International Business Machines Corporation

Guerbet

Merative US L.p.

Yi-Qing Wang

Giovanni John Jacques Palma

Dana Levanony

Pedro Luis Esquinas Fernandez

Moshe Raboh

Arkadiusz Sitek

Paul Dufort

Omid Bonakdar Sakhi

Thomas Binder

Mechanisms for identifying inconsistencies between volumes of medical images are provided. A plurality of volumes of medical images are received, each having a plurality of medical images of an anatomical structure. A plurality of first representation data structures are generated, each corresponding to a volume and having dimensional measurements of the anatomical structure at various locations. A reference data structure is generated, for the anatomical structure based on the first representation data structures, having second dimensional measurements derived from the first dimensional measurements. A discrepancy is detected between a second representation data structure and the reference data structure based on a comparison and a notification of the discrepancy where the notification identifies a type of the discrepancy.

Automated medical image and segmentation quality assessment for machine learning tasks

A method and system perform single phase and multi-phase contour refinement of lesions. The method includes receiving a three dimensional input mask; receiving input slices from the medical images including a lesion; cropping the input slices with the input mask; performing lesion contour refinement for the cropped input slices and the input mask to obtain a predicted mask; and storing the predicted mask that includes 3D lesion contour refinement. A multiphase method includes deforming the 3D input mask from the reference phase to a target phase or warping the input slices from the target phase to the reference phase before contour refinement. The warped images generate an output mask in the reference phase coordinate system that is then deformed to the target phase coordinate system for display.

Multi-phase object contour refinement

Methods and systems are provided for detecting focal lesions in multi-phase or multi-sequence medical imaging studies of medical images. One system includes memory for storing medical images and an electronic processor. The electronic processor is configured to: detect native candidate lesions for first phase computer tomography (CT) scans, detect candidate lesions for second phase CT scans, and project the candidate lesions detected for the first phase CT scans and/or the second phase CT scans on a same common domain. Once the candidate lesions are registered together, the electronic processor performs a matching algorithm for each of the native candidate lesions of the first phase CT scans with the registered candidate lesions from the second phase CT scans to determine presence and contours of valid lesions for the medical images, and to discard candidate lesions that are not acceptable.

Methods and systems for detecting focal lesions in multi-phase or multi-sequence medical imaging studies

A mechanism is provided for seed relabeling for seed-based slice-wise lesion segmentation. The mechanism receives a lesion mask for a three-dimensional medical image volume. The lesion mask corresponds to detected lesions in the medical image volume and wherein each detected lesion has a lesion contour. The mechanism generates a distance map for a given two-dimensional slice in the medical image volume based on the lesion mask. The distance map comprises a distance to a lesion contour for each voxel of the given two-dimensional slice. The mechanism performs local maxima identification to select a set of local maxima from the distance map such that each local maximum has a value greater than its immediate neighbor points. The mechanism performs seed relabeling based on the distance map and the set of local maxima to generate a set of seeds. Each seed represents a center of a distinct component of a lesion contour. The mechanism performs image segmentation on the lesion mask based on the set of seeds to form a split lesion mask.

Seed relabeling for seed-based segmentation of a medical image

A mechanism is provided to implement a trained machine learning computer model for determining z-wise lesion connectivity. The mechanism identifies, for a given slice in a three-dimensional medical image, a first lesion in the given slice and a second lesion in an adjacent slice in the three-dimensional medical image. The mechanism determines a first intersect value between the first lesion and the second lesion with respect to the first lesion and determines a second intersect value between the first lesion and the second lesion with respect to the second lesion. The mechanism determines whether the first lesion and the second lesion belong to the same three-dimensional lesion based on the first and second intersect values.

Logistic model to determine 3D z-wise lesion connectivity

A lesion detection and classification artificial intelligence (AI) pipeline comprising a plurality of trained machine learning (ML) computer models is provided. First ML model(s) process an input volume of medical images (VOI) to determine whether VOI depicts a predetermined amount of an anatomical structure. The AI pipeline determines whether criteria, such as a predetermined amount of an anatomical structure of interest being depicted in the input volume, are satisfied by output of the first ML model(s). If so, lesion processing operations are performed including: second ML model(s) processing the VOI to detect lesions which correspond to the anatomical structure of interest; third ML model(s) performing lesion segmentation and combining of lesion contours associated with a same lesion; and fourth ML models processing the listing of lesions to classify the lesions. The AI pipeline outputs the listing of lesions and the classifications for downstream computing system processing.

Lesion detection artificial intelligence pipeline computing system

A mechanism is provided in a data processing system for refining lesion contours with combined active contour and inpainting. The mechanism receives an initial segmented medical image having organ tissue including a set of object contours and a contour to be refined. The mechanism inpaints object voxels inside all contours of the set. The mechanism calculates an updated contour around the contour to be refined based on the in-painted object voxels to form an updated segmented medical image. The mechanism determines whether the updated segmented medical image is improved compared to the initial segmented medical image. The mechanism keeps the updated segmented medical image responsive to the updated segmented medical image being improved.

Refining lesion contours with combined active contour and inpainting

A lesion detection and classification artificial intelligence (AI) pipeline comprising a plurality of trained machine learning (ML) computer models is provided. First ML model(s) process an input volume of medical images (VOI) to determine whether VOI depicts a predetermined amount of an anatomical structure. The AI pipeline determines whether criteria, such as a predetermined amount of an anatomical structure of interest being depicted in the input volume, are satisfied by output of the first ML model(s). If so, lesion processing operations are performed including: second ML modal(s) processing the VOI to detect lesions which correspond to the anatomical structure of interest; third ML model(s) performing lesion segmentation and combining of lesion contours associated with a same lesion; and fourth ML models processing the listing of lesions to classify the lesions. The AI pipeline outputs the listing of lesions and the classifications for downstream computing system processing.

Growing community of inventors

Paris, France

Yi-Qing Wang