51 results about "Dose prediction" patented technology

The invention discloses a predication method for three-dimensional dose distribution in intensity modulated radiation therapy plans. The method includes steps of (1) collecting valid intensity modulated radiation therapy plan data and forming a case database; (2) dividing a PTV and different to-be-endangered organs of patients into a plurality of voxels according to the resolution ratio of CT images; (3) extracting anatomical characteristics of each patient in the database; (4) extracting dose characteristics of each patient in the database; (5) constructing an artificial neural network, inputting the anatomical characteristics and the dose characteristics of the patients, and learning the mapping relation between the anatomical characteristics and the dose characteristics by the aid of the artificial neural network, and obtaining a correlation model of the anatomical characteristics and the dose characteristics; (6) using the correlation model to predicate the three-dimensional dose distribution of a new patient. The application of said method is using the dose distribution predication method for dose prediction for to-be-endangered organs of patients and quality control is achieved. By adopting the above method, predication of three-dimensional dose distribution in intensity modulated radiation therapy plans can be realized and the method can be applied to a quality control link.

A dose prediction model can be determined for generating a dose distribution of a treatment plan for irradiating a target structure within a patient. Treatment plans from previous patients can be analyzed to determine D characteristic values to obtain a D dimensional point for each treatment plan. The treatment plans can be clustered based on the D dimensional points. The treatment plans of a cluster can then be used to determine a dose prediction model. A dose prediction model for patient can be selected from among multiple models. Characteristics about the patient can be used to determine a D dimensional point corresponding to the patient. The D-dimensional point can be used to select a model in comparison to D dimensional points of the models.

This invention provides an X-ray CT capable of presenting information of exposure to the operator by displaying X-ray dose information of each of regions of interest to be scanned by an X-ray CT apparatus, thereby encouraging reduction in exposure and optimization. X-ray dose information of each region to be scanned by a conventional scan (axial scan), a cine scan, a helical scan, or a variable-pitch helical scan of an X-ray CT apparatus is displayed so that the operator can recognize the X-ray dose information before acquisition of an image of a subject. The X-ray dose information can be predicated with higher precision and displayed by using a dose prediction value obtained by an interpolation value and an extrapolation value of the first or higher order on at least three or more kinds of phantom measurement values, not a simple prediction value such as a zero-th order interpolation value or a zero-th order extrapolation value obtained by using measurement values of two kinds of phantoms like in the present CTDI display.

The invention discloses a three-dimensional dose prediction method based on deep learning and prior plan. The method includes obtaining comprehensive treatment-related information such as medical modal image, three-dimensional dose distribution information, target region structure images and normal organ structure images by collecting and processing data of existing high-quality radiation therapyplan, and inputting the above information into a pre-built neural network model for repeated training and optimization. The neural network model can fully learn the dose distribution of a high-qualityprior plan, and the resulting neural network model can more accurately predict the three-dimensional dose. Thus, the efficiency and effectiveness of the radiotherapy plan output by the model can be greatly improved, and a good generalization ability is gained.

The present invention proposes a method for automatically creating a dose prediction model based on existing clinical knowledge that is accumulated from multiple sources without collaborators establishing communication links between each other. According to embodiments of the claimed subject matter, clinics can collaborate in creating a dose prediction model by submitting their treatment plans into a remote computer system (such as a cloud-based system) which aggregates information from various collaborators and produces a model that captures clinical information from all submitted treatment plans. According to further embodiments, the method may contain a step where all patient data submitted by a clinic is made anonymous or the relevant parameters are extracted and condensed prior to submitting them over the communications link in order to comply with local regulations.

The invention belongs to the field of radiotherapy plan evaluation and discloses a machine learning-based radiotherapy plan evaluation method. With the method adopted, the problems of the non-uniformity of standards and susceptibility to subjective deviation of a conventional technology which adopts manual operation to evaluate radiotherapy plans. The method includes the following steps that: pre-processing work such as voxel selection, voxel feature extraction and voxel data annotation, is performed on the DICOM (Digital Imaging and Communication) data of high-quality radiotherapy historicalplans; voxel data are used to train a machine learning model; the model outputs predicted dose values for each voxel in a plan to be evaluated; a two-dimensional DVH prediction curve and a three-dimensional voxel dose prediction distribution map are further generated for each crisis organ; and finally physicists actually evaluate the radiotherapy plan to be evaluated with the reference of the curves and distribution maps. The present invention also discloses a corresponding evaluation system suitable for the objective and accurate evaluation of a radiotherapy plan.

