71 results about "Dose prediction" patented technology

The invention proposes a voxel-level dose distribution prediction method, and the method comprises the steps: obtaining a detection image of a target object when each tracer is independently introduced under the condition that a plurality of tracers are independently introduced into the target object; performing image standardization processing on the detection image corresponding to each tracer agent to obtain a standardized image corresponding to each tracer agent; and taking the standardized images corresponding to the various tracers as input information of a deep learning model, and determining a comprehensive three-dimensional dose prediction result of the target object under the various tracers through the deep learning model. According to the technical scheme, the accuracy of dose prediction in nuclide therapy is improved, full-link automation from image acquisition to dose planning is realized, the generated dose distribution not only meets the clinical requirement of precise coverage of a target region, but also conforms to the radiotherapy principle of normal tissue protection, and reliable technical support is provided for personalized precise radiotherapy.

The invention discloses a patient historical state-based anesthetic dose simulation evaluation system, which is characterized by comprising a multi-source data integration and preoperative physiological feature quantification layer, a dose effect relationship simulation layer and a risk evaluation and recommendation output layer. The method has the following advantages and effects that the personalized pharmacokinetics-pharmacodynamics model is constructed by integrating multi-source historical data of a patient, vital sign change tracks under different anesthetic doses are simulated, the recommended dose interval is generated based on multi-objective optimization, the change from experience dependence to accurate quantification is realized, and the accuracy of the recommended dose interval is improved. The problem of inaccurate anesthetic dosage prediction caused by individual differences is effectively solved, and the safety and effectiveness of an anesthesia scheme are remarkably improved.

The invention discloses a machine learning-driven water body bacteria ultraviolet disinfection dosage prediction and optimization method, which comprises the following steps of: obtaining genetic information, environmental parameters and ultraviolet disinfection conditions of target bacteria, and numeralizing the genetic information into feature vectors; cleaning and coding the acquired data to form a feature matrix; based on the characteristic matrix, a machine learning algorithm is utilized to train an ultraviolet disinfection dosage prediction model, and the model takes the inactivation efficiency as a prediction target; according to bacterial genetic information, environmental parameters and ultraviolet conditions input by a user, utilizing the trained prediction model to predict the inactivation efficiency, scanning in a multi-dimensional parameter space, and determining an optimal ultraviolet disinfection condition combination meeting a target inactivation level; outputting an optimal ultraviolet disinfection condition combination; the method disclosed by the invention has cross-strain and cross-scene applicability.

The invention discloses a graded blasting construction method for a shallow-buried soft rock tunnel sensitive section. The method comprises the steps that measuring points are arranged in a tunnel chamber, a ground surface non-cavity section and a cavity section, and blasting vibration data are obtained; peak vibration velocity prediction models are established in a partitioned mode, an attenuation model is adopted in a chamber, a correction term is introduced into a non-cavity section of the earth surface to correct the attenuation model, and a multivariable model is established for the cavity section through dimensional analysis by integrating the horizontal distance, the elevation difference, the explosive quantity, the rock mass density and the tensile strength; determining a ground surface vibration velocity safety threshold, and quantifying a cavity vibration velocity amplification coefficient and an influence range thereof; based on the peak vibration velocity probability distribution and a preset safety level, establishing a safe dose prediction model and a safe range model; and dividing control areas according to the model, and implementing graded construction. According to the method, the vibration propagation law can be accurately described, the cavity amplification effect can be quantified, dynamic prediction of the blasting safety explosive quantity and range is achieved, and disturbance of blasting vibration to surrounding rock is effectively controlled through hierarchical management and control.

A dose prediction method and system, the method comprising: acquiring an original dose distribution of a radiotherapy object, original sign information and current sign information; obtaining a predicted Pareto optimization parameter through a dose prediction model according to the original dose distribution and the original sign information; and predicting a current dose distribution corresponding to the current sign information by using the dose prediction model according to the current sign information and the predicted Pareto optimization parameter.

