Diagnosis of Middle Ear Diseases Based on Convolutional Neural Network

An otoscope is traditionally used to examine the eardrum and ear canal. A diagnosis of otitis media (OM) relies on experience clinicians. If an examiner lacks experience, examination may be difficult time-consuming. This paper presents disease classification method using middle images based a convolutional neural network (CNN). Especially segmentation networks are classify otoscopic image into six classes: normal, acute (AOM), with effusion (OME), chronic (COM), congenital cholesteatoma (CC) traumatic perforations (TMPs). The Mask R-CNN utilized for extract region interest (ROI) from images. extracted ROIs as guiding features classification. transfer learning ensemble two CNN classifiers: EfficientNetB0 Inception-V3. proposed model was trained 5-fold cross-validation technique. evaluated achieved accuracy 97.29%.