The invention provides a three-dimensional dose prediction method for radiotherapy. The method includes the following steps of A, collecting radiotherapy planning data of past cases, and preprocessingthe data to obtain a feature image capable of being trained by a convolutional neural network; B, calculating the minimum distance from each voxel in a planning region to a target region boundary toobtain a target distance map; C, establishing a three-dimensional dose prediction network based on the convolutional neural network; D, training the three-dimensional dose prediction network through the feature image in step A to obtain an optimal three-dimensional dose prediction model through cross validation; E, inputting data of a patient to be irradiated into the three-dimensional dose prediction model to obtain a three-dimensional dose distribution map. The invention also provides a prediction system based on the method. The three-dimensional dose prediction method and system for radiotherapy have the advantages that by introducing deep learning of the convolutional neural network, the dependence on the personal experience of physicists is reduced, and the error of manual predictionis reduced.

The invention provides a warfarin individual anticoagulant pharmacogenomics detection kit suitable for Chinese population, which mainly comprises CYP2C9, VKORC1 and CYP4F2 related gene-type amplification primers, CYP2C9, VKORC1 and CYP4F2 related gene-type sequencing primers, a polymerase chain reaction (PCR) reagent and a sequencing reagent. The warfarin individual anticoagulant pharmacogenomicsdetection kit suitable for the Chinese population also can comprise a specification used for explaining a warfarin pharmacogenomics dose prediction model suitable for the Chinese population or / and a computer readable storage medium recording the warfarin pharmacogenomics dose prediction model suitable for the Chinese population. The warfarin pharmacogenomics detection kit provided by the invention is simple in preparation and is convenient in use. By adopting the kit, the warfarin anticoagulant therapy dose of Chinese patient population can be accurately estimated through detecting warfarin pharmacogenomics indexes, integrating clinical environmental factors and utilizing warfarin pharmacogenomics dose prediction model software suitable for the Chinese population, which is established by the invention.

The invention relates to the use of fetal tissues to generate a tissue model in a non-human animal. The tissue model comprises target tissues allowed to progress through development in vivo in a non-human host in order to obtain tissues having a mature phenotype that can be used to assess toxicity and / or efficacy of an agent.

The invention provides a drug dose prediction method and device, electronic equipment and a storage medium, is applied to the technical field of data processing, and can determine more accurate drug use doses for different patients. The method comprises the following steps: acquiring clinical original data of a target patient, wherein the clinical original data comprises demographic information, therapeutic drug use information, drug combination information, adjuvant therapy means, gene polymorphism and inspection information; performing data preprocessing on the clinical original data to obtain target feature data, the data preprocessing including at least one of data cleaning, data normalized coding and data screening; and inputting the target feature data into a drug dose prediction model to obtain the drug dose of the target patient in unit time.

The invention discloses an artificial intelligence guided dose prediction method and system, and the method comprises the steps: obtaining a medical image, stored in a preset format, of a patient; drawing the medical image to obtain a geometric anatomical structure; determining a prescription according to disease type information corresponding to the medical image, the geometric anatomical structure and a preset disease type-prescription template library; determining a radiotherapy irradiation angle according to the disease information, the geometric anatomical structure and the prescription;and inputting the disease information, the geometric anatomical structure, the prescription and the radiotherapy irradiation angle into a trained dose prediction model to obtain a radiotherapy dose result. By means of the technical scheme, full-automatic dose prediction is achieved, and the dose prediction efficiency and effect are improved.

The invention provides an intensity modulated radiation therapy plan three-dimensional dose distribution prediction method based on deep learning. The intensity modulated radiation therapy plan three-dimensional dose distribution prediction method comprises the following steps: collecting effective intensity modulated radiation therapy plan data to form a case database; extracting three-dimensional anatomical structure contour features of the region of interest of each patient from the case database; dividing the three-dimensional anatomical structure contour of the region of interest of the patient into a plurality of two-dimensional contour slice graphs; extracting dose features of each patient from the case database, and dividing the dose features into a plurality of two-dimensional dose slice distribution graphs; establishing a deep convolutional network, inputting the two-dimensional contour slice graph and the corresponding two-dimensional dose slice distribution graph of the patient, and obtaining an association model between anatomical structure contour features and dose features through model training; and predicting the three-dimensional dose distribution of the new patient by using the trained association model. By means of the intensity modulated radiation therapy plan three-dimensional dose distribution prediction method, the incidence relation between the anatomical structure features and the dose features can be effectively obtained, and the accuracy of dose prediction is improved.

