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 computer-implemented method for planning a thermal ablation of a target object within a biological body includes acquiring an object image within the body, determining an object position within the body from the image, determining external body surface position relative to the object position from the image, acquiring, for an initial set of ablation needles those of types for the ablation, and for each type, a set of characterizing features common to all needles of a same type, including a fixed and/or variable parameter. A neural ordinary differential equation algorithm receives a characterizing feature, external surface position, object position, algorithm for outputting an ablation plan, including a final set of needles for ablating the object, and for each needle of the final set, type, trajectory from the external surface, and optionally, a variable parameter value. The plan is provided through an interface to guide a clinician for object ablation.

Abstract: For quantification of blood flow by an ultrasound system, B-mode images generated with a multi-transmit, coherent image formation produce swirling or other speckle patterns in the blood regions. These patterns, as represented in specially formed B-mode images, are tracked over time to indicate two or three-dimensional velocity vectors of the blood at a B-mode resolution. Various visualizations may be provided at the same resolution, including the velocity flow field, flow direction, vorticity, vortex size, vortex shape, and/or divergence.

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: For robotically operating a catheter, translation and/or rotation manipulation is provided along the shaft or away from the handle, such as near a point of access to the patient. A worm drive arrangement may allow for both translation and rotation of the shaft. Some control may be provided by robotic manipulation of the handle, while other control (e.g., fine adjustments) are made by robotic manipulation of the shaft.

Abstract: Systems and methods for navigating a catheter in a patient using a robotic navigation system with risk management are provided. An input medical image of a patient is received. A trajectory for navigating a catheter from a current position to a target position in the patient is determined based on the input medical image using a trained segmentation network. One or more actions of a robotic navigation system for navigating the catheter from the current position towards the target position and a confidence level associated with the one or more actions are determined by a trained AI (artificial intelligence) agent and based on the generated trajectory and a current view of the catheter. In response to the confidence level satisfying a threshold, the one or more actions are evaluated based on a view of the catheter when navigated according to the one or more actions.

Abstract: For robotically operating a catheter, a medical catheter is controlled by rotation of the catheter as well as steering in one or more planes of a distal end of the catheter. To robotically rotate the catheter, a handle is rotated. The steering is performed separately using one or more knobs on the handle. The rotation of the handle complicates the robotic control of the knob. A mechanical decoupling is used so that rotation of the handle maintains the position of the knob relative to the handle. Gearing or transmission is used to avoid independent control of the knob and handle rotation. In an alternative or additional approach, the handle may be robotically controlled while also guiding the catheter shaft spaced away from the handle, allowing fine-tuned control of the catheter at the access point to the patient.

Abstract: A method and system for deep learning based cardiac electrophysiological model personalization is disclosed. Electrophysiological measurements of a patient, such as an ECG trace, are received. A computational cardiac electrophysiology model is personalized by calculating patient-specific values for a parameter of the computational cardiac electrophysiology model based at least on the electrophysiological measurements of the patient using a trained deep neural network (DNN). The parameter of the computational cardiac electrophysiology model corresponds to a spatially varying electrical cardiac tissue property.

Abstract: Systems and methods for automatically navigating a catheter in a patient are provided. An image of a current view of a catheter in a patient is received. A set of actions of a robotic navigation system for navigating the catheter from the current view towards a target view is determined using a machine learning based network. The catheter is automatically navigated in the patient from the current view towards the target view using the robotic navigation system based on the set of actions.

Abstract: Systems and methods for predicting a patient specific risk of cardiac events for cardiac arrhythmia are provided. A medical image sequence of a heart of a patient is received. Cardiac function features are extracted from the medical image sequence. Additional features are extracted from patient data of the patient. A patient specific risk of a cardiac event is predicted based on the extracted cardiac function features and the extracted additional 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: For robotically operating a catheter, a medical catheter is controlled by rotation of the catheter as well as steering in one or more planes of a distal end of the catheter. To robotically rotate the catheter, a handle is rotated. The steering is performed separately using one or more knobs on the handle. The rotation of the handle complicates the robotic control of the knob. A mechanical decoupling is used so that rotation of the handle maintains the position of the knob relative to the handle. Gearing or transmission is used to avoid independent control of the knob and handle rotation. In an alternative or additional approach, the handle may be robotically controlled while also guiding the catheter shaft spaced away from the handle, allowing fine-tuned control of the catheter at the access point to the patient.

Abstract: Systems and methods for generating an ablation map identifying target ablation locations on a heart of a patient are provided. One or more input medical images of a heart of a patient and a voltage map of the heart of the patient are received. An ablation map identifying target ablation locations on the heart is generated using one or more trained machine learning based models based on the one or more input medical images and the voltage map. The ablation map is output.

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: An artificial agent based cognitive operating room system and a method thereof providing automated assistance for a surgical procedure are disclosed. Data related to the surgical procedure from multiple data sources is fused based on a current context. The data includes medical images of a patient acquired using one or more medical imaging modalities. Real-time quantification of patient measurements based on the data from the multiple data sources is performed based on the current context. Short-term predictions in the surgical procedure are forecasted based on the current context, the fused data, and the real-time quantification of the patient measurements. Suggestions for next steps in the surgical procedure and relevant information in the fused data are determined based on the current context and the short-term predictions. The suggestions for the next steps and the relevant information in the fused data are presented to an operator.

Abstract: DCE MR images are obtained from a MR scanner and under a free-breathing protocol is provided. A neural network assigns a perfusion metric to DCE MR images. The neural network includes an input layer configured to receive at least one DCE MR image representative of a first contrast enhancement state and of a first respiratory motion state and at least one further DCE MR image representative of a second contrast enhancement state and of a second respiratory motion state. The neural network further includes an output layer configured to output at least one perfusion metric based on the at least one DCE MR image and the at least one further DCE MR image. The neural network with interconnections between the input layer and the output layer is trained by a plurality of datasets, each of the datasets having an instance of the at least one DCE MR image and of the at least one further DCE MR image for the input layer and the at least one perfusion metric for the output layer.

Abstract: An ultrasound imager provides for LAA closure guidance. Using ultrasound imaging allows for modeling over time (e.g., throughout a heart cycle). An anatomy model of the LAA over time is used to create a biomechanical model personalized to the patient. The personalized models and a model of one or more closure devices are used to select a closure device for the patient appropriate for the entire heart cycle and to guide placement of the selected closure device during an implantation.

Abstract: For non-invasive EP mapping, a sparse number of electrodes (e.g., 10 in a typical 12-lead ECG exam setting) are used to generate an EP map without requiring preoperative 3D image data (e.g. MR or CT). An imager (e.g., a depth camera) captures the surface of the patient and may be used to localize electrodes in any positioning on the patient. Two-dimensional (2D) x-rays, which are commonly available, and the surface of the patient are used to segment the heart of the patient. The EP map is then generated from the surface, heart segmentation, and measurements from the electrodes.

Abstract: For registration of medical images with deep learning, a neural network is designed to include a diffeomorphic layer in the architecture. The network may be trained using supervised or unsupervised approaches. By enforcing the diffeomorphic characteristic in the architecture of the network, the training of the network and application of the learned network may provide for more regularized and realistic registration.

Abstract: An ultrasound imager provides for LAA closure guidance. Using ultrasound imaging allows for modeling over time (e.g., throughout a heart cycle). An anatomy model of the LAA over time is used to create a biomechanical model personalized to the patient. The personalized models and a model of one or more closure devices are used to select a closure device for the patient appropriate for the entire heart cycle and to guide placement of the selected closure device during an implantation.