Synthetic Thyroid Healthcare Dataset : A Large-Scale Multidimensional Resource for Thyroid Disease Analysis, Risk Assessment and Clinical Prediction
Description
This data is a synthetic dataset of thyroid healthcare with 125 features, 34 binary, 28 categorical and 97 numerical, with 1 million data points, not AI generated but the synthetic generation process was held on the basis of probabilistic rules . It includes a variety of demographic, family, lifestyle, environmental, clinical, endocrine-metabolic, inflammatory, treatment, health care utilization and outcome-related characteristics. This dataset contains 10 thyroid phenotypes : Healthy, Hashimoto, Graves, overt hypo and hyperthyroidism, benign nodules, and mixed disorders. It also encompasses disease severity, the latent clinical states, symptom burden, treatment response, adherence, future stage of disease, healthcare utilization and expected clinical outcomes. The dataset is synthetic and is used for research within the fields of thyroid disease prediction, risk assessment, clinical outcome modeling, health care analytics, machine learning, deep learning and explainable AI.
Files
Institutions
- Chittagong University of Engineering & TechnologyChittagong, Chittagong