Dataset Overview
| Attribute | Description |
|---|---|
| Dataset Name | Structured data sets for specialized diseases |
| Total Scale | 3 million + structured data recording for the disease |
| Covered Disease Categories | 26 priority diseases (several diseases — respiratory system — nervous system — tumor — oral oral — cardiovascular, etc.) |
| Data Dimensions | Clinical medical records • Follow-up records • Examination of image data • Genomics • Surgeon teaching video • Health science video – |
| Follow-up cycle | Data on longitudinal follow-up for most diseases over 5 years, partially queuing up to 10 years + |
| Source | Authorized to operate (three-academic hospital + multi-centre clinical research) |
| Use Cases | Real World Research, Drug Research, Model Training, Model Assessment, Clinical Science |
| Supply Format | Hard Drive · Cloud Drive · API · Data Infrastructure |
Dataset fields
| Category | Field | Description |
|---|---|---|
| Pathological markings | Name of disease ICD-10/11 code ICD-10 Class of disease I.A. | Standardized system for classification and coding of pathological diseases |
| Information on cases | Number of cases • Criteria for entry · Exclusion • Time of start and end of data collection | The size of the patient queue and the criteria for admission |
| Clinical indicators | Key indicators · Mixional diseases • Drug use record • Surgery/operation | Multimodular Data Dimension Matrix |
| Follow-up indicators | The follow-up cycle . End events . Survival . Rate of failure | Longitudinal follow-up completeness assessment |
| Data Source | The collection agency • the authorization method • the quality level • the final update | Data traceability and quality control ratings |
Priority disease coverage
| Classification of diseases | Covered Disease Categories | Core Data Dimensions |
|---|---|---|
| Metabolic diseases | Diabetes Obesity | Blood sugar surveillance, HbA1c, complications, dietary interventions, 5 years + follow-up |
| Respiratory system | Chronic obstructive pulmonary disease (COPD) | Lung function, GOLD grade, acute severity, history of smoking, CAT rating |
| Neural system | Multiple sclerosis .. autoimmune encephalitis .. neuromyelitis optica .. cerebral infarction .. sleep disorders .. brain tumor | MRI lesion count, EDSS rating, AQP4 antibody, immunotherapy treatment record |
| Cancer | Lung cancer, liver cancer, prostate cancer, bladder cancer, thyroid cancer, cerebral glioma. | TNM staging, pathology, gene mutation, surgical approach, survival follow-up |
| Heart vessels. | Cerebral hemorrhage, cerebral vessels, cerebral infarction. | NIHSS rating, ASPECTS rating, thrombectomy window, endovascular intervention program |
| Other | Intracranial aneurysm, meningitis, oral 3D model, stroke, ultrasound screening, medical video | CSF examinations, oral 3D model, stroke type, fetal biometry, surgical instruction |
De-identified JSON Sample
{ "record_id": "SPC_DM_2024_0047219", "disease": {"name":"2 diabetes mellitus","icd10":"E11.900","category": "specialty"}, "demographics": {"gender":"M","age_group":"55-60","bmi":27.8}, "indicators": { "hba1c":[{"date":"2024-01-15","value":8.2},{"date":"2024-06-20","value":7.1}], "fpg_mmol":7.6,"egfr":78, "complications":["sample","sample"] }, "medications": [{"drug": "sample value","dose":"500mg bid"},{"drug": "sample value","dose":"10mg qd"}], "follow_up": {"duration_years":6,"last_date":"2025-03-10","status":"active"}, "data_source": "sample value", "review_status":"approved"
}{ "record_id": "SPC_LC_2024_0018923", "disease": {"name": "Technique name","icd10":"C34.900","category": "specialty"}, "diagnosis": { "date":"2023-08-04","tnm_stage":"T2aN1M0,Stage IIB", "pathology": "sample value", "ihc":{"TTF1":"+","NapsinA":"+","PDL1":"30%"} }, "genomics": {"EGFR":"19del","ALK":"-","KRAS":"G12C","TP53":"R273H"}, "treatment": [ {"type":"surgery","procedure":"sample+sample","date":"2023-08-18"}, {"type":"chemo","regimen":"sample+CNx4sample"} ], "survival": {"last_date":"2025-06-01","status":"disease_free","dfs_months":22}, "data_source": "sample value", "review_status":"approved"
}{ "record_id": "SPC_STROKE_2024_0000456", "disease": {"name": "Technique name","icd10":"I63.9","category": "specialty"}, "acute_measures": { "nihss_admission":16,"nihss_24h":8,"aspect_score":7, "onset_to_door_min":72,"door_to_puncture_min":85 }, "procedure": { "type": "sample value","vessel":"CNM1sample", "tici_grade":"2b-3","devices":["Solitaire FR 6x30mm"] }, "imaging": {"initial":"CNM1sample","post_procedure":"mTICI 3sample"}, "outcome": {"mrs_90day":2,"hemorrhagic_transformation": "sample value"}, "data_source": "sample value", "review_status":"approved"
