Thyroid Dataset Child
Description
This dataset contains 289,200 anonymized patient records developed to support research in machine learning, data mining, clinical decision support systems, and feature selection methodologies for thyroid disease screening and risk prediction. The dataset was designed to represent a diverse population with varying demographic, clinical, laboratory, and symptom-related characteristics commonly associated with thyroid disorders. The primary objective of this dataset is to facilitate the development, evaluation, and comparison of predictive models capable of identifying individuals at risk of thyroid dysfunction based on routinely available clinical information. The dataset is particularly suitable for studies involving classification, feature selection, model interpretability, ensemble learning, explainable artificial intelligence (XAI), and healthcare analytics. Each record consists of 15 variables, including demographic attributes, anthropometric measurements, thyroid hormone laboratory results, clinical symptoms, family history, autoimmune indicators, and diagnostic outcomes. The variables include: Age, Gender, Body Mass Index (BMI), Thyroid Stimulating Hormone (TSH), Free Thyroxine (FT4), Free Triiodothyronine (FT3), Fatigue, Tremor, Anxiety, Dry Skin, Family History, Cholesterol, Thyroid Peroxidase Antibodies (TPO Antibodies), and the target variable Diagnosis. The diagnosis variable indicates the presence or absence of thyroid disease and serves as the classification label for predictive modeling tasks.
Files
Institutions
- Bandung Institute of TechnologyWest Java, Bandung