Abstract: A system (100) includes an end point prediction engine (150) that predicts an end point (302) using a machine learning model (132) and one or more clinical report objects (152) for a patient, wherein the machine learning model inputs the one or more clinical report objects and outputs the predicted end point according to phrases or n-grams in the one or more clinical report objects. An end point visualization interface (160) visualizes the predicted end point (302) using a scorecard (162) or a timeline (164). An end point modeling engine (130) generates the machine learning model from training data that includes validated end points (122) and clinical report objects (116).

Abstract: A controller (122, 910/920) for interventional procedure optimization includes a processor (12210, 910) and a memory (12220, 920) that stores instructions. When executed by the processor, the instructions cause the controller (12210, 910) to implement a process that includes identifying (S210) anatomical characteristics from pre-interventional imagery of anatomy for each of multiple candidate types of an interventional procedure for the anatomy and comparing (S220) the anatomical characteristics with tool characteristics of candidate tools to use in each of the candidate types. The process also includes generating (S240) a feasibility report for each of the candidate types based on the identifying and the comparing. Each feasibility report includes a feasibility grade for each of the candidate types. The process also includes selecting (S260), based on the feasibility reports, an optimal interventional procedure type among the candidate types.

Abstract: The present disclosure is directed to methods and apparatus for validating and authenticating use of machine learning models. For example, various techniques are described herein to limit the vulnerability of machine learning models to attack and/or exploitation of the model for malicious use, and for detecting when such attack/exploitation has occurred. Additionally, various embodiments described herein promote the protection of sensitive and/or valuable data, for example by ensuring only licensed use is permissible. Moreover, techniques are described for version tracking, usage tracking, permission tracking, and evolution of machine learning models.

Abstract: Image co-registration entails: emitting energy, and responsively and dynamically acquiring images, the acquired images having a given dimensionality; selecting, repeatedly, and dynamically with the acquiring, from among the acquired images and, as a result of the selecting, changing, repeatedly, and dynamically, membership in a set of images of the given dimensionality; and, synchronously with the change, dynamically and iteratively registering the set to an image having a dimensionality higher than the given dimensionality. The co-registration is realizable with an imaging probe for the acquiring, and a tracking system for tracking a location, and orientation, of the probe, the registering being specialized for the acquiring with the probe being dynamically, during the acquiring, any one or more of angulated, rotated and translated.

Abstract: Image co-registration (S390) entails: emitting energy, and responsively and dynamically acquiring images (216), the acquired images having a given dimensionality; selecting, repeatedly, and dynamically with the acquiring, from among the acquired images and, as a result of the selecting, changing, repeatedly, and dynamically, membership in a set (226) of images of the given dimensionality; and, synchronously with the change (S376), dynamically and iteratively registering the set to an image having a dimensionality higher than the given dimensionality. The co-registration is realizable with an imaging probe (140) for the acquiring, and a tracking system for tracking a location, and orientation, of the probe, the registering being specialized for the acquiring with the probe being dynamically, during the acquiring, any one or more of angulated, rotated and translated.

Abstract: The present disclosure pertains to computer-assisted search of image slices. In some embodiments, an image slice having a detected finding (and representing a cross section of at least a portion of an individual) may be determined. A search space may be reduce to a subset of image slices respectively representing a cross section corresponding to the cross section represented by the determined image slice. The search space reduction may comprise filtering a set of image slices based on the cross section represented by the determined image slice. Image slice information related to the determined image slice and related one or more image slices (of the image slices subset) may be obtained based on the search space reduction. A determination of whether indications of the detected finding exist in the image slices (of the image slices subset) may be effectuated based on the image slice information.

Abstract: Methods and systems for correcting an examination report. The methods described herein extract examination and semantic data from an examination report, and identify any discrepancies between the extracted examination data and the extracted semantic data. The methods described herein then receive a resolution strategy regarding how to resolve any identified discrepancies and then resolve any identified discrepancies based on the resolution strategy.

Abstract: A radiology workstation (10) includes a computer (12) connected to receive a stack of radiology images of a portion of a radiology examination subject. The computer includes at least one display component (14) and at least one user input component (16). The computer includes at least one processor (22) programmed to: display selected radiology images of the stack of radiology images on the at least one display component; receive entry of a current radiology report via the at least one user input component and displaying the entered radiology report on the at least one display component; identify a radiology finding by at least one of (i) automated analysis of the stack of radiology images and (ii) detecting textual description of the radiology finding in the radiology report; identify or extract at least one key image from the stack of radiology images depicting the radiology finding; and embed or link the at least one key image with the radiology report.

Abstract: A segmentation selection system includes a transducer configured to transmit and receive imaging energy for imaging a subject. A signal processor is configured to process imaging data received to generate processed image data. A segmentation module is configured to generate a plurality of segmentations of the subject based on features or combinations of features of the imaging data and/or the processed image data. A selection mechanism is configured to select one of the plurality of segmentations that best meets a criterion for performing a task. A graphical user interface permits a user to select features or combinations of features of imaging data or processed image data to generate the plurality of segmentations and to select a segmentation that best meets criterion for performing a task.

Abstract: An intervention system employs an optical shape sensing tool (32) (e.g., a brachytherapy needle having embedded optical fiber(s)) and a grid (50, 90) for guiding an insertion of the optical shape sensing tool (32) into an anatomical region relative to a grid coordinate system. The intervention system further employs a registration controller (74) for reconstructing a segment or an entirety of a shape of the optical shape sensing tool (32) relative to a needle coordinate system, and for registering the needle coordinate system to the grid coordinate system as a function of a reconstructed segment/entire shape of the optical shape sensing tool (32) relative to the grid (50, 90) (i.e., reconstruction of a segment/entire shape of the OSS needle inserted into/through the grid serving as a basis for the grid/needle coordinate system registration).

