Abstract: For magnetic resonance (MR) reconstruction using artificial intelligence (AI), the AI-based reconstruction for MR imaging systems is offloaded to one or more servers. A remote server performs AI-based reconstruction. A library of recent, old, custom, and/or publicly available AI-based reconstruction processes may be rapidly deployed and available to the server, which has the memory and processing resources for AI-based reconstruction. Load balancing of the data and/or between servers may improve performance.

Abstract: Systems and methods for training an artificial intelligence-based system using self-supervised learning are provided. For each respective training medical image of a set of unannotated training medical images, the following steps are performed. A first augmented image is generated by applying a first augmentation operation to the respective training medical image. A second augmented image is generated by applying a second augmentation operation to the respective training medical image. A first representation vector is created from the first augmented image using an encoder network. A second representation vector is created from the second augmented image using the encoder network. The first representation vector is mapped to first cluster codes. The second representation vector is mapped to second cluster codes. The encoder network is optimized using the first and second representation vectors and the first and second cluster codes.

Abstract: Systems and methods for performing a medical imaging analysis task are provided. A pre-contrast medical image and a plurality of post-contrast medical images of an anatomical object of a patient are received. A mask of the anatomical object is generated based on at least one of 1) the pre-contrast medical image or 2) one or more of the plurality of post-contrast medical images. One or more processed post-contrast medical images each associated with a respective feature are generated based on the plurality of post-contrast medical images and the mask of the anatomical object. A medical imaging analysis task is performed using a machine learning based network based on the one or more processed post-contrast medical images and the pre-contrast medical image masked with the mask of the anatomical object. Results of the medical imaging analysis task are output.

Abstract: For AI-based recommendations in a service management system, the AI is machine trained using gamification. A model of the service management system is used in simulation to train a policy in reinforcement learning to implement strategies for improvement of KPI(s). By varying sampling of distribution of parameters of the model and/or varying the distributions of parameters used in the model, the policy learns to deal with a variety of situations using the simulations from the model. The resulting AI (machine-learned policy) is used to make recommendations for the service management system.

Abstract: Methods and systems for image registration using an intelligent artificial agent are disclosed. In an intelligent artificial agent based registration method, a current state observation of an artificial agent is determined based on the medical images to be registered and current transformation parameters. Action-values are calculated for a plurality of actions available to the artificial agent based on the current state observation using a machine learning based model, such as a trained deep neural network (DNN). The actions correspond to predetermined adjustments of the transformation parameters. An action having a highest action-value is selected from the plurality of actions and the transformation parameters are adjusted by the predetermined adjustment corresponding to the selected action. The determining, calculating, and selecting steps are repeated for a plurality of iterations, and the medical images are registered using final transformation parameters resulting from the plurality of iterations.

Abstract: A position prediction of a target is provided based on a segmentation of a context of the target. Alternatively, or additionally, to a spatial context of the target, it is also possible to consider a temporal context. A catheter tip can be tracked.

Abstract: CT scan parameters for performing a CT scan of an anatomical target region of a patient are determined and/or adjusted. An initial set of the CT scan parameters for starting to perform the CT scan is determined based on an initial set of attenuation curves associated with the anatomical target region of the patient. The initial set of attenuation curves are determined based on optical imaging data depicting the patient.

Abstract: Similar pre-stored medical datasets are identified by comparison with a current case dataset. A current case dataset is provided and includes radiological data of a patient. A number of pre-stored medical datasets each including radiological data of other patients are provided. Each case dataset is evaluated according to a predefined AI-based method to obtain a number of definitive features for that case dataset. The definitive features of the current case dataset are compared with the definitive features of each pre-stored medical dataset to identify a number of pre-stored medical datasets most similar to the current case dataset. The identified number of most similar pre-stored medical datasets are output.

Abstract: Systems and methods for automatically determining an assessment of a patient are provided. A patient is automatically interacted with, by a first trained machine learning based model, to acquire initial patient data. One or more risk factors associated with the patient are automatically determined, by a second trained machine learning based model, based on the received initial patient data. The patient is automatically interacted with, by the first trained machine learning based model, to acquire additional patient data based on the one or more determined risk factors. An assessment of the patient is automatically determined, by the second trained machine learning based model, based on the initial patient data and the additional patient data. The assessment of the patient is output.

Abstract: Provided are computer-implemented methods and systems for classifying a medical image data set. In particular, a method is provided comprising the steps of receiving the medical image dataset of a patient, of providing a first classification stage configured to classify the medical image dataset as normal or not-normal, of providing a second classification stage different than the second classification stage and configured to classify the medical image dataset as normal or not-normal, and of subjecting the medical image dataset to the first classification stage to classify the medical image dataset as normal or not-normal. Further, the method comprises subjecting the medical image dataset to the second classification stage to classify the medical image dataset as normal or not-normal, if the medical image dataset is classified as normal in the first classification stage.

