Methods and systems for implant identification using imaging data
Inventors
SUPRONO, Montry • Walter, Robert
Assignees
Publication Number
US-11080554-B2
Publication Date
2021-08-03
Expiration Date
2037-12-19
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Embodiments provide techniques, including systems and methods, for processing imaging data to identify an installed component. Embodiments include a component identification system that is configured to receive imaging data including an installed component, extract features of the installed component from the imaging data, and search a data store of components for matching reference components that match those features. A relevance score may be determined for each of the reference components based on a similarity between the image and a plurality of reference images in a component model of each of the plurality of reference components. At least one matching reference component may be identified by comparing each relevance score to a threshold relevance score and matching component information may be provided to an end-user for each matching reference component.
Core Innovation
The invention provides a system and method for identifying installed components, such as dental implants, using imaging data. The core technique involves receiving imaging data containing an installed component, extracting features from that imaging data, and then searching a database of reference components for those with matching features. The system determines a relevance score for each reference component based on the similarity between the imaging data and a set of reference images within a model for each reference component. Components with relevance scores exceeding a threshold are identified as matching, and information about these matches is provided to the end-user.
The problem addressed by the invention is the difficulty of identifying dental implants from X-rays due to limited and inconsistent imaging angles, as well as the vast variety of implant types and features available on the market. Conventional two-dimensional radiographs often fail to capture distinguishing features, making it time-intensive and sometimes impossible to conclusively identify installed implants.
The system utilizes feature extraction and image matching that operate across various angles and orientations of the implant, using reference models constructed from manufacturer data or previously classified images. The matching process is enhanced in efficiency by restricting comparisons to only relevant candidate components with extracted features similar to those in the imaging data. Additional processes allow for user clarification by providing guidance to capture images from perspectives that help resolve matches in ambiguous cases, and for updating component models using end-user feedback on match quality.
Claims Coverage
The patent contains multiple independent claims covering key inventive features related to automated identification of installed components using imaging data, relevance score computation, and feedback-driven model updates.
Automated component identification using imaging data and feature extraction
A system receives imaging data with an installed component, extracts a plurality of features from the imaging data, identifies a set of reference components matching those features from a database, and computes relevance scores for each by comparing the imaging data to reference images in a model for each component. Only those reference components with relevance scores above a threshold are identified as matches.
Similarity-based relevance scoring using reference image models
For each reference component, the method determines a relevance score by comparing the installed component in the image to a plurality of reference images, identifying the closest matching reference image for each, and calculating a similarity metric between the installed component and the reference component in the closest match.
Comprehensive system and device for interactive identification and updating of component models
The claims encompass a system comprising a component identification system and end-user computing device. The system can not only perform matching and provide component information, but also receive user selections for confirmed matches and update the component models to include new imaging data as reference images, enabling continual improvement in identification accuracy.
Machine learning classifier integration for enhanced component recognition
A component classifier algorithm, trained using a machine learning algorithm on classified reference images, is applied to the imaging data to classify the installed component based on size, placement, and shape. Resultant fit or match scores are used to determine relevance scores for candidate reference components, improving matching reliability.
In summary, the inventive features cover automated imaging-based component identification utilizing database reference models, relevance score computation using image similarity, user-driven clarification and model updating, and the integration of machine learning classifiers for improved recognition.
Stated Advantages
Enables efficient and accurate identification of installed implants from imaging data, regardless of orientation or angle.
Reduces the need for multiple X-rays, thereby decreasing patient exposure to radiation and saving time for healthcare providers.
Improves system efficiency by comparing only relevant candidates, conserving resources and increasing processing speed.
Allows for continual improvement in identification accuracy through end-user feedback and model updates.
Documented Applications
Identification of installed dental implants using radiographic or other imaging data.
Application in hospitals and medical fields for component identification in patients.
Ordering replacement implants online after identification of a matching piece of hardware.
Interested in licensing this patent?