The Use of Machine Learning Tools to Analyse Wideband Absorbance Immittance in Normal and Ears With Otitis Media With Effusion

Abstract

Background: Wideband Absorbance Immittance (WAI) has been available for more than a decade, however its clinical use still faces the challenges of limited understanding and poor interpretation of WAI results. This study aimed to develop Machine Learning (ML) tools to identify the WAI absorbance characteristics across different frequency-pressure regions in the normal middle ear and ears with otitis media with effusion (OME) to enable diagnosis of middle ear conditions automatically. Methods: A total of 672 sets of WAI sample data (423 normal middle ears and 249 ears with OME) were collected using the Titan IMP440. Data analysis included pre-processing of the WAI data, statistical analysis and classification model development, together with significant region extraction from the 2D frequency-pressure WAI images. Findings: Statistical analysis at each frequency-pressure point showed 88% of the data points to have significant differences in absorbance values between normal ears and ears with OME. The performance of the examined classifiers to automatically diagnose the OME and normal cases showed the accuracy at around 80%. Feature selection using RF classifiers and tests of statistical significance identified an area of importance at frequencies between 1090Hz to 2310 Hz and pressures from -40 to +90 daPa from the 2D frequency-pressure WAI images. Interpretation: ML tools appear to hold great potential for the automated diagnosis of middle ear diseases from WAI data. The identified key regions in the WAI provide guidance to practitioners to better understand and interpret WAI data and offer the prospect of quick and accurate diagnostic decisions. Funding Statement: This work is supported by Ser Cymru III Enhancing Competitiveness Infrastructure Award (MA/KW/5554/19), Great Britain Sasakawa Foundation (5826), Cardiff Metropolitan University Research Innovation Award and The Global Academies Research and Innovation Development Fund. Declaration of Interests: None declared. Ethics Approval Statement: The study was approved by Cardiff School of Sport and Health Sciences under the Cardiff Metropolitan University Ethics Framework (Ethical reference number: Sta-3013).