Oncology-focused relational dataset derived from MIMIC
This project comprises a curated, oncology-focused relational dataset derived from the MIMIC-IV v3.1 clinical database, one of the largest publicly available medical datasets that encompass detailed patient-level information at the Beth Israel Deaconess Medical Center. It isolates and structures information relevant to cancer patients to support downstream analysis, visualization, or machine learning workflows, and serves as a foundation for bridging clinical data (MIMIC-IV) with molecular profiles (e.g., TCGA).
- Extracted key tables from MIMIC-IV (
patients,admissions,diagnoses_icd,icustays,prescriptions,labevents) in a local PostgreSQL setup - Filtered all data to focus on oncology-relevant fields:
- Diagnoses using cancer-related ICD-9 (140–239) and ICD-10 (C00–D49) codes
- Lab values including blood counts, liver enzymes, and tumor markers
- Medications relevant to chemotherapy, immunotherapy, or hormonal therapy
- Defined a derived oncology cohort for efficient filtering based on ICD logic
- Built supporting views for treatment timelines and labeled lab test values
- Optimized for efficient querying via indexes and materialized views
Metadata
- Source: MIMIC-IV v3.1
- Institution: Beth Israel Deaconess Medical Center (2008–2022)
- Modules Used:
hosp,icu - Programming: PostgreSQL, Python (ETL), SQL (views/indexes)
- Focus: Oncology-relevant EHR subset
- Derived From: Diagnoses, labs, prescriptions, ICU stays
- Primary Cohort Definition: Patients with ICD-9/10 neoplasm codes
- Compliance: PhysioNet DUA 1.5.0, HIPAA-deidentified, time-shifted
Core Schema Structure
The dataset is structured using 3 broad types of tables:
| Category | Description | Examples |
|---|---|---|
| Main Tables | Core entities: patient demographics, admissions, ICU stays, cohort flag | oncology_patients, oncology_admissions, oncology_icustays, oncology_cohort |
| Fact Tables | Event-level records: diagnoses, labs, prescriptions, EMAR (med admin records) | oncology_diagnoses, oncology_labs, oncology_prescriptions, oncology_emar_detail |
| Reference Tables | Dictionaries or derived views for enrichment or summarization | oncology_icd_dict, oncology_labs_with_labels, oncology_emar_detail_with_times, oncology_treatment_windows |
Use Cases
- Track cancer patients longitudinally (lab values, medications, ICU stays)
- Model treatment response or outcomes
- Build dashboards or visualizations for clinical metrics
- Practice clinical data engineering and SQL workflows
- Integrate with genomics registries (e.g. TCGA) for multimodal analysis