}{ "record_id": "SPC_MS_2024_0001732", "disease": {"name": "Technique name","icd10":"G35","category": "specialty"}, "phenotype": "sample(RRMS)", "diagnosis_date":"2020-03-15", "indicators": { "edss_scores":[{"date":"2020-03","score":2.5},{"date":"2024-09","score":3.0}], "annual_relapse_rate":0.5, "mri_lesions":{"brain_t2":12,"spinal_cord":2,"gd_enhancing":0} }, "treatments": [ {"drug": "sample value","start":"2021-06","status":"ongoing"}, {"drug": "sample value","purpose":"relapse","use_count":1} ], "follow_up": {"duration_years":5,"status":"stable"}, "data_source":"CNMSCN", "review_status":"approved"
}{ "record_id": "SPC_ORAL_2024_0000982", "disease": {"name": "Technique name","icd10":"K07.6","category": "specialty"}, "scan_type": "sample value", "model_format":"STL", "measurements": { "arch_width_maxillary_mm":48.5,"arch_width_mandibular_mm":52.1, "crowding_index": "sample value","molar_relationship":"Angle Class I" }, "quality_metrics": {"mesh_vertices":145632,"scan_completeness":0.98}, "applications": ["sample","sample","CNCAD/CAM"], "data_source": "sample value", "review_status":"approved"
}Annotation Specifications
Structured extraction
Automatically extract + manual verification from electronic medical records to ensure field completeness and accuracy
Standardized encoding
ICD-10/11 code map, SNOMED CT alignment, LOINC testing code standardization
Expert Review
Regular screening and review by clinical experts to ensure data quality and clinical logic
Privacy De-identification
Full data automated + manual double de-identification, consistent with personal information protection regulations
AI Application Scenarios
Disease risk prediction
Models for risk prediction based on structured clinical indicators + long-term follow-up results, training on diabetes complications, cardiovascular events, etc.
Real World Studies (RWE)
Big real world data supports evidence-based research such as drug efficacy comparisons, safety evaluations, clinical trial simulations, etc.
Treatment Program Recommendations
Analysis of the Relationship Between the Real-World Drug Model and Treatment Efficacy, and Construction of a Recommendation Engine for Individualized Treatment
AI for Chronic Disease Management
Multiple long-term follow-up datasets supporting intelligent monitoring and management interventions for diabetes and slowly progressive diseases such as COPD
Clinical Decision Support
CDSS system trained on large-scale specialist data to support decision-making by primary care physicians
Drug Research and Development
Real-world data supports the optimization of new drug target discovery, patient stratification, and clinical trial inclusion criteria
Commercial Value
Accelerated Clinical and Scientific Output
High-quality structured specialist disease data reduce the retrospective research cycle from months to weeks, improving the efficiency of scientific translation.
Real-World Evidence
Supporting post-market evaluation and the adaptation of medicines/tools in compliance with NMPA/CDE real-world research guidelines.
Commercially Usable in Compliance
The full data volume is processed with dual de-identification and authorization, in compliance with the Personal Information Protection Law and the Data Security Law, enabling authorized commercial use.
Flexible and Customized Combinations
26 pathologies are available on a needs basis to support customizable combinations by individual disease, unit packaging, and specialist queue.
Frequently asked questions
How does the disease-specific dataset differ from a generic electronic medical record dataset?
The disease-specific dataset provides deeply structured field design, long-term follow-up, and summaries of key indicators for specific diseases. For example, diabetes includes not only routine diagnostic indicators but also longitudinal data such as glycated hemoglobin sequences, complication screening history, and annual follow-up assessments, suitable for clinical research-level analysis.
Can I request data for a specific disease?
Pricing varies depending on the number of cases, the duration of follow-up visits, and the richness of fields.
Are the data de-identified and ethically approved?
Data are used in accordance with authorized operating agreements.
Access a Complete Catalog of Disease-Specific Datasets
Contact our data experts to obtain a detailed list of pathologies and a commercial proposal.