A Comprehensive Dengue Dataset of Bangladesh
Description
Description: This dataset includes epidemiological and clinical data from 1,000 anonymized individuals in Dhaka, Bangladesh, collected during dengue outbreaks. It covers demographic details, living conditions, WHO-aligned serological tests, CBC parameters, and clinical symptoms to support public health monitoring, risk analysis, and diagnostic modeling for early dengue triage. Data Collection and Ethical Methodology: Data were collected through a cross-sectional study in hospital wards and households across 36 municipal areas of Dhaka. Approved by Bangladesh University’s institutional review board, the study followed the Declaration of Helsinki guidelines. Written informed consent was obtained from adult participants and guardians of minors. To ensure privacy, all personal identifiers were removed and replaced with unique identification codes. Clinical Diagnosis and Laboratory Testing: Clinical diagnosis was verified via lab testing and physician evaluation. Testing included NS1 antigen for acute infection and IgG/IgM antibodies to distinguish primary and secondary dengue cases. Platelet and white blood cell counts were measured using automated hematology analyzers. Environmental variables, such as house type and area development, were recorded through visual inspection and standardized survey forms. Variables in the Dataset The dataset is provided as a comma-separated values file (dataset.csv) containing 1000 records and 19 columns with no missing entries: Gender: Biological sex (Male or Female) Age: Patient age in completed years (8 to 65) Area: Residential area within Dhaka (36 specific municipal locations) AreaType: Urban infrastructure level (Developed or Undeveloped) HouseType: Housing construction category (Building, Tinshed, Other) District: Administrative district (Dhaka) Fever_Duration: Duration of active fever in days (0 to 7) Body_Temperature: Patient body temperature in degrees Celsius (36.2 to 40.5) Platelet_Count: Total blood platelet count per microliter WBC_Count: Total white blood cell count per microliter Headache: Presence of acute headache (1 = Yes, 0 = No) Retro_Orbital_Pain: Pain behind the eyes (1 = Yes, 0 = No) Myalgia: Muscle aches and body pain (1 = Yes, 0 = No) Rash: Cutaneous rash or petechiae (1 = Yes, 0 = No) Joint_Pain: Severity of joint pain (Absent, Moderate, Severe) NS1: Dengue non-structural protein 1 antigen test (1 = Positive, 0 = Negative) IgG: Dengue IgG antibody test (1 = Positive, 0 = Negative) IgM: Dengue IgM antibody test (1 = Positive, 0 = Negative) Outcome: Final clinical diagnosis (1 = Confirmed Dengue, 0 = Non-Dengue) Potential Applications The dataset can be used by researchers, data scientists, and healthcare agencies for: 1. Machine learning models for dengue risk assessment using vital signs and blood markers. 2. Spatial analysis of transmission patterns across urban housing environments. 3. Clinical research on symptom profiles and laboratory thresholds for dengue diagnosis.
Files
Steps to reproduce
Steps to reproduce the data collection and curation process: Participant Screening and Ethical Approval Obtain institutional ethics approval from relevant review boards prior to initiating surveys. Screen patients presenting with acute febrile illness and suspected dengue symptoms at hospital outpatient and inpatient wards, as well as residents in surrounding community areas across urban Dhaka. Provide an informational study sheet and obtain written informed consent from adult participants and legal guardians of pediatric patients. Clinical Symptom and Vital Signs Assessment Conduct direct structured interviews using questionnaires adapted from World Health Organization dengue case definition forms. Measure and record body temperature using calibrated digital clinical thermometers in degrees Celsius. Record the number of consecutive days of active fever. Document the presence of common dengue-associated symptoms including headache, retro-orbital pain, myalgia, cutaneous rash, and joint pain categorized by severity (absent, moderate, or severe). Diagnostic Serology and Hematological Testing Collect blood samples following standard hospital phlebotomy protocols. Perform complete blood counts using calibrated automated hematology analyzers to obtain total blood platelet counts and white blood cell (WBC) counts per microliter. Perform serological testing using standard rapid diagnostic test (RDT) kits or enzyme-linked immunosorbent assays (ELISA) to determine NS1 antigen status (for acute viral presence) and IgG and IgM antibody status (for primary vs. secondary exposure classification). Final diagnosis outcome is confirmed based on combined clinical presentation and laboratory confirmation. Environmental and Housing Classification Record the participant residential location within municipal ward boundaries across Dhaka. Conduct on-site visual assessment of the patient household to classify the primary structural construction material (concrete building, tin-shed structure, or other semi-permanent materials). Classify the surrounding neighborhood infrastructure as developed or undeveloped based on paved road access, drainage, and local utilities. Data Quality Control and De-identification Cross-check field survey questionnaire forms against physical laboratory diagnostic slips to ensure consistency. Remove all personal identifying information, including names, contact numbers, and specific street addresses. Assign a unique anonymized record identifier to each entry. Check the compiled records for entry errors, duplicates, and missing values. Export the finalized cleaned dataset into a standard UTF-8 encoded CSV file (dataset.csv) for statistical analysis in Python, R, SPSS, or Stata.