Abstract: A method is for generating modified medical images. An embodiment of the method includes receiving a first medical image displaying an abnormal structure within a patient, and applying a trained inpainting function to the first medical image to generate a modified first medical image, the trained inpainting function being trained to inpaint abnormal structures within a medical image. The method includes determining an abnormality patch based on the first medical image and the modified first medical image; receiving a second medical image of the same type as the first medical image; and including the abnormality patch into the second medical image to generate a modified second medical image. A method is for detecting abnormal structures using a trained detection function trained based on modified second medical images. Systems, computer programs and computer-readable media related to those methods are also disclosed.

Abstract: Systems and methods for determining an optimal position of one or more ablation electrodes are provided. A current state of an environment is defined based on a mask of one or more anatomical objects and one or more current positions of one or more ablation electrodes. The one or more anatomical objects comprise one or more tumors. For each particular AI (artificial intelligence) agent of one or more AI agents, one or more actions for updating the one or more current positions of a respective ablation electrode of the one or more ablation electrodes in the environment are determined based on the current state using the particular AI agent. A next state of the environment is defined based on the mask and the one or more updated positions of the respective ablation electrode.

Abstract: Systems and methods for generating synthesized medical images of a tumor are provided. A 3D mask of an anatomical structure generated from a 3D medical image and a 3D image of a plurality of concentric spheres are received. A 3D mask of a tumor is generated based on the 3D mask of the anatomical structure and the 3D image of the plurality of concentric spheres using a first 3D generator network. A 3D intensity map of the tumor is generated based on the 3D mask of the tumor and the 3D image of the plurality of concentric spheres using a second 3D generator network. A 3D synthesized medical image of the tumor is generated based on one or more 2D slices of the 3D intensity map of the tumor and one or more 2D slices of the 3D medical image using a 2D generator network. The 3D synthesized medical image of the tumor is output.

Abstract: Systems and methods for generating a synthetic image are provided. An input medical image in a first modality is received. A synthetic image in a second modality is generated from the input medical image. The synthetic image is upsampled to increase a resolution of the synthetic image. An output image is generated to simulate image processing of the upsampled synthetic image. The output image is output.

Abstract: Systems and methods for occlusion detection in medical images are provided. An input medical image of one or more vessels in an anatomical object of a patient is received. One or more anatomical landmarks are identified in the input medical image. A first patch and one or more additional patches are extracted from the input medical image based on the identified one or more anatomical landmarks. The first patch and the one or more additional patches depict different portions of the anatomical object. Features are extracted from the first patch and the one or more additional patches using a machine learning based feature extractor network. An occlusion in the one or more vessels is detected in the first patch based on the extracted features with or without modeling features on a probability distribution function. Results of the detecting are output.

Abstract: For reconstruction of an image in MRI, unsupervised training (i.e., data-driven) based on a scan of a given patient is used to reconstruct model parameters, such as estimating values of a contrast model and a motion model based on fit of images generated by the models for different readouts and times. The models and the estimated values from the scan-specific unsupervised training are then used to generate the patient image for that scan. This may avoid artifacts from binning different readouts together while allowing for scan sequences using multiple readouts.

Abstract: Systems and methods for predicting risk for a medical event associated with evaluating or treating a patient for a disease are provided. Input medical imaging data and patient data of a patient are received. The input medical imaging data includes abnormality patterns associated with a disease. Imaging features are extracted from the input medical imaging data using a trained machine learning based feature extraction network. One or more of the extracted imaging features are normalized. The one or more normalized extracted imaging features and the patient data are encoded into features using a trained machine learning based encoder network. Risk for a medical event associated with evaluating or treating the patient for the disease is predicted based on the encoded features.

Abstract: Methods and systems for image registration using an intelligent artificial agent are disclosed. In an intelligent artificial agent based registration method, a current state observation of an artificial agent is determined based on the medical images to be registered and current transformation parameters. Action-values are calculated for a plurality of actions available to the artificial agent based on the current state observation using a machine learning based model, such as a trained deep neural network (DNN). The actions correspond to predetermined adjustments of the transformation parameters. An action having a highest action-value is selected from the plurality of actions and the transformation parameters are adjusted by the predetermined adjustment corresponding to the selected action. The determining, calculating, and selecting steps are repeated for a plurality of iterations, and the medical images are registered using final transformation parameters resulting from the plurality of iterations.

Abstract: Systems and methods for generating a synthesized medical image are provided. An input medical image is received. A synthesized segmentation mask is generated. The input medical image is masked based on the synthesized segmentation mask. The masked input medical image has an unmasked portion and a masked portion. An initial synthesized medical image is generated using a trained machine learning based generator network. The initial synthesized medical image includes a synthesized version of the unmasked portion of the masked input medical image and synthesized patterns in the masked portion of the masked input medical image. The synthesized patterns is fused with the input medical image to generate a final synthesized medical image.

Abstract: For prediction of response of radiation therapy, radiomics are used for unsupervised machine training of an encoder-decoder network to predict based on input of image data, such as computed tomography image data and from segmentation. The trained encoder is then used to generate latent representations to be used as input to different classifiers or regressors for prediction of therapy responses, such as one classifier to predict response for an organ at risk and another classifier to predict another type of response for the organ at risk or to predict a response for the tumor.