The source code for Towards Sustainable Compressive Population Health: A GAN-based Year-By-Year Imputation Method
Published on ACM Transactions on Computing for Healthcare 2022
Thank you for your interest in our work, we have uploaded the Section 4 DATA OBSERVATION AND INTUITION and all the code for the model here.
Install python, tensorflow. We use Python 3.7, Tensorflow 1.14.0. See requirements.txt
The three open-world Chronic Diseases Prevalence Datasets can be downloaded at:
- UK-Obesity Dataset: https://digital.nhs.uk/data-and-information/publications/statistical/quality-and-outcomes-framework-achievement-prevalence-and-exceptions-data
- US-Hypertension Dataset: https://www.cdc.gov/500cities/
- Taiwan-Diabetes Dataset: https://dep.mohw.gov.tw/DOS/cp-2519-3480-113.html
The "DATA" folder contains the downloaded raw data set and the pre-processed normalised data.
All the hyper-parameters and steps are included in the ./EXPERIMENTS/UAA-GAIN/main.py file, you can run it directly.
All other baseline methods are also in the "EXPERIMENTS" folder.
Due to space limitations in the paper, it is not possible to show the results of the experiments for all years. Here we show all the results of the experiments in graphs：
- UK-Obesity results：
- US-Hypertension results：
- Taiwan-Diabetes results：
@article{feng2022towards,
title={Towards Sustainable Compressive Population Health: A GAN-based Year-By-Year Imputation Method},
author={Feng, Yujie and Wang, Jiangtao and Wang, Yasha and Chu, Xu},
journal={ACM Transactions on Computing for Healthcare},
year={2022},
publisher={ACM New York, NY}
}