A Survey on Deep Learning-based Eardrum Segmentation

Abstract

Eardrum segmentation is an important part of the field of medical image segmentation, aiming to help doctors make accurate diagnoses and reduce their workload. Computer tech- niques have been widely used in this scenario. Due to the rapid development of deep learning, it has been replacing traditional computer vision methods and is widely used for segmenting medical images, several models have been used. In this paper, we investigate and analyze recent research related to deep learning on the eardrum segmentation scenario, introduce some basic background knowledge, and classify models, algorithms, and methods. We group the research by data feature. Finally, we speculate on the future direction of development and outlook.