The invention provides a flexible solar cell irradiation attenuation coefficient estimation method comprising the following steps: orbital parameters are combed, and spacecraft orbital parameters and space environment parameters are obtained; based on the orbit parameters and the space environment parameters, establishing an irradiation dose calculation model of the flexible solar cell; performing a ground equivalent test based on the irradiation dose calculated by the irradiation dose calculation model; and determining the irradiation attenuation coefficient of the solar cell in combination with test data and historical data. According to the method for estimating the irradiation attenuation coefficient of the flexible solar cell, the irradiation dose calculation model suitable for the flexible solar cell is constructed, the on-orbit performance attenuation coefficient of the solar cell array can be accurately obtained, and the accuracy of irradiation dose estimation is improved; scientific basis is provided for solar cell array design and on-orbit energy management, and reliability and design rationality of a spacecraft energy system are effectively improved.

The invention discloses a proton treatment plan verification and optimization device and method and related equipment, and relates to the technical field of radiotherapy. The device comprises an image acquisition module which is used for acquiring a PET image of a patient after completing at least one proton treatment irradiation; the dose prediction module is used for mapping the treated PET image into actual deposited dose distribution in the body of the patient based on the trained generative model so as to predict actual distribution dose distribution; and the adjustment and optimization module is used for generating an adjusted fractional treatment plan when the deviation between the predicted actual distribution dose distribution and the initial planned dose distribution is greater than a deviation threshold value. Based on in-vivo measurement information, the actual deposition dose is accurately verified and dynamically optimized, and accurate and safe proton therapy is achieved.

The application provides a kind of based on deep learning's absorbed dose prediction method, device, equipment and medium, it is related to medical image processing technical field, comprising: obtaining the treatment of patient's pre-test multi-modal data;According to clinical examination data and the mask map of the target tumor region and normal key organ region of pre-set drug, space coding is carried out, and the biomarker feature map of the patient to be tested is constructed;According to the pre-treatment medical image and biomarker feature map, a preset first deep learning model is used for prediction, to generate the simulated treatment initial medical image of the patient to be tested;According to the simulated treatment initial medical image, a preset second deep learning model is used for prediction, to generate the post-treatment predicted dosimetry parameter map of the patient to be tested.The application can improve the accuracy of nuclide absorbed dose prediction evaluation.

This invention relates to the field of medical data-driven dose prediction technology, and discloses a method, system, device, and storage medium for individualized dose prediction of hypothyroidism based on medical data. The method includes: constructing a standardized feature vector based on the child's weight, age in days, corrected age in months, current L-T4 dose, TSH value, FT4 value, previous TSH value, TSH rate of change, feeding method, month of consultation, etiology of hypothyroidism, comorbidity status, previous dose adjustment magnitude, and age at which TSH first reached target levels; extracting TSH dynamic trajectory features from the child's TSH time-series data from previous follow-ups; obtaining a basic recommended dose using a gradient boosting decision tree model constructed with counterfactual filtering training data; and correcting the basic recommended dose to obtain an individualized recommended dose. This method improves the prediction accuracy of the gradient boosting decision tree model and allows the individualized recommended dose to simultaneously take into account multiple clinical confounding factors.

The embodiment of the present application relates to the technical field of imaging detection, in particular to a component irradiation dose prediction method, an imaging system and a storage medium, the prediction method is applied to the imaging system, and the imaging system comprises a ray source and a detector. Through obtaining the actually measured dose of the concerned component under the irradiation of the ray source, fitting the irradiation intensity distribution of the ray source and the actually measured dose, obtaining the ray intensity dose distribution, mapping the path of the relative position change of the concerned component to the path of the height layer of the detector based on the ray intensity dose distribution, obtaining the irradiation dose of the concerned component under the standard height and the standard current, and finally correcting the irradiation dose according to the actual situation of the concerned component, the predicted irradiation dose of the concerned component is obtained, in this way, the irradiation dose of the concerned component can be accurately predicted by combining the ray intensity distribution of the ray source and the actual situation of the concerned component, the accuracy of the irradiation dose prediction is improved, and the irradiation dose error is reduced.

