to the imaging unit and can rotate together with the imaging unit; and a brush that is electrically connected to the slip ring by contacting the slip ring. The slip ring and the contact part of the brush are disposed in a sealed space filled with a liquid lubricant.

The invention discloses a chest image diagnosis method and system based on multi-modal sign collection, and relates to the technical field of medical image.The method comprises the steps that a chest image of a patient is obtained, electrocardiosignals and blood oxygen saturation data are synchronously collected, and an associated radiology report is obtained; the image lesion features and the frequency domain rhythm template are combined for processing, motion artifacts are eliminated through a frequency domain decoupling equation, and refined image features are output; inputting the refined image features and the text pathological semantic features into a bidirectional attention mechanism to generate fusion features, and splicing the oxyhemoglobin saturation data and the text pathological semantic features into a sign-text vector; and inputting the fusion feature and the sign-text joint vector into a multi-task loss function, and outputting a structured diagnosis report. According to the method, accurate elimination of motion artifacts is achieved through a frequency domain decoupling equation, and coupling calculation is conducted on an electrocardio rhythm template and image lesion features in a frequency domain space.

The present invention relates to the field of image processing technology, and in particular to a method for image diagnosis and positioning of a supplementary motor area based on a resting state, comprising: acquiring a user's brain magnetic resonance image based on an MRI scanner, creating a database, and storing the acquired user's brain magnetic resonance image; acquiring the user's brain magnetic resonance image, and removing high-frequency physiological noise and low-frequency drift in the user's brain magnetic resonance image; acquiring the supplementary motor area, and performing low-frequency pulsation amplitude analysis, local consistency analysis, and functional connection abnormality analysis; the present invention utilizes brain maps for preliminary positioning and functional connection mutual verification, can more accurately determine the position of the supplementary motor area, reduce positioning errors, and evaluate the state of the supplementary motor area from two aspects of local brain activity and functional connection abnormality analysis, and based on this method, reduces the interference of subjective factors, realizes the quantification of diagnostic indicators, and makes the results more objective and reliable.

The invention discloses a few-sample epidemic disease image diagnosis algorithm based on comparative learning and a DS evidence theory. The algorithm comprises the following steps: S1, data set preparation: collecting chicken images of normal chickens without epidemic disease tags as a pre-training data set; collecting an epidemic disease image of an abnormal chicken with an epidemic disease tag as a training data set; epidemic disease labels corresponding to the epidemic disease images are label sets; s2, pre-training: performing comparative learning through a query encoder and a momentum encoder based on a MoCoV2 comparative learning framework to obtain a trained query encoder; s3, training: constructing a classification model, carrying out refined training through the training data set, and carrying out supervised learning through the label set to obtain a trained classification model; s4, diagnosis result fusion: inputting a plurality of to-be-diagnosed images into the trained classification model to obtain probability distribution of epidemic disease categories corresponding to the to-be-diagnosed images; and based on the DS evidence theory and the probability distribution, obtaining the prediction probability of the epidemic disease category of the to-be-diagnosed chicken.

The invention discloses a medical image privacy protection-oriented structure perception adaptive steganography method, which comprises the following steps of: in a training stage, generating a diagnosis related region mask based on a pre-trained IMIS-Net, extracting an embeddable suppression mask, and obtaining an initial embedding probability under the minimum distortion constraint; and the self-adaptive adjustment of structure perception is carried out on the pixel-level embedding probability in combination with a reinforcement learning normal form, so that the ROI is protected preferentially and a high-risk region is inhibited on the premise of meeting the set capacity, and the steganalysis resistance and perception quality are improved. And after the training is completed, in a reasoning stage, the masks and probabilities are firstly generated or quoted, then action probabilities are obtained through a trained strategy network, then probability consistency is mapped into STC cost, real embedding is completed by the STC, and a secret-containing medical image is obtained. According to the invention, on the premise that the availability of image diagnosis is not reduced, safe embedding and hidden transmission of sensitive information are realized.

The invention relates to the technical field of computer vision and industrial equipment fault diagnosis, in particular to a fault image diagnosis system for a remote controller of a bolter miner driven by a CMC chip. Comprising a data sensing module for synchronously acquiring a CMC chip internal logic state vector and a real-time observation image, and preprocessing to obtain target visual data; the generative reconstruction module maps the logic state vector to a high-dimensional space to generate an ideal feature map; the visual analysis module is used for extracting an observation feature map through a convolutional neural network, generating an interference distribution mask based on frequency domain texture statistics, and identifying an unstructured environmental noise region; the difference analysis module is used for calculating a semantic residual image by combining the ideal feature map, the observation feature map and the interference distribution mask, and generating an image difference analysis result; according to the method, ideal features are generated by using the internal logic state vector of the CMC, and the residual calculation of the coal dust noise area is shielded by the interference mask, so that the false alarm caused by the environmental noise is effectively eliminated, and the closed-loop verification that what you see is what you control is realized.

The invention provides a pneumonia CT image diagnosis model training method, diagnosis method and equipment, and the training method comprises the steps: inputting a 3D chest CT image into a multi-task deep learning model, enabling a shared encoder in the model to extract multi-scale feature data, and enabling a connection module and a decoder to obtain pneumonia focus region prediction result data according to the multi-scale feature data, the classification head obtains pneumonia diagnosis type prediction result data according to the multi-scale feature data; determining the joint loss of the model in the current iteration round and updating model parameters; and if the current multi-task deep learning model satisfies a training termination condition, outputting the current model as a pneumonia CT image diagnosis model. According to the method, the problems of low model feature utilization rate and low pneumonia diagnosis process efficiency caused by incapability of simultaneously completing focus segmentation and type classification due to task simplification of an existing pneumonia diagnosis model can be solved.