# 🧠 PyBrain - Unified Healthcare Intelligence Platform
[](https://www.python.org/downloads/)
[](LICENSE)
[](https://pypi.org/project/pybrain/)
PyBrain is the intelligence layer of the BrainSAIT Healthcare Unification Platform, providing AI-powered data harmonization, clinical NLP, and decision support for building next-generation healthcare systems.
## 🚀 Features
- **AI-Powered Data Harmonization**: Automatically maps and transforms data across different healthcare standards
- **Clinical NLP Engine**: Extracts structured data from unstructured clinical notes with medical language understanding
- **Federated Learning Framework**: Enables privacy-preserving AI model training across healthcare institutions
- **Real-time Decision Support**: Provides evidence-based recommendations using ensemble AI models
- **Predictive Analytics**: Forecasts patient outcomes, resource needs, and population health trends
## 📦 Installation
```bash
pip install pybrain
```
For development:
```bash
pip install pybrain[dev]
```
For all ML features:
```bash
pip install pybrain[ml,nlp]
```
## 🔧 Quick Start
### Basic Usage
```python
from pybrain import AIEngine, DataHarmonizer
# Initialize AI engine
ai = AIEngine()
# Extract entities from clinical text
clinical_note = "Patient presents with type 2 diabetes, prescribed metformin 500mg twice daily"
entities = ai.extract_clinical_entities(clinical_note)
print(entities)
# {'conditions': ['Diabetes'], 'medications': ['Metformin'], ...}
# Harmonize HL7v2 data to FHIR
harmonizer = DataHarmonizer()
hl7_data = {
"PID": {
"5": {"1": "Smith", "2": "John"},
"7": "19800415",
"8": "M"
}
}
fhir_patient = harmonizer.harmonize_to_fhir(hl7_data, "hl7v2", "Patient")
```
### AI-Powered Risk Assessment
```python
from pybrain import AIEngine, DecisionEngine
ai = AIEngine()
decision_engine = DecisionEngine()
# Patient data
patient_data = {
"age": 65,
"conditions": ["diabetes", "hypertension"],
"medications": ["metformin", "lisinopril"],
"bmi": 28.5
}
# Predict clinical risks
risk_score = ai.predict_risk_score(patient_data)
print(f"Overall risk score: {risk_score:.2f}")
# Get clinical recommendations
recommendations = decision_engine.evaluate_patient(patient_data)
print("Clinical alerts:", recommendations["alerts"])
```
### Population Health Analytics
```python
from pybrain import AnalyticsEngine
analytics = AnalyticsEngine()
# Analyze population trends
population_data = [
{"patient": {"id": "1", "birthDate": "1960-01-01"}, "observations": [...]},
{"patient": {"id": "2", "birthDate": "1975-05-15"}, "observations": [...]}
]
metrics = analytics.calculate_population_metrics(population_data)
print(f"High-risk patients: {metrics['risk_distribution']['high']}")
print(f"Recommendations: {metrics['recommendations']}")
```
### CLI Usage
```bash
# Analyze clinical text
pybrain analyze -t "Patient has hypertension and diabetes"
# Harmonize data files
pybrain harmonize -i patient.json -f hl7v2 -r Patient -o patient_fhir.json
# Start API server
pybrain serve --port 8000
```
## 🏗️ Architecture
PyBrain is designed as a modular, scalable platform:
```
pybrain/
├── core/
│ ├── ai/ # AI models and engines
│ ├── harmonizer/ # Data harmonization
│ ├── analytics/ # Analytics engine
│ ├── decision/ # Decision support
│ └── knowledge/ # Knowledge graphs
├── connectors/ # External system connectors
├── models/ # Pre-trained models
└── utils/ # Utilities
```
## 🤝 Integration with PyHeart
PyBrain works seamlessly with PyHeart for complete healthcare system unification:
```python
from pybrain import AIEngine
from pyheart import FHIRClient
# Use PyHeart for data access
client = FHIRClient("https://fhir.example.com")
patient_data = client.get_patient("12345")
# Use PyBrain for intelligence
ai = AIEngine()
risk_score = ai.predict_risk_score(patient_data)
if risk_score > 0.8:
print("High-risk patient - immediate intervention required")
```
## 🧪 Key Capabilities
### Clinical NLP
- Medical entity extraction
- Clinical concept normalization
- FHIR-compliant text processing
- Multi-language support
### AI-Powered Analytics
- Risk stratification
- Readmission prediction
- Fall risk assessment
- Medication adherence prediction
### Data Harmonization
- HL7v2 to FHIR transformation
- Custom EHR format mapping
- Terminology services integration
- Quality validation
### Decision Support
- Clinical rule engine
- Evidence-based recommendations
- Drug interaction checking
- Population health insights
## 📚 Documentation
Full documentation available at: https://pybrain.readthedocs.io
## 🧪 Testing
```bash
# Run tests
pytest
# With coverage
pytest --cov=pybrain
```
## 🤝 Contributing
We welcome contributions! Please see our Contributing Guide for details.
## 📄 License
PyBrain is licensed under the Apache License 2.0. See LICENSE for details.
