Documentation Index
Fetch the complete documentation index at: https://mintlify.com/sagar-grv/ayush-synapse/llms.txt
Use this file to discover all available pages before exploring further.
Traditional Indian medicine—Ayurveda, Siddha, and Unani—uses its own clinical terminology that has no direct equivalent in international biomedical coding systems like ICD-11. This creates an interoperability gap: an EMR system built around ICD codes cannot natively interpret a diagnosis recorded in NAMASTE codes, and vice versa. Ayush Synapse bridges this gap by maintaining a structured, FHIR R4-compliant mapping layer between the NAMASTE coding system and two ICD-11 target vocabularies—TM2 (Traditional Medicine Module 2) and Biomedicine—so that clinical records created in any AYUSH facility can be exchanged with any FHIR-compatible system without loss of meaning.
NAMASTE Codes
NAMASTE stands for National AYUSH Morbidity and Standardized Terminologies Electronic. It is a standardized coding system developed by India’s Ministry of AYUSH to represent the morbidity terms used across the three main traditional medicine disciplines:
- Ayurveda — classical Indian medicine based on tridosha (Vata, Pitta, Kapha) principles
- Siddha — a Tamil system of medicine rooted in alchemical and herbal traditions
- Unani — a Greco-Arabic tradition of medicine practised widely in South Asia
Each NAMASTE code identifies a single clinical condition within one of these disciplines. The code itself follows the format NAM-NNN (e.g., NAM-001), making it easy to sort and filter by system while maintaining a unique namespace for every condition.
NamasteCode Model
The NamasteCode Python class (defined in app/models/namaste_code.py) represents a single NAMASTE entry in-memory and exposes a FHIR serialisation method:
class NamasteCode:
def __init__(self, code, display, definition, system):
self.code = code
self.display = display
self.definition = definition
self.system = system # Ayurveda, Siddha, Unani
def to_fhir_codesystem(self):
return {
"code": self.code,
"display": self.display,
"definition": self.definition
}
def __repr__(self):
return f"NamasteCode(code={self.code}, display={self.display}, system={self.system})"
| Field | Type | Description |
|---|
code | string | Unique identifier in NAM-NNN format |
display | string | Human-readable name of the condition |
definition | string | Plain-language clinical description of the condition |
system | string | Originating discipline — one of Ayurveda, Siddha, or Unani |
Example NAMASTE Codes
The following codes are drawn directly from sample-namaste.csv and represent conditions across all three AYUSH disciplines:
| Code | Display | Definition | System |
|---|
| NAM-001 | Common Cold | An upper respiratory tract infection caused by various viruses | Ayurveda |
| NAM-007 | Jwara | Fever or elevated body temperature | Ayurveda |
| NAM-004 | Joint Pain | Pain in one or more joints | Siddha |
| NAM-005 | Digestive Issues | Problems with digestion or gastrointestinal function | Unani |
| NAM-021 | Madhumeha | Diabetes mellitus | Ayurveda |
| NAM-025 | Pandu | Anemia or pallor | Ayurveda |
| NAM-071 | Surra | Surra or trypanosomiasis | Unani |
| NAM-097 | Yara | Malaria | Unani |
ICD-11 Codes
The International Classification of Diseases, 11th Revision (ICD-11), published by the World Health Organization, is the global standard for recording diagnoses in clinical and research settings. Ayush Synapse works with two ICD-11 chapters that are relevant to traditional and complementary medicine:
- TM2 — Traditional Medicine Module 2: A dedicated ICD-11 module that provides codes for conditions recognised in East Asian, South Asian, and other traditional medicine systems. TM2 codes follow the format
TM2-NNN.
- Biomedicine: The conventional biomedical chapter of ICD-11. Biomedicine codes follow the format
BM-NNN. Mapping NAMASTE codes to Biomedicine targets allows AYUSH records to participate in conventional biomedical workflows.