The invention discloses a radiotherapy plan automatic optimization design method based on machine learning, and relates to the technical field of radiotherapy, and the method comprises the steps: firstly collecting and preprocessing radiotherapy core data to generate a training set, training feature extraction, dose prediction and optimization adjustment of a model; processing to-be-processed data through a feature extraction model, screening three optimal similar cases in combination with cosine similarity, and extracting effective features; based on the feature vectors and the effective features, outputting an initial dose plan, constraint parameters and a deposition matrix by a dose prediction model; and finally, constructing a target function, and iteratively optimizing the radiation field parameters and the photon flux by adopting an improved ant colony algorithm and a conjugate gradient method. According to the method, clinical experience and individual differences are fused, the radiotherapy plan is efficiently and accurately generated, and the quality stability and clinical suitability of the plan are improved.

The present invention relates to a system for predicting and managing radiation dose and, more specifically, to a system for predicting and managing radiation dose, wherein a simulation can be implemented in advance by modeling a work site and a worker and reflecting the work site and the worker in a scenario so that the radiation dose of the worker during radiation work in a nuclear power plant can be predicted before the work is begun.

This invention discloses an intelligent adaptive dose verification calculation method and system. The intelligent adaptive dose verification calculation method includes: obtaining preliminary dose distribution data for a patient based on a deep learning-based dose prediction model; dynamically assigning variance reduction parameters to particles in a Monte Carlo simulation based on the high and low dose distribution information in the preliminary dose distribution data, wherein the variance reduction parameters include simulation weights, with simulation weights assigned to particles predicted to be in high-dose regions being less than those assigned to particles predicted to be in low-dose regions; and using a Monte Carlo dose calculation engine, simulating the planned beam based on the dynamically assigned variance reduction parameters to obtain high-precision verification dose distribution data. This invention's intelligent adaptive dose verification calculation method ensures the physical authenticity of the final verification dose distribution and the credibility of the "gold standard."

The invention provides a dose prediction model training method and device, equipment and a storage medium, and belongs to the technical field of artificial intelligence. According to the method, the dose prediction model for determining the dose distribution based on the target imaging image is trained, so that when the dose distribution is determined, the dose distribution can be rapidly determined through the dose prediction model, multiple iterative calculations are not needed, the time consumption is obviously shortened, and the efficiency is improved. Moreover, training is performed based on a gradient of a model loss function with respect to a model parameter, and the model loss function is not only related to a difference between a predicted dose distribution and a target dose distribution, but also related to a difference between an imaging image predicted based on the predicted dose distribution and a target imaging image. In this way, after the dose prediction model is obtained through training, photoetching is carried out through the determined dose distribution, the quality of an imaging image can be improved, that is, the dose distribution is determined through the dose prediction model, efficiency is high, and accuracy is high.

The application discloses a three-dimensional radiotherapy dose prediction method based on distance prior and local ranking loss. In view of the target boundary leakage and optimization instability problems existing in the traditional model, the method adopts a SwinUNETR network, and a continuous spatial distance map is injected at the input end to assist the model in establishing the physical law of radiation attenuation. At the same time, a boundary-aware local ranking loss is proposed, which accurately applies a one-way topological penalty to the 3mm high-risk area at the junction of the target area and the organ at risk, truncates the dose leakage and promotes the energy to gather in the target area. Experiments prove that the method makes the core target area coverage index D95 jump to 42.798Gy, and the irradiated volume V50 of the organ at risk is reduced to below 0.005%, realizing high-precision and high-safety radiotherapy dose distribution prediction.

Systems and methods for determining multi-field doses for radiation therapy are provided. 1) one or more medical images of a patient and 2) a fluence map for each of a plurality of fields for radiation therapy of the patient are received. Fluence-related information is determined for each of the plurality of fields based on the fluence map. Fluence-related information for the plurality of fields is aggregated. A multi-field dose of the patient is determined based on the one or more medical images and the aggregated fluence-related information using a machine learning-based dose prediction network. And outputting the multi-field dose.

The application discloses a head and neck cancer radiotherapy dose prediction method based on cross perception fusion, comprising the following steps: (1) establishing a training data set containing a patient CT image, a planning target volume, an organ at risk and a real dose distribution; (2) constructing a generative adversarial network based on cross perception fusion, introducing a cross perception fusion module in the generator, realizing the deep interaction and fusion of local and global features, and introducing a multi-scale cross window self-attention coding network to constrain the difference between the generated and real doses at multiple scales; (3) constructing a hybrid loss function based on the discriminator and the multi-scale cross window self-attention coding network; (4) training the generative adversarial network by using the training data set; and (5) testing the data to be tested by using the trained network, and the output of the generator is the dose prediction result. The application can realize the direct generation of the head and neck cancer radiotherapy dose distribution, assist doctors in formulating accurate radiotherapy plans, and improve the radiotherapy effect.

The invention provides a pelvic tumor multi-center, multi-tumor-species and multi-prescription dose prediction method and related equipment, and is applied to the technical field of data processing. The method comprises the following steps: preprocessing a training sample set to generate a training sample set with target feature data; processing a preset dose distribution processing model based on a SwinTransform network architecture based on the training sample set with the target feature data, and generating a target dose distribution processing model; performing feature extraction processing on the CT image information of the target user to generate a target feature vector; processing the physiological state information of the target user to generate a dose limiting factor of the target user; processing the physiological parameter information of the target user to generate a dynamic dose influence factor of the target user; and processing the target feature vector, the dose limiting factor of the target user and the dynamic dose influence factor of the target user based on a target dose distribution processing model to generate target dose distribution information.

The application relates to the technical field of medical image processing, and discloses a nasopharyngeal carcinoma radiotherapy dose prediction system and method of a cascade network, which solves the technical problem that nasopharyngeal carcinoma radiotherapy dose prediction lacks global-local structure collaborative modeling and adaptive cross-stage feature interaction in the prior art. A first-stage three-dimensional U-Net network captures anatomical global features, a second-stage improved MedNeXt network performs dose fine prediction through residual learning, explicit division avoids the multi-scale feature extraction limitation of a single network, and the overall dose score and the DVH score are respectively improved; through instance normalization, the first-stage feature activation distribution is stabilized, after being spliced with original input, the convolution adaptive layer is seamlessly integrated into the second stage, the dimension mismatch and the gradient saturation problem are solved, and flexible multi-stage feature reuse is supported.

The application provides a lung cancer radiotherapy plan generation method and system based on dose prediction and auxiliary contours, and belongs to the technical field of medical devices. Lung cancer radiotherapy plan case data is acquired. A trained dose prediction model is used to process the radiotherapy plan case data to obtain dose distribution results of a target region and an organ at risk. Anisotropic auxiliary contours are generated according to the predicted dose distribution and combined with dose constraints required by reverse optimization. According to the auxiliary contours and the dose distribution, dose limits of the target region, the organ at risk and the auxiliary contours are obtained. Reverse optimization is performed according to the auxiliary contours and the dose limit values to generate a lung cancer radiotherapy plan. The minimum distance distribution from patient voxels to the target region is included in model training, which improves the dose prediction accuracy. The auxiliary contours are generated using the dose distribution, the optimization conditions / targets are more reasonable, and the design time is shortened. It is beneficial to the protection of the organ at risk lung and the rapid improvement of the plan design level of units lacking of radiotherapy experience.