Abstract: The present disclosure is directed to methods and apparatus for validating and authenticating use of machine learning models. For example, various techniques are described herein to limit the vulnerability of machine learning models to attack and/or exploitation of the model for malicious use, and for detecting when such attack/exploitation has occurred. Additionally, various embodiments described herein promote the protection of sensitive and/or valuable data, for example by ensuring only licensed use is permissible. Moreover, techniques are described for version tracking, usage tracking, permission tracking, and evolution of machine learning models.

Abstract: A method (100) for generating a domain-specific training set, comprising: generating (130) a generic corpus comprising a plurality of tokenized documents, comprising: (i) parsing (132) a document retrieved from the generic corpus; (ii) preprocessing (134) the parsed document; (iii) tokenizing (136) the preprocessed document; and (iv) storing (138) the tokenized document in the generic corpus; generating (140) an ontology database of tokenized entries, comprising: (i) parsing (142) an ontology entry retrieved from an ontology; (ii) preprocessing (144) the parsed entry; (iii) tokenizing (146) the preprocessed entry; and (iv) storing (148) the tokenized entry in the ontology database; querying (150), using domain-specific tokenized entries from the ontology database, the tokenized documents in the generic corpus; identifying (160), based on the query, a plurality of tokenized documents specific to the domain; and storing (170), in a training set database, the identified tokenized documents as a training set spec

Abstract: A segmentation selection system includes a transducer (14) configured to transmit and receive imaging energy for imaging a subject. A signal processor (26) is configured to process imaging data received to generate processed image data. A segmentation module (50) is configured to generate a plurality of segmentations of the subject based on features or combinations of features of the imaging data and/or the processed image data. A selection mechanism (52) is configured to select one of the plurality of segmentations that best meets a criterion for performing a task.

Abstract: Methods and systems for selecting a treatment for a patient. The system extracts an incidental finding from a record associated with a patient and an associated follow-up recommendation. The system then determines whether any inconsistencies exist between the follow-up recommendation from the report and a follow-up recommendation prescribed by institutional or administrative guidelines. Any inconsistencies may then be resolved to guarantee an appropriate workflow for patient care.

Abstract: A system for planning and performing a repeat interventional procedure is provided which includes a registration device and an image generation device which map current targets in a reference image for a first interventional procedure to at least one guidance image acquired from a different imaging modality. Biopsy locations are recorded to the guidance images during the first interventional procedure and the biopsy locations are mapped to the reference image to provide a planning image for use in a subsequent interventional procedure on the patient. In a subsequent interventional procedure, the prior planning image (124) may be registered to a current reference image and the prior biopsy locations and prior and current targets are mapped to a guidance image acquired from a different imaging modality. Biopsy locations are then mapped to the guidance image and mapped back to the current reference image.

Abstract: An imaging apparatus (24) images an introduction element (17) like a needle or a catheter for performing a biopsy or a brachytherapy. The introduction element (17) includes at least one ultrasound receiver (21) arranged at a known location. An ultrasound probe (12) for being inserted into a living being (2) emits ultrasound signals for acquiring ultrasound data of an inner part (19) of the living being (2). A first tracking unit (3) tracks the location of the introduction element (17) based on a reception of the emitted ultrasound signals by the at least one ultrasound receiver (21). An imaging unit (4) generates an indicator image showing the inner part (19) and an indicator of the introduction element (17) based on the tracked location. A display (5) displays the indicator image providing feedback about the location of the introduction element (17).

Abstract: There is provided a computer-implemented method for detecting an incidental finding in a radiology report associated with an imaging examination of a subject. The method comprises acquiring a first finding from the radiology report, the first finding being indicative of at least one of: an imaging observation, a disorder, and an abnormality identified from the imaging examination, acquiring information associated with the radiology report, wherein the acquired information associated with the radiology report comprises a reason for performing the imaging examination, a modality of the imaging examination, and a body part of the subject examined in the imaging examination, and determining whether the first finding is an incidental finding based on the acquired information associated with the radiology report, wherein the incidental finding is an unanticipated finding not related to a diagnostic inquiry associated with the imaging examination.

Abstract: A system for planning and performing a repeat interventional procedure includes a registration device (116) and an image generation device (120) which map current targets in a reference image (104) for a first interventional procedure to at least one guidance image (112) acquired from a different imaging modality. Biopsy locations are recorded to the guidance images during the first interventional procedure and the biopsy locations are mapped to the reference image to provide a planning image for use in a subsequent interventional procedure on the patient. In a subsequent interventional procedure, the prior planning image (124) may be registered to a current reference image (126) and the prior biopsy locations and prior and current targets are mapped to a guidance image acquired from a different imaging modality. Biopsy locations are then mapped to the guidance image (115) and mapped back to the current reference image.

Abstract: A system for registration feedback includes a segmentation module configured to segment a relevant three-dimensional structure in a first image to be registered and to segment a same relevant structure in a second image to be registered to provide three-dimensional segmentations of the first and second images. A registration module is configured to register the three-dimensional segmentations of the first and second images by applying a registration transformation to one of the three-dimensional segmentations to map one three-dimensional segmentation onto the coordinate space of the other.

Abstract: The present system is configured to obtain training information related to patients, and obtain a user input indicating prediction criteria that are to be used by a prediction model for generating patient-related predictions. The prediction criteria include which and how many prediction-contributing features are to be used by the prediction model for generating patient-related predictions. The system is configured to generate the prediction model based on the prediction criteria and the training information; and generate, based on the prediction model and patient information associated with a patient, a prediction related to a health outcome of the patient. The system is also configured to cause display of the prediction and other predictions, wherein the display comprises a scaled display of two or more of the prediction-contributing features.