Table 3 Comparison of recent studies on vestibular disorder classification

Vivar et al. (2021)

Development of Base-ml tool for classifying vestibular disorders

Multiple algorithms including Random Forest and deep learning models

Base-ml achieved accuracy up to 92.5% in classifying vestibular disorders

MD, BPPV, VM, PPPD

Ahmadi et al. (2020)

Differentiation between central and peripheral vestibular disorders

Linear and non-linear methods including Random Forest and deep learning models

Machine learning outperformed traditional diagnostic scores in acute vestibular syndromes

Stroke, VEST

Wang et al. (2024)

Differentiation between vestibular migraine and Menière’s disease

Multiple algorithms including AdaBoost and Random Forest

Achieved 85.77–97.81% accuracy in differentiating vestibular migraine and Menière’s disease

MD, VM

Strobl et al. (2021)

Identification of key variables for classifying vestibular disorders based on patient history

Classification trees for identifying key variables

Key variables identified for distinguishing vestibular disorders with almost 50% accuracy

MD, VM, BPPV, PPPD, vestibular paroxysmia

Raponi et al. (2024)

Use of AI for enhancing diagnosis and differentiation of vestibular disorders

Machine learning algorithms applied to anamnestic questionnaire data

AI models showed high accuracy in supporting diagnosis and differentiation of vestibular disorders

MD, BPPV, AVD, VM

This Study

AI-based classifier for 6 common vestibular disorders

CatBoost machine learning model

Achieved overall accuracy of 88.4%

MD, BPPV, VEST, HOD, VM, PPPD