Dark Mode
Diabetes Patient Health Records
Patient Health Records & Digital Health
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset provides a straightforward collection of patient records, specifically curated for classification problems related to diabetes. It encompasses over one hundred individual patient entries, each detailing a variety of health-related metrics. The primary purpose is to offer a clear and accessible resource for developing models that predict diabetes diagnoses based on clinical and lifestyle factors.
Columns
- Age: The patient's age, measured in years. Values typically range from 12 to 75 years, with an average age of 42.
- Gender: Identifies the patient's gender, categorised as either male or female. The distribution shows 53% male and 47% female patients.
- BMI: Represents the patient's Body Mass Index, indicating weight relative to height. The mean BMI is approximately 35.4.
- Blood pressure: Records the patient's blood pressure, given in mmHg. Categories include High (62%), Normal (30%), and Other (8%).
- FBS: Denotes the patient's fasting blood sugar level, in mg/dL. The average FBS is 163 mg/dL.
- HbA1c: Measures the patient's haemoglobin A1c, reflecting average blood sugar control over the past three months. The mean HbA1c is about 7.89.
- Family history of diabetes: A boolean indicator specifying whether the patient has a family history of diabetes. 40% of records indicate a family history.
- Smoking: A boolean indicator showing whether the patient smokes. 62% of patients are recorded as smokers.
- Diet: Describes the patient's dietary habits, categorised as either poor or healthy. 62% of patients report a poor diet.
- Exercise: Indicates whether the patient exercises regularly. 62% of patients do not exercise regularly.
- Diagnosis: The patient's final diagnosis, indicating either diabetes or no diabetes. 24% of patients in the dataset have a diabetes diagnosis.
Distribution
The dataset is provided in a CSV format, specifically named 'Diabetes Classification.csv', and is approximately 6.18 kB in size. It contains 128 valid patient records, structured across 11 distinct columns. There are no missing or mismatched values reported for any of the columns, ensuring data integrity.
Usage
This dataset is ideally suited for:
- Developing and testing machine learning models for diabetes classification.
- Identifying key risk factors associated with diabetes onset.
- Educational purposes in data science and healthcare analytics.
- Research into predictive healthcare solutions.
Coverage
The dataset focuses on demographic information of patients, including age (12-75 years), gender, BMI, and various health markers related to diabetes risk. Geographic and time range details are not specified within the available sources.
License
Attribution 4.0 International (CC BY 4.0).
Who Can Use It
- Data Scientists and Machine Learning Engineers: To build, train, and evaluate classification models.
- Healthcare Researchers: For studying correlations between lifestyle factors, clinical measurements, and diabetes.
- Students and Educators: As a practical case study for learning about medical datasets and classification algorithms.
- Public Health Analysts: To understand population-level health trends and risk profiles related to diabetes.
Dataset Name Suggestions
- Diabetes Patient Health Records
- Diabetic Patient Classification Data
- Healthcare Diabetes Prediction Dataset
- Patient Diabetes Status Dataset
- Clinical Diabetes Risk Factors Dataset
Attributes
Original Data Source: Diabetes Patient Health Records
QUALITY