Monday, July 15, 2024

data sources for AI in heathcare


https://www.nimhd.nih.gov/resources/schare/


  1. Jackson Heart Study (JHS):

    • The JHS focuses on cardiovascular disease (CVD) causes among African Americans.
    • With over 5,300 African American participants in Jackson, Mississippi, it’s one of the largest initiatives in this field.
    • The dataset covers diverse domains relevant to CVD, including demographics, anthropometrics, medication usage, conditions (e.g., hypertension, diabetes), lipid profiles, biomarkers, genetics, and more.
  2. ScHARe Data Ecosystem:

    • This ecosystem merges JHS data with area-level SDoH variables.
    • SDoH factors (like community resilience, socioeconomic status, and environmental conditions) play a crucial role in health outcomes.
    • By incorporating these factors, your AI/ML models can account for biases and enhance fairness.

1. **Social determinants of health data**:

   - URL: [Social Determinants of Health Data](https://healthdata.gov/dataset/social-determinants-health)


2. **Genomic data**:

   - URL: [National Center for Biotechnology Information (NCBI)](https://www.ncbi.nlm.nih.gov/)

   - URL: [Ensembl Genome Browser](https://www.ensembl.org/)


3. **Data on PTSD and burnout among clinicians**:

   - URL: [National Institute of Mental Health (NIMH) Data Archive](https://nda.nih.gov/)


4. **Referral networks data**:

   - This data may be more specific and proprietary, typically found through healthcare providers or specific research collaborations.


5. **Psychiatric patient data**:

   - URL: [National Institute of Mental Health (NIMH) Data Archive](https://nda.nih.gov/)


6. **Health records**:

   - URL: [HealthData.gov](https://www.healthdata.gov/)

   - URL: [Centers for Medicare & Medicaid Services (CMS)](https://www.cms.gov/Research-Statistics-Data-and-Systems/Research-Statistics-Data-and-Systems)


7. **Clinical trial data**:

   - URL: [ClinicalTrials.gov](https://clinicaltrials.gov/)


8. **Population health data**:

   - URL: [World Health Organization (WHO) Global Health Observatory](https://www.who.int/data/gho)

   - URL: [HealthData.gov](https://www.healthdata.gov/)


9. **Electronic health records**:

   - Typically proprietary, but some de-identified data sets can be found at:

     - URL: [MIMIC-III Clinical Database](https://mimic.physionet.org/)


10. **Survey data**:

    - URL: [CDC Behavioral Risk Factor Surveillance System (BRFSS)](https://www.cdc.gov/brfss/index.html)

    - URL: [National Health Interview Survey (NHIS)](https://www.cdc.gov/nchs/nhis/index.htm)


11. **Public health datasets**:

    - URL: [HealthData.gov](https://www.healthdata.gov/)

    - URL: [Centers for Disease Control and Prevention (CDC) Data and Statistics](https://www.cdc.gov/datastatistics/)


12. **Machine learning datasets**:

    - URL: [UCI Machine Learning Repository](https://archive.ics.uci.edu/ml/index.php)

    - URL: [Kaggle Datasets](https://www.kaggle.com/datasets)


13. **Biomedical data**:

    - URL: [Bioinformatics.org](https://www.bioinformatics.org/)

    - URL: [NIH Database of Genotypes and Phenotypes (dbGaP)](https://www.ncbi.nlm.nih.gov/gap)


14. **Hospital records**:

    - Typically proprietary, but aggregated data can be found at:

      - URL: [HealthData.gov](https://www.healthdata.gov/)

      - URL: [American Hospital Association (AHA) Data](https://www.aha.org/data)


15. **Mental health data**:

    - URL: [National Institute of Mental Health (NIMH) Data Archive](https://nda.nih.gov/)


16. **AI-generated data**:

    - This data is typically generated within research projects or specific AI applications, but some examples can be found in open repositories:

      - URL: [OpenAI](https://openai.com/)

      - URL: [Hugging Face Datasets](https://huggingface.co/datasets)


No comments:

Post a Comment