## 🌟 Acknowledgments
Built with ❤️ by the BrainSAIT Healthcare Innovation Lab
Special thanks to the open-source healthcare community and all contributors.
---
**Together with PyHeart, PyBrain is building the future of intelligent healthcare.**
Raw data
{
"_id": null,
"home_page": null,
"name": "brainsait-pybrain",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "BrainSAIT Healthcare Innovation Lab <healthcare@brainsait.com>",
"keywords": "healthcare, ai, fhir, clinical-nlp, medical-ai, health-informatics, interoperability, decision-support, federated-learning, healthcare-analytics",
"author": null,
"author_email": "\"Dr. Fadil\" <fadil@brainsait.com>, BrainSAIT Team <team@brainsait.com>",
"download_url": "https://files.pythonhosted.org/packages/da/fc/42692e8a32207613212c0a818c83768779cac7693aaa1e10832df8f4c84c/brainsait_pybrain-0.1.0.tar.gz",
"platform": null,
"description": "# \ud83e\udde0 PyBrain - Unified Healthcare Intelligence Platform\n\n[](https://www.python.org/downloads/)\n[](LICENSE)\n[](https://pypi.org/project/pybrain/)\n\nPyBrain is the intelligence layer of the BrainSAIT Healthcare Unification Platform, providing AI-powered data harmonization, clinical NLP, and decision support for building next-generation healthcare systems.\n\n## \ud83d\ude80 Features\n\n- **AI-Powered Data Harmonization**: Automatically maps and transforms data across different healthcare standards\n- **Clinical NLP Engine**: Extracts structured data from unstructured clinical notes with medical language understanding\n- **Federated Learning Framework**: Enables privacy-preserving AI model training across healthcare institutions\n- **Real-time Decision Support**: Provides evidence-based recommendations using ensemble AI models\n- **Predictive Analytics**: Forecasts patient outcomes, resource needs, and population health trends\n\n## \ud83d\udce6 Installation\n\n```bash\npip install pybrain\n```\n\nFor development:\n```bash\npip install pybrain[dev]\n```\n\nFor all ML features:\n```bash\npip install pybrain[ml,nlp]\n```\n\n## \ud83d\udd27 Quick Start\n\n### Basic Usage\n\n```python\nfrom pybrain import AIEngine, DataHarmonizer\n\n# Initialize AI engine\nai = AIEngine()\n\n# Extract entities from clinical text\nclinical_note = \"Patient presents with type 2 diabetes, prescribed metformin 500mg twice daily\"\nentities = ai.extract_clinical_entities(clinical_note)\nprint(entities)\n# {'conditions': ['Diabetes'], 'medications': ['Metformin'], ...}\n\n# Harmonize HL7v2 data to FHIR\nharmonizer = DataHarmonizer()\nhl7_data = {\n \"PID\": {\n \"5\": {\"1\": \"Smith\", \"2\": \"John\"},\n \"7\": \"19800415\",\n \"8\": \"M\"\n }\n}\nfhir_patient = harmonizer.harmonize_to_fhir(hl7_data, \"hl7v2\", \"Patient\")\n```\n\n### AI-Powered Risk Assessment\n\n```python\nfrom pybrain import AIEngine, DecisionEngine\n\nai = AIEngine()\ndecision_engine = DecisionEngine()\n\n# Patient data\npatient_data = {\n \"age\": 65,\n \"conditions\": [\"diabetes\", \"hypertension\"],\n \"medications\": [\"metformin\", \"lisinopril\"],\n \"bmi\": 28.5\n}\n\n# Predict clinical risks\nrisk_score = ai.predict_risk_score(patient_data)\nprint(f\"Overall risk score: {risk_score:.2f}\")\n\n# Get clinical recommendations\nrecommendations = decision_engine.evaluate_patient(patient_data)\nprint(\"Clinical alerts:\", recommendations[\"alerts\"])\n```\n\n### Population Health Analytics\n\n```python\nfrom pybrain import AnalyticsEngine\n\nanalytics = AnalyticsEngine()\n\n# Analyze population trends\npopulation_data = [\n {\"patient\": {\"id\": \"1\", \"birthDate\": \"1960-01-01\"}, \"observations\": [...]},\n {\"patient\": {\"id\": \"2\", \"birthDate\": \"1975-05-15\"}, \"observations\": [...]}\n]\n\nmetrics = analytics.calculate_population_metrics(population_data)\nprint(f\"High-risk patients: {metrics['risk_distribution']['high']}\")\nprint(f\"Recommendations: {metrics['recommendations']}\")\n```\n\n### CLI Usage\n\n```bash\n# Analyze clinical text\npybrain analyze -t \"Patient has hypertension and diabetes\"\n\n# Harmonize data files\npybrain harmonize -i patient.json -f hl7v2 -r Patient -o patient_fhir.json\n\n# Start API server\npybrain serve --port 8000\n```\n\n## \ud83c\udfd7\ufe0f Architecture\n\nPyBrain is designed as a modular, scalable platform:\n\n```\npybrain/\n\u251c\u2500\u2500 core/\n\u2502 \u251c\u2500\u2500 ai/ # AI models and engines\n\u2502 \u251c\u2500\u2500 harmonizer/ # Data