The invention provides a plan implementation method and apparatus based on predicted dose guidance and Gaussian process optimization, and relates to the field of radiotherapy technology. The method comprises: calculating a predicted dose of a case by using a dose prediction model; calculating a plan score of the predicted dose according to the scoring rule as an optimal plan score; determining a plurality of sets of plan parameters based on the a priori database according to organ anatomy information; calculating corresponding plan scores of the plurality of sets of plan parameters, and constituting a Gaussian data set; and based on the Gaussian data set, calculating new plan parameters by using the Gaussian process, calculating the corresponding plan scores, adding the calculated scores to the Gauss data set, iteratively executing the step, and finally calculating the intensity-modulated optimization results under the plan parameters corresponding to the highest score in the Gauss data set. By the use of the predictive model to predict the dose distribution of the case to optimize the guidance, the quality of the plan is guaranteed, and by the use of the Gaussian process to calculate the posterior distribution based on the a priori data, the number of trial and error is reduced, thereby speeding up the optimization.

The invention provides a rectal cancer radiotherapy plan automatic design method based on deep learning. The method comprises the steps: establishing a U-Net neural network used for deep learning, andestablishing a case database, wherein the case database is clinical intensity modulated radiotherapy plan data of rectal cancer; performing deep learning on the case database to train the U-Net neural network; transmitting a img file of the CT positioning data to the trained U-Net neural network to obtain a prediction target area and prediction dose distribution output by the U-Net neural network; obtaining a dose objective function according to the prediction target area and the prediction dose distribution; and designing a radiotherapy plan by using a Pinnacle planning system according to the prediction target area and the dose objective function. Therefore, the rectal cancer radiotherapy plan automatic design method based on deep learning integrates the target area sketching technology, the dose prediction technology and the automatic planning technology, and the full-automatic design process of the individualized radiotherapy plan is realized in combination with the Pinnacle plansystem.

The invention discloses a warfarin dose prediction method and prediction device, and the method comprises the steps of: extracting a sample set from a database, carrying out the normalization of the sample set, and dividing the normalized sample set into a plurality of groups; sequentially selecting one group as a first verification set and the rest groups as a training set, constructing a singlemodel according to the training set and a preset algorithm, verifying the single model through the currently selected first verification set to obtain an error of the single model, and, according to the error corresponding to the same preset algorithm, calculating the precision corresponding to the preset algorithm; selecting the algorithm with the precision meeting the requirement as the optimalalgorithm; adjusting the parameters of the optimal algorithm through grid search; training according to the sample set and the optimal algorithm obtained after parameter adjustment to obtain a warfarin dose prediction model; and acquiring user information, and inputting the user information into the warfarin dose prediction model to obtain a warfarin dose corresponding to the user information. Byadopting the method, the prediction accuracy can be improved.

The invention discloses a three-dimensional dose prediction method and system in a personalized precise radiotherapy plan. The method comprises the following steps of: 1, acquiring radiotherapy related information such as an electronic computed tomography image, a dangerous organ structure mask image and a three-dimensional dose distribution image; 2, performing data preprocessing operation on the images in the step 1; 3, inputting the acquired image data into a two-stage generator network to generate a three-dimensional dose distribution image and a confidence map; 4, confronting the three-dimensional dose distribution prediction image and the three-dimensional dose distribution real image by adopting a Markov discriminator; 5, jointly optimizing a prediction model through a reconstruction loss function, a reconstruction loss function with the confidence coefficient weight and a confrontation loss function; and 6, generating three-dimensional dose distribution by using the trained prediction model. Through the three-dimensional dose prediction method and system in the personalized precise radiotherapy plan provided by the technical schemes of the invention, manual intervention in the radiotherapy plan can be reduced, the dose prediction precision is improved, and personalized precise radiotherapy is realized.

