A non-invasive, automated diagnosis of Meni\xe8re\u2019s disease using radiomics and machine learning on conventional magnetic resonance imaging: A multicentric, case-controlled feasibility study

Philippe Lambin,Stephanie Vanden Bossche,Marly F. J. A. van der Lubbe,Akshayaa Vaidyanathan,Marc van Hoof,Tjasse D. Bruintjes,Alida A. Postma,Elske L. van den Burg,Raymond van de Berg,Monique A. L. Bilderbeek-Beckers,Patrick F. M. Dammeijer,Marjolein de Wit,Vincent Van Rompaey

doi:10.1007/s11547-021-01425-w

Abstract

PurposeThis study investigated the feasibility of a new image analysis technique (radiomics) on conventional MRI for the computer-aided diagnosis of Menière’s disease.Materials and methodsA retrospective, multicentric diagnostic case–control study was performed. This study included 120 patients with unilateral or bilateral Menière’s disease and 140 controls from four centers in the Netherlands and Belgium. Multiple radiomic features were extracted from conventional MRI scans and used to train a machine learning-based, multi-layer perceptron classification model to distinguish patients with Menière’s disease from controls. The primary outcomes were accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the classification model.ResultsThe classification accuracy of the machine learning model on the test set was 82%, with a sensitivity of 83%, and a specificity of 82%. The positive and negative predictive values were 71%, and 90%, respectively.ConclusionThe multi-layer perceptron classification model yielded a precise, high-diagnostic performance in identifying patients with Menière’s disease based on radiomic features extracted from conventional T2-weighted MRI scans. In the future, radiomics might serve as a fast and noninvasive decision support system, next to clinical evaluation in the diagnosis of Menière’s disease.

Highlights

Menière’s disease (MD) is a multifactorial condition of the inner ear characterized by recurrent episodes of vertigo and fluctuating aural symptoms like hearing loss, aural fullness, and tinnitus
MD is strongly associated with the classical histological finding known as endolymphatic hydrops (EH), which is a distention of the endolymphatic compartment of the labyrinth [1]
To further explore the application of radiomics, the objective of this study was to develop a computer-aided diagnostic tool for MD by using a radiomics approach combined with machine learning (ML)

Summary

Introduction

Menière’s disease (MD) is a multifactorial condition of the inner ear characterized by recurrent episodes of vertigo and fluctuating aural symptoms like hearing loss, aural fullness, and tinnitus. MD is strongly associated with the classical histological finding known as endolymphatic hydrops (EH), which is a distention of the endolymphatic compartment of the labyrinth [1]. The consistent finding of EH in temporal. Extended author information available on the last page of the article bones of patients with MD [2, 3] led to defining EH as the pathological basis of MD. It marked the beginning of a diagnostic challenge, as EH could only be identified post-mortem. MD remained a clinical diagnosis, and different symptom-based classification methods emerged over time [4]

Objectives

Methods

Findings

Discussion

Conclusion