This invention discloses an X-ray thermoacoustic imaging reconstruction method based on Fourier neural operators, belonging to the field of X-ray imaging technology. This invention rapidly constructs a multimodal X-ray thermoacoustic dataset covering different sampling conditions; fully utilizes the frequency domain global feature extraction advantages of Fourier operators to effectively improve the accuracy of forward physics model simulation and dose prediction; the dual-head Fourier structure effectively enhances the stepwise correction effect of absorption coefficient and dose distribution, making the reconstructed image clearer and more quantitatively accurate; lightweight training and recursive parameter updates reduce computational resource consumption, enabling the system to achieve real-time or near-real-time imaging capabilities, fully realizing a closed-loop system from initial data simulation, network reconstruction to dose distribution verification, and exhibiting high clinical adaptability.

The invention provides a lung cancer IMRT multi-prescription dosage prediction method and device based on a beam path distance map, and is applied to the technical field of data processing. The method is developed around lung cancer IMRT multi-prescription dosage prediction, CT images, organ outlines and other data of a plurality of patients are collected firstly, and data set division, resampling, normalization and data enhancement preprocessing are carried out; a signed distance map is generated based on the beam path, a cascade model CasU-Net-BDM containing GD-Net and RD-Net is constructed, and training is completed by taking MAE as a loss function and adopting an Adam optimizer and the like; evaluating the performance of the model through indexes such as voxel MAE, HI and CI and a visual means; and finally, inputting patient related data, outputting an accurate multi-prescription dose distribution prediction result by the trained model, and being suitable for various IMRT and SIB prescription scenes.

The invention relates to the technical field of medical image processing, discloses a cascaded network nasopharyngeal carcinoma radiotherapy dose prediction system and method, and solves the technical problem that nasopharyngeal carcinoma radiotherapy dose prediction in the prior art lacks collaborative modeling and self-adaptive cross-stage feature interaction for a global-local structure. According to the method, in the first stage, the three-dimensional U-Net network captures anatomical global features, in the second stage, the improved MedNeXt network carries out dosage refined prediction through residual learning, explicit division avoids the limitation of multi-scale feature extraction of a single network, and the overall dosage score and the DVH score are improved respectively; first-stage feature activation distribution is stabilized through instance normalization, and is seamlessly integrated to a second stage through a convolution adaptation layer after being spliced with original input, so that the problems of dimension mismatching and gradient saturation are solved, and flexible multi-stage feature multiplexing is supported.

The invention relates to the technical field of radiotherapy optimization, and discloses an interactive radiotherapy planning system optimization method. The method comprises the following steps: acquiring medical image data and real-time body surface movement monitoring data of a patient, constructing a three-dimensional dose flux distribution diagram according to the medical image data, and generating an anatomical displacement feature set according to the real-time body surface movement monitoring data; inputting the three-dimensional dose flux distribution diagram and the anatomical displacement feature set into a pre-constructed multi-modal dose prediction model to obtain a first prediction result of dose distribution deviation; extracting radiation biological characteristic parameters according to the three-dimensional dose flux distribution diagram, and matching the radiation biological characteristic parameters with a preset radiation damage threshold model to obtain a second prediction result of dose safety assessment; and generating a dynamic optimization weight based on the first prediction result and the second prediction result, and adjusting treatment plan parameters according to the dynamic optimization weight to realize radiotherapy optimization.

The invention discloses a single pencil beam dose simulation method for an LSTM model, and the method comprises the steps: collecting a dynamic image slice through a high-frequency imaging device, marking a timestamp, and synchronously generating a time sequence dose data set related to an organization form; fusing image deformation and dose characteristics by using a double-flow LSTM network, predicting high-precision dose distribution and outputting a deformation compensation factor; dynamically adjusting a dose threshold and a voxel weight according to the tissue deformation degree, and strengthening the prediction precision of a deformation sensitive area; and combining real-time image feedback to optimize pencil beam parameters, intercepting image slices in real time, and using a dynamic LSTM model to predict high-precision dose distribution online. The predicted dose is corrected in real time through the deformation compensation factor and compared with the actual monitoring dose, and an error is fed back to the model; and finally, the pencil beam dose distribution is superposed to generate dynamic Monte Carlo dose distribution, so that the dose prediction real-time performance and accuracy are improved, and the radiotherapy treatment safety and effectiveness are enhanced.

