A Deep Neural Network-Based Method for Prediction of Dementia Using Big Data

Abstract

The rise in dementia among the aging Korean population will quickly create a financial burden on society, but timely recognition of early warning for dementia and proper responses to the occurrence of dementia can enhance medical treatment. Health behavior and medical service usage data are relatively more accessible than clinical data, and a prescreening tool with easily accessible data could be a good solution for dementia-related problems. In this paper, we apply a deep neural network (DNN) to prediction of dementia using health behavior and medical service usage data, using data from 7031 subjects aged over 65 collected from the Korea National Health and Nutrition Examination Survey (KNHANES) in 2001 and 2005. In the proposed model, principal component analysis (PCA) featuring and min/max scaling are used to preprocess and extract relevant background features. We compared our proposed methodology, a DNN/scaled PCA, with five well-known machine learning algorithms. The proposed methodology shows 85.5% of the area under the curve (AUC), a better result than that using other algorithms. The proposed early prescreening method for possible dementia can be used by both patients and doctors.

Keywords: deep learning; deep neural network; dementia; feature extraction; prediction; principal component analysis.

Figure 1

The data selection of the study population from KNHANES.

Figure 2

The Flow Chart of the proposed DNN/scaled PCA approach.

Figure 3

Diverse Feature scaling plots with the PCA: class 0 is non-dementia patients (blue triangle) and class 1 is dementia patients (red rectangular): (a) 10th and 11th PCA with quantile transformer scaler; (b) 17th and 18th PCAs with min/max scaler; (c) 19th and 20th PCAs with min/max Scaler; (d) 9th and 20th PCAs with standard scaler; (e) 8th and 9th PCAs without scaler; (f) 12th and 13th PCAs without scaler.

Figure 4

The percentage of variance in PCA-min/max-transformer scaler.

Figure 5

The architecture of the proposed DNN.

Figure 6

Comparison of ROC curves of six classification methods.