harmonization\n\u2502 \u251c\u2500\u2500 analytics/ # Analytics engine\n\u2502 \u251c\u2500\u2500 decision/ # Decision support\n\u2502 \u2514\u2500\u2500 knowledge/ # Knowledge graphs\n\u251c\u2500\u2500 connectors/ # External system connectors\n\u251c\u2500\u2500 models/ # Pre-trained models\n\u2514\u2500\u2500 utils/ # Utilities\n```\n\n## \ud83e\udd1d Integration with PyHeart\n\nPyBrain works seamlessly with PyHeart for complete healthcare system unification:\n\n```python\nfrom pybrain import AIEngine\nfrom pyheart import FHIRClient\n\n# Use PyHeart for data access\nclient = FHIRClient(\"https://fhir.example.com\")\npatient_data = client.get_patient(\"12345\")\n\n# Use PyBrain for intelligence\nai = AIEngine()\nrisk_score = ai.predict_risk_score(patient_data)\n\nif risk_score > 0.8:\n print(\"High-risk patient - immediate intervention required\")\n```\n\n## \ud83e\uddea Key Capabilities\n\n### Clinical NLP\n- Medical entity extraction\n- Clinical concept normalization\n- FHIR-compliant text processing\n- Multi-language support\n\n### AI-Powered Analytics\n- Risk stratification\n- Readmission prediction\n- Fall risk assessment\n- Medication adherence prediction\n\n### Data Harmonization\n- HL7v2 to FHIR transformation\n- Custom EHR format mapping\n- Terminology services integration\n- Quality validation\n\n### Decision Support\n- Clinical rule engine\n- Evidence-based recommendations\n- Drug interaction checking\n- Population health insights\n\n## \ud83d\udcda Documentation\n\nFull documentation available at: https://pybrain.readthedocs.io\n\n## \ud83e\uddea Testing\n\n```bash\n# Run tests\npytest\n\n# With coverage\npytest --cov=pybrain\n```\n\n## \ud83e\udd1d Contributing\n\nWe welcome contributions! Please see our Contributing Guide for details.\n\n## \ud83d\udcc4 License\n\nPyBrain is licensed under the Apache License 2.0. See LICENSE for details.\n\n## \ud83c\udf1f Acknowledgments\n\nBuilt with \u2764\ufe0f by the BrainSAIT Healthcare Innovation Lab\n\nSpecial thanks to the open-source healthcare community and all contributors.\n\n---\n\n**Together with PyHeart, PyBrain is building the future of intelligent healthcare.**\n",
"bugtrack_url": null,
"license": "Apache-2.0",
"summary": "Unified Healthcare Intelligence Platform - AI-powered healthcare data harmonization and decision support",
"version": "0.1.0",
"project_urls": {
"Changelog": "https://github.com/brainsait/pybrain/blob/main/CHANGELOG.md",
"Documentation": "https://pybrain.readthedocs.io",
"Homepage": "https://github.com/brainsait/pybrain",
"Issues": "https://github.com/brainsait/pybrain/issues",
"Repository": "https://github.com/brainsait/pybrain"
},
"split_keywords": [
"healthcare",
" ai",
" fhir",
" clinical-nlp",
" medical-ai",
" health-informatics",
" interoperability",
" decision-support",
" federated-learning",
" healthcare-analytics"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "f82c1f5e811de632eb02be64816be9f5aff34756da36583683336dae27815d67",
"md5": "05af8f3c2d5f1ad2ce1f01aec7ebcfad",
"sha256": "15b7f2e3f8a55935d96812f1d13ded1a1aa022e12846091f9596c565e31c5c5e"
},
"downloads": -1,
"filename": "brainsait_pybrain-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "05af8f3c2d5f1ad2ce1f01aec7ebcfad",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 23911,
"upload_time": "2025-08-02T17:15:56",
"upload_time_iso_8601": "2025-08-02T17:15:56.214193Z",
"url": "https://files.pythonhosted.org/packages/f8/2c/1f5e811de632eb02be64816be9f5aff34756da36583683336dae27815d67/brainsait_pybrain-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "dafc42692e8a32207613212c0a818c83768779cac7693aaa1e10832df8f4c84c",
"md5": "5cba31c15960188b300cedfe5defcf73",
"sha256": "010ac8796c2055504d215d217317296bd248355597742556a9bdffdaa02a9cf2"
},
"downloads": -1,
"filename": "brainsait_pybrain-0.1.0.tar.gz",
"has_sig": false,
"md5_digest": "5cba31c15960188b300cedfe5defcf73",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 22605,
"upload_time": "2025-08-02T17:15:57",
"upload_time_iso_8601": "2025-08-02T17:15:57.645433Z",
"url": "https://files.pythonhosted.org/packages/da/fc/42692e8a32207613212c0a818c83768779cac7693aaa1e10832df8f4c84c/brainsait_pybrain-0.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-08-02 17:15:57",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "brainsait",
"github_project": "pybrain",
"github_not_found": true,
"lcname": "brainsait-pybrain"
}
JSON