This invention discloses a method, device, and equipment for evaluating dose limits for organs at risk in tumor radiotherapy planning based on spatial distance analysis, relating to the field of tumor radiotherapy technology. Specifically, based solely on geometric statistics, before dose calculation, a three-dimensional Euclidean distance transformation is used to obtain the shortest distance distribution from all voxels of the organs at risk to the target area after anisotropic normalization, directly forming an individualized dose limit and weight template. The output of this invention is a distance-voxel histogram and its derived cumulative proportions and rates of change, from which several geometric indicators are extracted. Combined with a preset set of geometric rules, individualized suggestions for dose limit thresholds and optimization weights corresponding to different doses are directly obtained. Compared with existing technologies, this invention does not perform dose prediction, nor does it require generating auxiliary contours or performing forward / backward dose calculations. Instead, it provides individualized and executable dose limit and weight suggestions before dose calculation, significantly reducing optimization iterations and improving consistency, with highly interpretable results.

The application discloses a tumor radiotherapy dose prediction method based on a generative adversarial network, comprising the following steps: (1) establishing a training data set comprising a patient CT image, a planning target volume, an organ at risk and a real dose distribution; (2) constructing a generative adversarial network based on an adaptive weight loss, wherein a CNN-Transformer hybrid structure is adopted to establish a generator, and an adaptive weight adjustment strategy is introduced into a discriminator to dynamically adjust the loss weights of real data and generated data; (4) training the generative adversarial network by using the training data set; (5) testing the data to be tested by using the trained network, and the output of the generator is the dose prediction result. The application can effectively solve the balance problem of the generator and the discriminator in the adversarial training, improve the performance and stability of the GAN model, and improve the precision of the tumor radiotherapy dose prediction.

The control circuitry receives a radiotherapy planning system request for the image data to support a particular radiotherapy planning step (typically for a particular corresponding patient). The control circuit then accesses specific image data that is particularly suitable for supporting a specific radiotherapy planning step and transmits the specific image data in response to a radiotherapy planning system request. The illustrative examples of the specific radiotherapy planning steps include but are not limited to a contour drawing step, a segmentation step, a dose prediction step, a dose calculation step and the like. By one approach and as an illustrative example, the foregoing particular image data may include patient image information including a segmented structure.

The invention relates to the crossing field of anesthetic medicine and artificial intelligence, and discloses a remazolam anesthetic dosage AI prediction system, which comprises a data acquisition module, a data processing module, a data processing module and a data processing module, and is characterized in that the data acquisition module is used for acquiring multi-dimensional clinical data of a target patient; the feature engineering module is used for generating a standardized feature data set based on the multi-dimensional clinical data; the improved StochDiff time sequence prediction module is used for outputting an initial individualized drug delivery scheme of the remazolam and prediction distribution of vital signs and sedation depth of a patient after drug delivery through a pre-trained improved StochDiff time sequence prediction model based on the standardized feature data set; the dose dynamic correction module is used for carrying out online dynamic correction on the initial individualized administration scheme based on patient data collected in real time in an operation and updating administration parameters in real time; and the safety early warning module is used for realizing pre-grading early warning of adverse anesthesia events based on a model prediction result and real-time data in an operation. Therefore, individualized precise administration and anesthesia safety guarantee are realized.

Systems and methods for predicting radiation dose distribution and planning a radiotherapy treatment are disclosed. An exemplary system includes a memory to store a computational model (such as a trained machine learning model), a dose prediction engine to predict a dose profile, and a treatment planning circuit. The dose prediction engine executes a dose simulation to determine a preliminary dose profile at a first statistical uncertainty level, applies the determined preliminary dose profile to the computational model to predict a refined dose profile at a second statistical uncertainty level lower than the first statistical uncertainty level. Based at least in part on refined dose profile, the treatment planning system can generate or update a radiotherapy treatment plan for use in a radiation treatment session.