Icd11Code Model
The Icd11Code Python class (defined in app/models/icd11_code.py) represents a single ICD-11 entry:
class Icd11Code:
def __init__(self, code, display, definition, chapter):
self.code = code
self.display = display
self.definition = definition
self.chapter = chapter # TM2 or Biomedicine
def to_fhir_codesystem(self):
return {
"code": self.code,
"display": self.display,
"definition": self.definition
}
def __repr__(self):
return f"Icd11Code(code={self.code}, display={self.display}, chapter={self.chapter})"
| Field | Type | Description |
|---|
code | string | Unique identifier — TM2-NNN for TM2, BM-NNN for Biomedicine |
display | string | Human-readable name of the ICD-11 concept |
definition | string | Clinical definition of the concept |
chapter | string | ICD-11 chapter — either TM2 or Biomedicine |
Example ICD-11 Codes
| Code | Display | Definition | Chapter |
|---|
| TM2-001 | Common Cold | Acute viral rhinitis | TM2 |
| TM2-002 | Headache | Pain in head or neck | TM2 |
| TM2-003 | Fever | Elevated body temperature | TM2 |
| TM2-004 | Joint Pain | Arthralgia | TM2 |
| TM2-005 | Digestive Issues | Gastrointestinal symptoms | TM2 |
| BM-001 | Acute rhinitis | Acute inflammation of the nasal mucosa | Biomedicine |
| BM-002 | Tension-type headache | Headache caused by muscle tension | Biomedicine |
| BM-004 | Arthralgia | Joint pain without inflammation | Biomedicine |
| BM-005 | Dyspepsia | Indigestion or stomach discomfort | Biomedicine |
Concept Mapping
A ConceptMap is the bridge between a NAMASTE source code and one or more ICD-11 target codes. Each mapping record captures not only which codes correspond but also the nature and confidence of that correspondence.
ConceptMap Model
The ConceptMap Python class (defined in app/models/concept_map.py) models a single directional mapping:
class ConceptMap:
def __init__(self, source_code, target_code, equivalence, source_system, target_system, confidence=1.0):
self.source_code = source_code
self.target_code = target_code
self.equivalence = equivalence # Types: equivalent, equal, wider, subsumes, etc.
self.source_system = source_system
self.target_system = target_system
self.confidence = confidence # 0.0 to 1.0
def to_fhir_conceptmap(self):
return {
"sourceCode": self.source_code,
"targetCode": self.target_code,
"equivalence": self.equivalence,
"confidence": self.confidence
}
def __repr__(self):
return f"ConceptMap(source={self.source_code}, target={self.target_code}, equivalence={self.equivalence})"
| Field | Type | Description |
|---|
source_code | string | NAMASTE code being mapped (e.g., NAM-001) |
target_code | string | ICD-11 code that the source maps to (e.g., TM2-001) |
equivalence | string | Semantic relationship between the two codes (see table below) |
source_system | string | Always NAMASTE in the current dataset |
target_system | string | Either ICD-11 TM2 or ICD-11 Biomedicine |
confidence | float | Mapping confidence score, from 0.0 (no confidence) to 1.0 (certainty) |
Equivalence Types
The equivalence field follows the FHIR ConceptMapEquivalence value set:
| Value | Meaning |
|---|
equivalent | The source and target concepts are clinically and semantically equivalent. |
equal | The source and target concepts are exactly the same (identical definitions and scope). |
wider | The target concept is broader than the source; the source is a subset of the target. |
subsumes | The target concept subsumes the source; the source is a specialisation of the target. |
Confidence Score
The confidence field is a floating-point value between 0.0 and 1.0 that reflects the certainty of the mapping:
- 0.9 – 1.0: High confidence — codes share near-identical clinical meaning and scope.
- 0.75 – 0.89: Moderate confidence — codes are clinically related but may differ in scope or granularity.
- Below 0.75: Low confidence — the mapping is approximate and should be reviewed by a clinical expert.
Example Mapping
The mapping below illustrates how NAM-001 (Ayurvedic Common Cold) is linked to TM2-001 (ICD-11 TM2 Common Cold) with high confidence:
ConceptMap(
source_code="NAM-001",
target_code="TM2-001",
equivalence="equivalent",
source_system="NAMASTE",
target_system="ICD-11 TM2",
confidence=0.9
)
NAM-011 (Arsha — piles or hemorrhoids) maps to BM-001 with a confidence of 0.87.
Coverage Statistics
The Ayush Synapse MVP ships with 85 pre-built concept mappings derived from the sample-namaste.csv dataset of 100 NAMASTE codes. This gives an out-of-the-box mapping coverage of 85%.
| Metric | Value |
|---|
| Total NAMASTE codes in dataset | 100 |
| Pre-built mappings | 85 |
| Coverage percentage | 85% |
| Target systems covered | TM2, Biomedicine |
| Minimum confidence score | 0.75 |
| Maximum confidence score | 0.95 |
In the current MVP, the WHO ICD-11 API integration (WhoApiService) returns sample data. Production deployment can integrate with the live WHO API by setting the WHO_API_KEY environment variable and replacing the placeholder fetch_tm2_codes() and fetch_biomedicine_codes() methods in app/services/who_api_service.py with real HTTP calls to https://icd.who.int/dev11.