Analysis of Surfaces Using Constrained Regression Models

Abstract

Abstract. We present a study of the relationship between the changes in the shape of the human ear due to jaw movement and acoustical feedback (AF) in hearing aids. In particular, we analyze the deformation field of the outer ear associated with the movement of the mandible (jaw bone) to understand its effect on AF and identify local regions that play a significant role. Our data contains ear impressions of 42 hearing aid users, in two different positions: open and closed mouth, and survey data including information about experienced discomfort due to AF. We use weighted support vector machines (WSVM) to investigate the separation between the presence and lack of AF and achieve classification accuracy of 80% based on the deformation field. To robustly localize the regions of the deformation field that significantly contribute to AF we employ logistic regression penalized with elastic net (EN). By visualizing the selected variables on the mean surface, we provide clinical interpretations of the results.

The tongue and ear interface

“…When the mouth is open, the Mandible Condyle slides forward from its location and creates a small void. The tissue surrounding the ear canal fills this void, expanding the ear canal [3] as illustrated in Figure 4a.…”

An Average of the Human Ear Canal: Recovering Acoustical Properties via Shape Analysis

“…One of the first in-vivo investigations of the shape of the ear-canal was published by Paulsen et al [6] who created the first artificial shape model of the human ear [7] based on the impressions of 29 subjects and expert annotations of the surfaces of the impressions. A more detailed model was presented by Darkner et al [8] and extended to a model of the dynamic behavior of the ear canal [9]. Both papers used a fully automated alignment procedure [10,11].…”

“…Further analysis and modeling of the shape of the ear canal has been carried out in [12,13,14] including attempts at building a model for use in automated production of hearing aids. Darkner et al [9] used the shape of the ear and a penalized regression model to correlate shape changes due to ear motion with the perceived level of comfort by the subject wearing the hearing aid, however with inconclusive results. It has been been debated whether it is possible to locate the exact transition of the canal-bone junction based on ear impression alone.…”

Use of magnetic resonance image registration to estimate displacement in the human earcanal due to the insertion of in-ear devices

“…Benacchio et al (2018) recently proposed a method based on a common image processing tool to assess the earcanal displacement field from medical images of both open and occluded earcanals. The aforementioned tool, called the image registration method, had already been used in the past to compute the deformation of the earcanal due to the opening of the mouth (Darkner et al, 2007;Darkner et al, 2008) and to obtain an average earcanal geometry extracted from 45 human earcanal magnetic resonance (MR) images (Darkner et al, 2017). Contrary to previous studies (Oliveira et al, 1992;Pirzanski, 2006) where the magnitude of the displacement was determined only at some specific anatomical landmarks or in planes corresponding to cross sections of the earcanal, the method proposed by Benacchio et al (2018) provides the displacement vector field (both magnitude and direction) of the entire earcanal wall.…”