Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Predictive Analytics for Dementia: Machine Learning on Healthcare Data

Shafiul Ajam Opee, Nafiz Fahad, Anik Sen, Rasel Ahmed, Fariha Jahan, Md. Kishor Morol, Md Rashedul Islam

TL;DR

Predictive dementia analytics are addressed by comparing supervised ML classifiers (KNN, LDA, QDA, Gaussian Process) with SMOTE and TF-IDF preprocessing on a rich health dataset, plus a hybrid CNN-RNN model for multi-modal data. LDA achieves the highest generalization with a testing accuracy of 98%, supported by strong associations to APOE-e4 and diabetes, while interpretability is highlighted as essential for clinical adoption. The study demonstrates the value of feature engineering and data balancing for dementia prediction and explores a hybrid architecture to fuse numeric and textual information. Overall, the work advances interpretable ML approaches for early dementia risk stratification and points to explainable AI and broader data integration as key future directions.

Abstract

Dementia is a complex syndrome impacting cognitive and emotional functions, with Alzheimer's disease being the most common form. This study focuses on enhancing dementia prediction using machine learning (ML) techniques on patient health data. Supervised learning algorithms are applied in this study, including K-Nearest Neighbors (KNN), Quadratic Discriminant Analysis (QDA), Linear Discriminant Analysis (LDA), and Gaussian Process Classifiers. To address class imbalance and improve model performance, techniques such as Synthetic Minority Over-sampling Technique (SMOTE) and Term Frequency-Inverse Document Frequency (TF-IDF) vectorization were employed. Among the models, LDA achieved the highest testing accuracy of 98%. This study highlights the importance of model interpretability and the correlation of dementia with features such as the presence of the APOE-epsilon4 allele and chronic conditions like diabetes. This research advocates for future ML innovations, particularly in integrating explainable AI approaches, to further improve predictive capabilities in dementia care.

Predictive Analytics for Dementia: Machine Learning on Healthcare Data

TL;DR

Predictive dementia analytics are addressed by comparing supervised ML classifiers (KNN, LDA, QDA, Gaussian Process) with SMOTE and TF-IDF preprocessing on a rich health dataset, plus a hybrid CNN-RNN model for multi-modal data. LDA achieves the highest generalization with a testing accuracy of 98%, supported by strong associations to APOE-e4 and diabetes, while interpretability is highlighted as essential for clinical adoption. The study demonstrates the value of feature engineering and data balancing for dementia prediction and explores a hybrid architecture to fuse numeric and textual information. Overall, the work advances interpretable ML approaches for early dementia risk stratification and points to explainable AI and broader data integration as key future directions.

Abstract

Dementia is a complex syndrome impacting cognitive and emotional functions, with Alzheimer's disease being the most common form. This study focuses on enhancing dementia prediction using machine learning (ML) techniques on patient health data. Supervised learning algorithms are applied in this study, including K-Nearest Neighbors (KNN), Quadratic Discriminant Analysis (QDA), Linear Discriminant Analysis (LDA), and Gaussian Process Classifiers. To address class imbalance and improve model performance, techniques such as Synthetic Minority Over-sampling Technique (SMOTE) and Term Frequency-Inverse Document Frequency (TF-IDF) vectorization were employed. Among the models, LDA achieved the highest testing accuracy of 98%. This study highlights the importance of model interpretability and the correlation of dementia with features such as the presence of the APOE-epsilon4 allele and chronic conditions like diabetes. This research advocates for future ML innovations, particularly in integrating explainable AI approaches, to further improve predictive capabilities in dementia care.
Paper Structure (29 sections, 10 equations, 13 figures, 2 tables)

This paper contains 29 sections, 10 equations, 13 figures, 2 tables.

Table of Contents

  1. INTRODUCTION
  2. LITERATURE REVIEW
  3. Supervised Learning Techniques
  4. Deep Learning and Hybrid Approaches
  5. Key Insights from Existing Work
  6. METHODOLOGY
  7. Data Collection
  8. Proposed Architecture
  9. Data Preprocessing
  10. Handling Missing Values
  11. Encoding Categorical Variables
  12. Smote Oversampling
  13. Feature Engineering
  14. TF-IDF Vectorization
  15. Scaling Numerical Features
  16. ...and 14 more sections

Figures (13)

  • Figure 1: Proposed Architecture
  • Figure 2: Methodological Description of Hybrid Neural Network
  • Figure 3: Correlation between Dementia and Features.
  • Figure 4: Family history of dementia in our dataset.
  • Figure 5: Dementia Patients' Chronic Conditions.
  • ...and 8 more figures