The invention relates to a nasopharyngeal carcinoma three-dimensional dose distribution prediction method based on segmentation task assistance. The nasopharyngeal carcinoma three-dimensional dose distribution prediction method specifically comprises the steps of collecting an original nasopharyngeal carcinoma image and marking the original nasopharyngeal carcinoma image; constructing a dose distribution prediction model, wherein the prediction model comprises an auxiliary segmentation network, a dose prediction network and an adversarial network, the segmentation network and the prediction network share encoder network parameters, shared representation information between a segmentation task and a dose prediction task are obtained through joint training of the segmentation task and the dose prediction task, the feature expression capability of a shared encoder is enhanced, and the network is promoted to mine essential features with an auxiliary function for dose prediction in the segmentation task to the maximum extent under limited training data. Meanwhile, in order to effectively utilize feature information of a prediction decoder under different scales, a multi-scale iterativefusion IMF strategy is provided at a prediction task decoder end, and a more accurate prediction result is obtained.

The invention relates to a multi-view multi-scale lymph node false positive inhibition modeling method. The invention discloses a radiotherapy automatic plan design system and a construction method thereof, and relates to the field of radiotherapy plan systems, and the system comprises a plan design auxiliary contour generation module, a prescription setting module, a radiation field adding module, a deep neural network dose prediction module, and an optimization objective function generation and plan design module. The deep neural network dose prediction module is used for providing a reasonable dose design target for a reverse optimization process according to data obtained by the same disease type; after the deep neural network model is trained, the dose distribution condition of a radiotherapy patient can be quickly predicted within several minutes, and radiotherapy plan design is automatically carried out, so the working efficiency of a radiotherapy doctor is effectively improved, and the formulation of a radiotherapy scheme of the patient is accelerated.

The invention provides a radiotherapy automatic planning system, which comprises an information input unit, a parameter generation unit, a dose prediction unit, an evaluation unit and an output unit, and is characterized in that the information input unit is used for inputting patient information including medical image information, contour sketching information and prescription dose information; the parameter generation unit generates initialization parameters based on a first neural network model according to the medical image information, the contour sketching information and the prescription dosage information, and outputs the initialization parameters to the TPS, so that the TPS can determine a radiotherapy plan by using the initialization parameters; the dose prediction unit predicts the dose distribution of the radiotherapy plan generated by the TPS based on a second neural network model according to the medical image information and the contour sketching information; the evaluation unit compares the dose distribution of the radiotherapy plan determined by the TPS with the predicted dose distribution to generate an evaluation result; and the output unit outputs the radiotherapy plan generated by the TPS according to the evaluation result. The efficiency of designing the radiotherapy plan can be improved.

The invention provides a heparin dose prediction method and device, and the method comprises the steps: obtaining the clinical observation data of a patient for injection, and extracting a parameter value corresponding to a heparin dose influence parameter contained in the clinical observation data according to a preset heparin dose influence parameter; respectively combining the parameter value with each normalized heparin dose value in a preset normalized heparin dose set to obtain a combined parameter value matched with the input of a preset treatment classification model; inputting the combined parameter value into a preset treatment classification model to obtain a probability value that the combined parameter value belongs to a normal treatment in treatment classification, and obtaining a target combined parameter value corresponding to the maximum probability value; and predicting the heparin dose of the patient to be injected according to the heparin dose range corresponding tothe normal treatment and the normalized heparin dose value in the target combination parameter value. The prediction accuracy of the heparin dose can be improved.

The invention discloses a dose prediction method and device for robot radiotherapy equipment. The method comprises the steps: performing model training, and building a patient model body according to a medical image of a patient; according to parameters of a treatment head of robot radiotherapy equipment, calculating dose distribution H of a single radiation field of a patient die body by using a first radiotherapy dose calculation method; according to the parameters of the treatment head of the robot radiotherapy equipment, calculating dose distribution L of the single radiation field of the patient die body by using a second radiotherapy dose calculation method; taking the L and the medical image of the patient as inputs, taking the H as an output, and sending the inputs into a deep learning neural network for training to obtain a radiotherapy dose prediction network; predicting doses, and calculating the dose distribution L * of the single radiation field of any patient die body by using a second radiotherapy dose calculation method; and obtaining the dose distribution H * of the single radiation field of any patient die body, predicted by the output end of the radiotherapy dose prediction network. The technical effect that one algorithm is used for predicting the calculation result output by the other algorithm is achieved through the deep learning network.

The invention discloses a system and method for dose calculation in generating radiation treatment plans. A first dose calculated using a first set of fluence maps and a first (faster) dose predictionmodel is accessed. A second fluence map is accessed. The second fluence map is projected onto the first set of fluence maps to determine a set of scalars and a residual value. When the residual valuesatisfies a criterion, a second dose is calculated using the first dose prediction model, the set of scalars, and the second fluence map. When the residual value does not satisfy the criterion, the second dose is calculated using a second (more accurate) dose prediction model and the second fluence map.

The present invention relates to the field of dose prediction in radiotherapy, and proposes an automatic radiotherapy dose prediction method based on a deep neural network, comprising the following steps: firstly, obtaining a CT image of a patient; secondly, delineating a patient's target area and organs at risk based on the CT image; Then, calculate the non-modulated dose of the patient, where the non-modulated dose is used to provide global information; then, build a dose prediction model based on deep neural network based on CT images, target volumes, organs at risk and non-modulated dose; finally, use depth-based The dose prediction model of neural network can automatically predict the radiotherapy dose. The present invention can automatically predict patient dose distribution based on deep neural network, the process does not need to manually extract features or set parameters, and the predicted dose distribution can accelerate the formulation of radiotherapy plan for patients, and the model uses non-modulated dose in the input part , which can provide more global information for accurate prediction of dose distribution.

The invention provides a method for generating a deliverable radiotherapy plan according to three-dimensional space dose distribution. The method comprises the following steps of preprocessing obtained patient radiotherapy related data used for generating the deliverable radiotherapy plan, and predicting a field dose in each radiation field direction through a preset field dose prediction model; and further based on the predicted field dose, predicting a fluence map of each corresponding radiation field direction through a preset fluence map prediction model, and finally, generating a deliverable radiotherapy plan based on the predicted fluence map of each radiation field direction and radiotherapy equipment parameters, wherein the generated deliverable radiotherapy plan comprises the information of each radiation field direction and related parameters of an accelerator, so that the generated deliverable radiotherapy plan is the deliverable plan which can be executed by a radiotherapy accelerator. Since a deep convolutional neural network is adopted to predict the field dose and the fluence map, the dosimetry features of original three-dimensional space dose distribution are reserved through a high-dimensional feature automatic extraction mode, dose fidelity before and after conversion is high, and generation speed is high.

The invention provides a method for predicting the three-dimensional dose distribution of an intensity-modulated radiotherapy plan based on deep learning. The method includes: collecting effective intensity-modulated radiotherapy planning data to form a case database; extracting the three-dimensional anatomical structure contour features of each patient's region of interest from the case database; dividing the three-dimensional anatomical structure contour of the patient's region of interest into several binary Two-dimensional contour slice map; extract the dose characteristics of each patient from the case database, and divide it into several two-dimensional dose slice distribution maps; build a deep convolutional network, input the patient's two-dimensional contour slice map and the corresponding two-dimensional dose slice distribution Figure, the association model between anatomical structure contour features and dose features is obtained through model training; the three-dimensional dose distribution of new patients is predicted using the trained association model. By using the method of the present invention, the correlation between anatomical structure features and dose features can be effectively obtained, and the accuracy of dose prediction can be improved.

The invention discloses a predication method for three-dimensional dose distribution in intensity modulated radiation therapy plans. The method includes steps of (1) collecting valid intensity modulated radiation therapy plan data and forming a case database; (2) dividing a PTV and different to-be-endangered organs of patients into a plurality of voxels according to the resolution ratio of CT images; (3) extracting anatomical characteristics of each patient in the database; (4) extracting dose characteristics of each patient in the database; (5) constructing an artificial neural network, inputting the anatomical characteristics and the dose characteristics of the patients, and learning the mapping relation between the anatomical characteristics and the dose characteristics by the aid of the artificial neural network, and obtaining a correlation model of the anatomical characteristics and the dose characteristics; (6) using the correlation model to predicate the three-dimensional dose distribution of a new patient. The application of said method is using the dose distribution predication method for dose prediction for to-be-endangered organs of patients and quality control is achieved. By adopting the above method, predication of three-dimensional dose distribution in intensity modulated radiation therapy plans can be realized and the method can be applied to a quality control link.