Classification of Otitis Media Images
- Detailed Technology Description
- None
- *Abstract
-
BackgroundManual classification of biological and biomedical images by experts is time consuming, prone to subjectivity, and depicts limited intra and inter observer reproducibility. Hence there is a need for developing an accurate, efficient, automated system for the classification of biomedical images. The inventors have considered the application of the classification problem in detecting the stages of otitis media in images. Thus automated classification of otitis media images will aid in the proper diagnosis of the different stages of otitis media, therefore reducing the over-treatment of improperly diagnosed patients. TechnologyThe system integrates multiresolution techniques used to decompose the input images, feature extraction, the classification system, and the weighting algorithm. The goal is to distinguish between three classes of ear conditions: Normal, AOM, and OME. The Normal class represents a healthy ear. AOM is defined as the presence of fluid in the middle ear, in association with symptoms and signs of acute illness such as otalgia, otorrhea and fever. OME is defined as the presence of fluid in the middle ear, in the absence of signs or symptoms of acute infection. Application1. Enhanced capability to diagnosis otitis mediaAdvantages1. Classification works in conjunction with the digital otoscope2. Allows less subjectivity in the diagnosis3. System provides immediate feedback to the clinicianStage of DevelopmentSoftware and algorithms developedCopyright
- *Principal Investigator
-
Name: Alejandro Hoberman
Department: Med-Pediatrics
Name: Jelena Kovacevic
Department: Electrical & Computer Engineering
- Country/Region
- USA
For more information, please click Here