Machine learning detection and classification of maxillofacial bone lesions in CBCT

A computer-implemented method comprising: receiving a plurality of CBCT scans, each comprising a series of axial slices, wherein the CBCT scans are associated with a cohort of subjects comprising a first subgroup of subjects having each one or more maxillofacial bone lesions, and a second subgroup o...

Full description

Saved in:
Bibliographic Details
Main Authors Aslan, Ragda Abdalla, Nadler, Chen, Yeshua, Talia, Leichter, Itzhak
Format Patent
LanguageEnglish
Published 26.12.2023
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:A computer-implemented method comprising: receiving a plurality of CBCT scans, each comprising a series of axial slices, wherein the CBCT scans are associated with a cohort of subjects comprising a first subgroup of subjects having each one or more maxillofacial bone lesions, and a second subgroup of subjects having no maxillofacial bone lesions; applying a feature extraction operation to extract a set of features from the axial slices in each of the CBCT scans; at a training stage, training a machine learning model on a training dataset comprising: (i) all of the extracted sets of features, and (ii) annotations indicating boundaries of bone lesions in the axial slices, to obtain a trained machine learning model configured to detect and segment a bone lesion in an axial slice from a CBCT scan.
Bibliography:Application Number: US202318312260