IMPROVING STROKE PREDICTION ON IMBALANCED CLINICAL DATA USING CTGAN AND TVAE: A SYNTHETIC DATA APPROACH

Majid Liaquat; CD Nugent; I Cleland; Naveed Khan

Abstract

Synthetic data (SD) have been evaluated and adopted in different domains and areas, especially in health. To conduct this study, we chose tabular data on stroke prediction, available in [3]. The dataset contains 11 clinical features, including the
last column, positive = 1 and negative = 0 for stroke. We chose this dataset because of its imbalance, which will be a perfect fit for implementing the generating techniques to know how well the real data resemble these. For generating SD, we use two techniques known as conditional tabular GAN (CTGAN) and tabular variational autoencoder (TVAE), which have different numbers of epochs and batch sizes. We further evaluated the results with three Machine Learning (ML) models as a benchmark with real data. The results highlight data generated with CTGAN (epochs=1500, batch size=500) perform
better with an accuracy score of 0.995 on random forest (RF) and Support Vector Machine (SVM).

Original language	English
Pages	28
Number of pages	28
Publication status	Published (in print/issue) - 1 May 2025
Event	Northern Ireland Biomedical Engineering Society Annual Symposium (NIBES) 2025 - Queens University, Belfast, Northern Ireland - Queens University, Belfast, United Kingdom Duration: 1 May 2025 → 1 May 2025

Conference

Conference	Northern Ireland Biomedical Engineering Society Annual Symposium (NIBES) 2025 - Queens University, Belfast, Northern Ireland
Country/Territory	United Kingdom
City	Belfast
Period	1/05/25 → 1/05/25

Keywords

Synthetic Data generation
Imbalance data
Clinical data
Data Accuracy

Access to Document

NIBES 2025

Cite this

@conference{065adf6a744d4909a4268be1dece030b,

title = "IMPROVING STROKE PREDICTION ON IMBALANCED CLINICAL DATA USING CTGAN AND TVAE: A SYNTHETIC DATA APPROACH",

abstract = "Synthetic data (SD) have been evaluated and adopted in different domains and areas, especially in health. To conduct this study, we chose tabular data on stroke prediction, available in [3]. The dataset contains 11 clinical features, including the last column, positive = 1 and negative = 0 for stroke. We chose this dataset because of its imbalance, which will be a perfect fit for implementing the generating techniques to know how well the real data resemble these. For generating SD, we use two techniques known as conditional tabular GAN (CTGAN) and tabular variational autoencoder (TVAE), which have different numbers of epochs and batch sizes. We further evaluated the results with three Machine Learning (ML) models as a benchmark with real data. The results highlight data generated with CTGAN (epochs=1500, batch size=500) perform better with an accuracy score of 0.995 on random forest (RF) and Support Vector Machine (SVM).",

keywords = "Synthetic Data generation, Imbalance data, Clinical data, Data Accuracy",

author = "Majid Liaquat and CD Nugent and I Cleland and Naveed Khan",

year = "2025",

month = may,

day = "1",

language = "English",

pages = "28",

note = "Northern Ireland Biomedical Engineering Society Annual Symposium (NIBES) 2025 - Queens University, Belfast, Northern Ireland ; Conference date: 01-05-2025 Through 01-05-2025",

}

IMPROVING STROKE PREDICTION ON IMBALANCED CLINICAL DATA USING CTGAN AND TVAE: A SYNTHETIC DATA APPROACH. / Liaquat, Majid ; Nugent, CD ; Cleland, I et al.
2025. 28 Abstract from Northern Ireland Biomedical Engineering Society Annual Symposium (NIBES) 2025 - Queens University, Belfast, Northern Ireland, Belfast, United Kingdom.

Research output: Contribution to conference › Abstract

TY - CONF

T1 - IMPROVING STROKE PREDICTION ON IMBALANCED CLINICAL DATA USING CTGAN AND TVAE: A SYNTHETIC DATA APPROACH

AU - Liaquat, Majid

AU - Nugent, CD

AU - Cleland, I

AU - Khan, Naveed

PY - 2025/5/1

Y1 - 2025/5/1

N2 - Synthetic data (SD) have been evaluated and adopted in different domains and areas, especially in health. To conduct this study, we chose tabular data on stroke prediction, available in [3]. The dataset contains 11 clinical features, including the last column, positive = 1 and negative = 0 for stroke. We chose this dataset because of its imbalance, which will be a perfect fit for implementing the generating techniques to know how well the real data resemble these. For generating SD, we use two techniques known as conditional tabular GAN (CTGAN) and tabular variational autoencoder (TVAE), which have different numbers of epochs and batch sizes. We further evaluated the results with three Machine Learning (ML) models as a benchmark with real data. The results highlight data generated with CTGAN (epochs=1500, batch size=500) perform better with an accuracy score of 0.995 on random forest (RF) and Support Vector Machine (SVM).

AB - Synthetic data (SD) have been evaluated and adopted in different domains and areas, especially in health. To conduct this study, we chose tabular data on stroke prediction, available in [3]. The dataset contains 11 clinical features, including the last column, positive = 1 and negative = 0 for stroke. We chose this dataset because of its imbalance, which will be a perfect fit for implementing the generating techniques to know how well the real data resemble these. For generating SD, we use two techniques known as conditional tabular GAN (CTGAN) and tabular variational autoencoder (TVAE), which have different numbers of epochs and batch sizes. We further evaluated the results with three Machine Learning (ML) models as a benchmark with real data. The results highlight data generated with CTGAN (epochs=1500, batch size=500) perform better with an accuracy score of 0.995 on random forest (RF) and Support Vector Machine (SVM).

KW - Synthetic Data generation

KW - Imbalance data

KW - Clinical data

KW - Data Accuracy

UR - https://drive.google.com/file/d/1vkVJFHVzAMBb1BO3f9NXsvb5oVVktuWR/view

UR - https://sites.google.com/view/nibes/events/previous-symposia

M3 - Abstract

SP - 28

T2 - Northern Ireland Biomedical Engineering Society Annual Symposium (NIBES) 2025 - Queens University, Belfast, Northern Ireland

Y2 - 1 May 2025 through 1 May 2025

ER -

IMPROVING STROKE PREDICTION ON IMBALANCED CLINICAL DATA USING CTGAN AND TVAE: A SYNTHETIC DATA APPROACH

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint