Dark Mode
Renal Disease Risk Factors Bangladesh
Patient Health Records & Digital Health
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
the prediction of Chronic Kidney Disease (CKD). It delves into a significant medical challenge where renal capacities decline, leading to kidney damage. The dataset originates from real patient information collected in Bangladesh, specifically from Enam Medical College in Savar, Dhaka. It provides a valuable resource for understanding CKD risk factors and developing predictive models. It is important to note that the raw data has not undergone any preprocessing, so initial data preparation will be necessary before applying machine learning algorithms.
Columns
The dataset contains 29 distinct columns, capturing a variety of clinical and demographic indicators:
- bp (Diastolic): Diastolic blood pressure readings, with common values observed.
- bp limit: Blood pressure limits.
- sg: Specific gravity of urine, indicating kidney's ability to concentrate urine.
- al: Albumin levels in urine, a marker for kidney damage.
- class: Categorical variable indicating whether a patient has CKD or not.
- rbc: Red blood cell presence in urine.
- su: Sugar levels in urine.
- pc: Pus cell presence in urine.
- pcc: Pus cell clump presence in urine.
- ba: Bacteria presence in urine.
- bgr: Blood glucose random measurements.
- bu: Blood urea levels.
- sod: Sodium levels in blood.
- sc: Serum creatinine levels, an indicator of kidney function.
- pot: Potassium levels in blood.
- hemo: Haemoglobin levels.
- pcv: Packed cell volume.
- rbcc: Red blood cell count.
- wbcc: White blood cell count.
- htn: Hypertension (high blood pressure) status.
- dm: Diabetes mellitus status.
- cad: Coronary artery disease status.
- appet: Appetite status.
- pe: Pedal oedema (swelling of feet/ankles) status.
- ane: Anaemia status.
- grf: Glomerular filtration rate, a key measure of kidney function.
- stage: The stage of CKD progression.
- affected: Indicates if the patient is affected by CKD.
- age: The age of the patient.
Distribution
The dataset is provided in a CSV format, with a file size of 34.15 kB. It consists of 29 columns and approximately 202 records. Most variables have 201 valid entries, with a minimal number of missing values (typically 1). The 'affected' and 'age' columns have 202 valid entries with no missing data. There are no anticipated updates to this dataset.
Usage
This data is ideal for developing machine learning algorithms for the early detection and risk prediction of CKD. It can be utilised in medical research to identify significant risk factors, for training predictive models to assist in diagnosis, and in health analytics to understand disease patterns within the Bangladeshi population.
Coverage
The data provides insights into Bangladeshi patients, having been collected from a medical college in Dhaka, Bangladesh. The specific time range of data collection is not detailed, but it represents real patient information from that geographic region.
License
Attribution 4.0 International (CC BY 4.0) License.
Who Can Use It
This data is suitable for researchers, data scientists, healthcare analysts, and students interested in medical diagnostics and predictive health modelling. It offers a foundation for developing AI-driven tools to combat chronic kidney disease, for studying disease epidemiology, and for public health initiatives aimed at kidney health.
Dataset Name Suggestions
- Bangladeshi CKD Patient Data
- Renal Disease Risk Factors Bangladesh
- Dhaka CKD Patient Metrics
- Kidney Function Prediction Dataset
- Chronic Kidney Disease Analytics (Bangladesh)
Attributes
Original Data Source: Renal Disease Risk Factors Bangladesh
QUALITY