Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Attention-Based Deep Learning for Early Parkinson's Disease Detection with Tabular Biomedical Data

Olamide Samuel Oseni, Ibraheem Omotolani Obanla, Toheeb Aduramomi Jimoh

TL;DR

This work tackles early Parkinson's disease detection from structured tabular data by benchmarking four models—MLP, Gradient Boosting, TabNet, and SAINT—on a public voice-based PD dataset from UCI. It demonstrates that SAINT's dual attention mechanism (intra- and inter-sample) delivers superior discriminative performance, achieving high weighted metrics and the strongest MCC and AUC-ROC among the evaluated models. The study provides evidence for the applicability of attention-based deep learning to clinical prediction tasks with tabular data and discusses the need for larger, multimodal datasets to improve generalizability and clinical utility. Overall, SAINT shows promise as a diagnostic aid for early PD, though validation on diverse datasets and considerations of interpretability remain essential for real-world deployment.

Abstract

Early and accurate detection of Parkinson's disease (PD) remains a critical challenge in medical diagnostics due to the subtlety of early-stage symptoms and the complex, non-linear relationships inherent in biomedical data. Traditional machine learning (ML) models, though widely applied to PD detection, often rely on extensive feature engineering and struggle to capture complex feature interactions. This study investigates the effectiveness of attention-based deep learning models for early PD detection using tabular biomedical data. We present a comparative evaluation of four classification models: Multi-Layer Perceptron (MLP), Gradient Boosting, TabNet, and SAINT, using a benchmark dataset from the UCI Machine Learning Repository consisting of biomedical voice measurements from PD patients and healthy controls. Experimental results show that SAINT consistently outperformed all baseline models across multiple evaluation metrics, achieving a weighted precision of 0.98, weighted recall of 0.97, weighted F1-score of 0.97, a Matthews Correlation Coefficient (MCC) of 0.9990, and the highest Area Under the ROC Curve (AUC-ROC). TabNet and MLP demonstrated competitive performance, while Gradient Boosting yielded the lowest overall scores. The superior performance of SAINT is attributed to its dual attention mechanism, which effectively models feature interactions within and across samples. These findings demonstrate the diagnostic potential of attention-based deep learning architectures for early Parkinson's disease detection and highlight the importance of dynamic feature representation in clinical prediction tasks.

Attention-Based Deep Learning for Early Parkinson's Disease Detection with Tabular Biomedical Data

TL;DR

This work tackles early Parkinson's disease detection from structured tabular data by benchmarking four models—MLP, Gradient Boosting, TabNet, and SAINT—on a public voice-based PD dataset from UCI. It demonstrates that SAINT's dual attention mechanism (intra- and inter-sample) delivers superior discriminative performance, achieving high weighted metrics and the strongest MCC and AUC-ROC among the evaluated models. The study provides evidence for the applicability of attention-based deep learning to clinical prediction tasks with tabular data and discusses the need for larger, multimodal datasets to improve generalizability and clinical utility. Overall, SAINT shows promise as a diagnostic aid for early PD, though validation on diverse datasets and considerations of interpretability remain essential for real-world deployment.

Abstract

Early and accurate detection of Parkinson's disease (PD) remains a critical challenge in medical diagnostics due to the subtlety of early-stage symptoms and the complex, non-linear relationships inherent in biomedical data. Traditional machine learning (ML) models, though widely applied to PD detection, often rely on extensive feature engineering and struggle to capture complex feature interactions. This study investigates the effectiveness of attention-based deep learning models for early PD detection using tabular biomedical data. We present a comparative evaluation of four classification models: Multi-Layer Perceptron (MLP), Gradient Boosting, TabNet, and SAINT, using a benchmark dataset from the UCI Machine Learning Repository consisting of biomedical voice measurements from PD patients and healthy controls. Experimental results show that SAINT consistently outperformed all baseline models across multiple evaluation metrics, achieving a weighted precision of 0.98, weighted recall of 0.97, weighted F1-score of 0.97, a Matthews Correlation Coefficient (MCC) of 0.9990, and the highest Area Under the ROC Curve (AUC-ROC). TabNet and MLP demonstrated competitive performance, while Gradient Boosting yielded the lowest overall scores. The superior performance of SAINT is attributed to its dual attention mechanism, which effectively models feature interactions within and across samples. These findings demonstrate the diagnostic potential of attention-based deep learning architectures for early Parkinson's disease detection and highlight the importance of dynamic feature representation in clinical prediction tasks.
Paper Structure (17 sections, 6 equations, 7 figures)

This paper contains 17 sections, 6 equations, 7 figures.

Table of Contents

  1. Introduction
  2. Related Works
  3. Machine Learning Approaches for Early PD Detection
  4. Deep Learning Approaches for Early PD Detection
  5. Attention Mechanisms in Deep Learning for Biomedical Data
  6. SAINT Model and Advances in Attention-based Models for Tabular Data
  7. Research Gap and Motivation for this Study
  8. Materials and Methods
  9. Dataset
  10. Exploratory Data Analysis
  11. Description of Methods
  12. Techniques
  13. Model Evaluation
  14. Evaluation Metrics
  15. Results and Interpretations
  16. ...and 2 more sections

Figures (7)

  • Figure 1: Proposed PD early detection modelling framework
  • Figure 2: Exploratory analysis of dataset features.
  • Figure 3: SAINT Architecture, including pre-training and training pipeline introduced by saint
  • Figure 4: Pipeline for Early Parkinson’s Disease Detection
  • Figure 5: Comparison of performance metrics for different models.
  • ...and 2 more figures