International Business Machines Corporation

Merative US L.p.

Mark D Bronkalla

Arkadiusz Sitek

Yiting Xie

Murray A Reicher

Marwan Sati

Sun Young Park

Dustin Michael Sargent

Ben Graf

Grant Covell

James G Thompson

Benedikt Werner Graf

Larissa Christina Schudlo

Prasanth Prasanna

Rebecca Trunnell-Hyman

Amanda Long

James Boritz

John Hansen

David Joseph Richmond

Lilla Boroczky

Yufan Guo

James H Thompson

Amin Katouzian

Pedro Luis Esquinas Fernandez

Colin Compas

Michael Trambert

David Richmond

Amy Chenault

Karl Diedrich

Jennifer C Quattrone-Brown

Weber Marett

Systems and methods for selecting a prior comparison study. One system includes an electronic processor configured to, for a medical image study associated with a patient, select a prior comparison image study. The electronic processor is also configured to automatically determine, based on monitored user interaction with the selected prior comparison image study, a usefulness of the selected prior comparison image study. The electronic processor is also configured to automatically update a selection model based on the usefulness of the prior comparison image study to a user.

Intelligent automatic selection of a prior comparison study

Systems and methods for selectively processing image studies with an artificial intelligence system. One system includes an electronic processor configured to select an image study awaiting review and update a workflow status of the image study to a first status indicating that the image study has been claimed for review by the artificial intelligence system. The electronic processor is also configured to apply at least one of the plurality of rules to the image study to determine whether the image study is applicable for processing by the artificial intelligence system, and, in response to determining the image study is not applicable for processing by the artificial intelligence system based on the at least one of the plurality of rules, update the workflow status associated with the image study to a second status to make the image study available for claiming by a manual reviewer or another artificial intelligence system.

Enhancing medical imaging workflows using artificial intelligence

A method, a computer system, and a computer program product for marketplace creation and model training in a blockchain network is provided. Embodiments of the present invention may include uploading images to a marketplace. Embodiments of the present invention may include recording references to the images on a blockchain ledger. Embodiments of the present invention may include measuring local and global annotations. Embodiments of the present invention may include determining an image quality threshold has been met. Embodiments of the present invention may include providing an annotation quality metric to the marketplace. Embodiments of the present invention may include pre-processing and standardizing the images. Embodiments of the present invention may include training a model. Embodiments of the present invention may include recording the model in the blockchain ledger. Embodiments of the present invention may include providing the model to the marketplace.

Method, system, and media for machine learning annotation and image marketplace using blockchain ledgers

An approach for a computer program to receive image data of a subject including at least a portion of a spine of the subject and a chronological age of the subject. The approach includes the computer program pre-processing the image data including at least a portion of a spine. The approach includes determining an apparent age of the spine or a portion of the spine of the subject using a trained artificial intelligence deep learning algorithm.

Estimating spinal age

Methods and systems of summarizing medical data. One system includes an electronic processor configured to analyze medical data to extract a medical concept and a plurality of additional attributes of the medical concept and store the medical concept and the plurality of additional attributes. The electronic processor is configured to generate a first medical summary associated with the patient, where the first medical summary is based on the stored medical concept and at least a first additional attribute included in the stored plurality of additional attributes. The electronic processor is configured to receive a user interaction with the first medical summary. The electronic processor is configured to generate a second medical summary associated with the patient based on the user interaction, the second medical summary is based on the stored medical concept and at least a second additional attribute included in the stored plurality of additional attributes.

Dynamic medical summary

Automated peer review of medical imagery is provided. In some embodiments, at least one finding is determined for a present study. The present study has a subject anatomy. Based on the at least one finding and the subject anatomy, at least one prior study is selected. The at least one prior study has subject anatomy related to the subject anatomy of the present study and does not include the at least one finding. The at least one prior study is provided to a user for review with respect to the at least one finding.

Automated peer review of medical imagery

A method, system, and computer program product provide disease simulation in synthetic projection imagery. The method obtains first medical imaging data of a first imaging type as source imaging data. A second imaging type to be generated from the source imaging data is identified. The method identifies a parameter set for the second imaging type. Second medical imaging data is modeled from the first medical imaging data based on the parameter set. A set of synthetic images is generated from the first medical imaging data based on the modeled second medical imaging data.

Disease simulation in medical images

One or more processors may identify a missing image in the set of multi-view images. Each of the images is associated with a particular view type. The one or more processors may generate, utilizing a replacement AI model, a replacement image for the missing image in the set of multi-view images. The replacement image is generated utilizing an AI model trained to generate a replacement image using training images from two or more time-adjacent sets of images. The one or more processors may identify a duplicate image in the set of multi-view images. The one or more processors may generate, utilizing a quality generative AI model, a characteristic improved image based on the duplicate image for the set of multi-view images. The one or more processors may output the replacement image and the characteristic improved image.

Study-level multi-view processing system

Methods and systems for detecting a dissection in surface of an elongated structure in a three dimensional medical image. One system includes an electronic processor configured to receive the three dimensional medical image and determine a periphery of the elongated structure included in the three dimensional medical image. The electronic processor is also configured to generate a non-contrast image representing the periphery of the elongated structure and superimpose a contrast image associated with the three dimensional image on top of the non-contrast image to generate a superimposed image. The electronic processor is also configured to detect at least one dissection in the elongated structure using the superimposed image and output a medical report identifying the at least one dissection detected in the elongated structure.

Vascular dissection detection and visualization using a superimposed image

Systems and methods for generating electronic reports for medical images. One system includes an electronic processor configured to access a plurality of prior reports associated with a user and automatically generate a mapping using machine learning based on the plurality of prior reports. The mapping associates language included in the plurality of prior reports with at least one section of a report. The system stores the mapping to a memory, receives input from the user for an electronic report associated with a medical image, accesses the mapping from the memory, automatically determines a section in the electronic report associated with the input based on the mapping, and automatically inserts text into the section in the electronic report based on the input.

Automatically associating user input with sections of an electronic report using machine learning

In an approach to automatic detection of vertebral dislocations, training data is received, where the training data includes an ordered sequence of radiographic image patches of vertebrae and intervertebral spaces, and location data for each vertebra and each intervertebral space. Location deep learning models are trained to detect a location of each vertebra and each intervertebral space from the training data. Classification deep learning models are trained to classify an ordered sequence of image patches to identify vertebral anomalies of the vertebrae and the intervertebral spaces from the training data. Responsive to receiving radiographic image files, the location deep learning models and the classification deep learning models are applied to the radiographic image files to create a condition assessment.

Automatic detection of vertebral dislocations

Retrospective review of electronic medical records is provided. In some embodiments, a retrospective review of a plurality of electronic medical records is performed. The retrospective review includes searching for electronic medical records relevant to a medical condition. Based on the retrospective review, a plurality of studies most relevant to a present study is determined. Each of the plurality of studies are flagged for supplemental review in a worklist.

Automatic retrospective review of electronic medical records

Measurement of medical images as a hybrid cloud service is provided. In various embodiments, pre-trained parameters are received at a client from a remote server. A local cognitive system is instantiated using the pre-trained parameters. The cognitive system is applied to evaluate a medical image. A result is sent to the remote server for training of a remote cognitive system.

Growing community of inventors

Waukesha, WI, United States of America

Mark D Bronkalla