What question did this study set out to answer?

This study aims to evaluate the effectiveness of a multimodal machine learning framework for detecting and staging Parkinson’s disease.

Stage classification and report generation of Parkinson’s disease using multimodal data and machine learning

View Full Paper

Ask AI

Bookmark

Authors

Exploratory multimodal machine learning framework detects and stages Parkinson’s disease, highlighting its diagnostic utility.

This study aims to evaluate the effectiveness of a multimodal machine learning framework for detecting and staging Parkinson’s disease.
Used gait, voice, and MRI data integrated through a machine learning framework.
Extracted biomechanical features from planar-pressure recordings and applied XGBoost for gait classification.
Utilized Wavelet Scattering Transforms for speech data analysis and focused on 30 brain region volumes for MRI.
The gait pipeline achieved a cross-validated classification rate of 90.15% using subject-wise partitioning.
The speech pipeline reported an exploratory accuracy of 95.5% but had limitations due to speaker-level leakage.
Multimodal aggregation reached about 92% accuracy under a majority voting scheme, demonstrating proof-of-concept potential.

Kshithij et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

View Full Paper

Ask AI

Bookmark

Authors

Exploratory multimodal machine learning framework detects and stages Parkinson’s disease, highlighting its diagnostic utility.

This study aims to evaluate the effectiveness of a multimodal machine learning framework for detecting and staging Parkinson’s disease.
Used gait, voice, and MRI data integrated through a machine learning framework.
Extracted biomechanical features from planar-pressure recordings and applied XGBoost for gait classification.
Utilized Wavelet Scattering Transforms for speech data analysis and focused on 30 brain region volumes for MRI.
The gait pipeline achieved a cross-validated classification rate of 90.15% using subject-wise partitioning.
The speech pipeline reported an exploratory accuracy of 95.5% but had limitations due to speaker-level leakage.
Multimodal aggregation reached about 92% accuracy under a majority voting scheme, demonstrating proof-of-concept potential.

Kshithij et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark