Dataset Overview
Definition：Health Checkup Datasetsis provided byChangsha Langhui Information Technology Co., Ltd.The data comprises 5 million genuine structured examinations obtained through authorized compliance transfer channels, sourced from 50+ collaborating hospitals (including 100 tertiary hospitals), with supporting documentation attached.Direct Authorization from Three Grade-A Tertiary HospitalsandCity-Level Comprehensive Authorization DocumentsThe data covers all age groups, with a 1:1 balanced sample of women and men, and medical examinations exceeding $1,000 to ensure quality and representativeness.
| Dataset Name | Medical Examination Datasets (including population stratification) |
| Total Data Volume | Five million real medical examinations. |
| Data Type | Structured Data |
| Data Source | Resale (partnerships with more than 50 hospitals, including Fudan Top-100 hospitals) |
| License Type | Direct authorization from the hospital. Direct Authorization from Three Grade-A Tertiary Hospitals + City-Level Comprehensive Authorization Documents |
| Delivery Cycle | 2 weeks - 1 month |
| Gender distribution | Male:female ≈1:1, even sampling |
| Age Layer | 10% under 18, 20% 18-30 years, 30% 30-40 years, 30% 40-60 years, 10% over 60 years |
| Amount of medical examination | Total amount of medical examination: £1,000 (middle-high-end medical examination package) |
Core Data Fields
Each medical examination record contains more than 80 structured fields covering the full medical examination dimension from basic information to video-check.
| Field Category | Field Count | Typical Fields |
|---|---|---|
| Demographic information | 8 | Sex, year of birth, age, height, weight, BMI, waist circumference, blood pressure (systolic/diastolic) |
| Blood protocol. | 15 | White blood cell count, red blood cell count, hemoglobin, platelet count, neutrophil percentage, lymphocyte percentage, etc. |
| Biochemical Total | 22 | ALT, AST, GGT, uric acid, eGFR, total cholesterol, triglycerides, LDL-C, fasting blood glucose, HbA1c |
| tumor markers | 12 | AFP, CEA, CA19-9, CA125, PSA Total/Free/PSA ratio (M), CA15-3 (F) |
| Image testing | 15 | CT-CV and Lung-RADS, carotid ultrasound IMT and stenosis, echocardiogram EF and E/A ratio. |
| Endoscopy | 8 | Gastroscopy findings, pathological results, H. pylori status; colonoscopy, size and location of lesions |
| Integrated assessment | 10 | Main abnormalities, risk assessment ranking, year-over-year comparison data, health recommendations |
Data Sample Showcase
The following is a sample of structured medical data after de-identification, showing typical field structure and data formats.
{ "anon_id": "PHY-2023-A8F2K9M4X7", "report_id": "RPT-S-BJXH-20230415-0032", "meta": { "quality_grade": "S", "source_institution": "partner medical checkup centers", "institution_tier": "premium", "checkup_date": "2023-04-15", "report_pages": 42, "original_price_cny": 6800 }, "demographics": { "gender": "male", "birth_year": 1975, "age_at_checkup": 48, "height_cm": 172.5, "weight_kg": 78.3, "bmi": 26.3, "waist_cm": 92, "blood_pressure_mmhg": {"systolic": 138, "diastolic": 88} }, "lab_results": { "blood_routine": {"wbc_10e9_L": 6.8, "hgb_g_L": 152, "plt_10e9_L": 218}, "biochemistry": { "alt_U_L": 42, "ggt_U_L": 68, "cr_umol_L": 88, "ua_umol_L": 420, "egfr_ml_min": 88, "tc_mmol_L": 5.82, "tg_mmol_L": 2.35, "ldl_c_mmol_L": 3.68, "fpg_mmol_L": 6.2, "hba1c_percent": 6.0 }, "tumor_markers": { "afp_ng_ml": 3.2, "cea_ng_ml": 2.1, "ca19_9_U_ml": 15.8, "psa_total_ng_ml": 2.8, "psa_ratio": 0.21 } }, "imaging": { "chest_ct": { "findings": "sample，sample，sample4mm，sample，Lung-RADS 2sample。", "lung_rads": "2", "nodule_count": 3, "largest_nodule_mm": 4 }, "carotid_us": { "findings": "sample，sample8.2x2.5mm，lumen stenosis ≈25%。", "imt_left_mm": 1.1, "stenosis_percent": 25 } }, "summary": { "main_findings": [ "sample", "sample（sample25%）", "sample（IFG），HbA1cCNelevated", "mixed hyperlipidemia", "sample（Lung-RADS 2sample）" ], "risk_stratification": "sample value" }
}AI Application Scenarios
Health Risk Assessment Model
Training of multi-dimensional health risk assessment models for cardiovascular disease, diabetes, tumors, and other health conditions based on large-scale medical data, enabling individualized risk prediction and stratified management.
Early disease prediction
Using rich biochemical indicators and imaging data, early warning models for diseases are being developed.
AI Interpretation of Medical Reports
Training the Health Management Master Model to automate the reading of medical examination reports, providing user-friendly health analysis and individualized health advice.
Health Management Assistant AI
Health management dialogue AI, based on 5 million balanced sampling cases, covers health counseling and management advice generation across all ages and genders.
Health Insurance Fees and Health Economics
Analyses of the consumption characteristics of medical examinations and disease spectrum for different age and gender groups, which provide data support for the actuarial and public health policies of health insurance.
Longitudinal Analysis of Chronic Disease Management
Through comparison of medical data over the years, training of chronic disease prediction models is carried out to achieve precision management of chronic diseases such as diabetes, hypertension, and hyperlipidemia.
Project value
Scale advantage — 5 million graded, balanced samples
Five million cases on a super-large scale ensure statistical visibility. The 1:1 ratio of women to men and five age groups (10–30% each) provide the AI model with sufficient sample diversity to effectively avoid gender and age biases.
Compliance Advantage — Triple Tertiary Hospital Authorization Guarantee
Unlike a single authorized source, this data set is accompanied by three tertiary hospitals with full city-level authorization documents, which form a multilayered legal firewall. This provides a significant advantage in compliance reviews for medical AI data.
Quality Advantage — $1,000+ High- to Mid-End Medical Examinations
The total cost of medical examination records exceeds $1,000, covering a comprehensive package of high- and mid-level examinations, including CT, ultrasound, tumor markers, etc.
Scenario Advantage — Health Management AI Across the Full Chain
This data set supports AHMA's full-chain training from disease prediction to report interpretation. It covers the full closed loop before (risk assessment), during (assisted diagnosis), and after (report interpretation) the examination.
Frequently asked questions
Need customized health data solutions?
Changsha Langhui Information Technology Co., Ltd. Datasets API platform, which provides one-stop medical data services from data collection to annotation to authorization.
Business consultation • Data sample preview • Customization solutions