AmareshHebbar/ayurveda-icd-sft

Ayurveda / Traditional Medicine to ICD-10 Bridge

Part of the AxisMapper Medical AI Suite — 16 domain-specific SFT datasets for fine-tuning medical LLMs.

Built by AmareshHebbar | Studio Ilios / Humanova Minds

---

What this dataset does

Traditional / Ayurvedic medicine concepts → nearest ICD-10-CM code

Why download this

Bridge traditional Indian medicine (Ayurveda, Unani, Siddha, Homeopathy) to ICD-10 for ABDM integration, ABHA health records, and integrative medicine clinical documentation.

Dataset stats

Split	Rows
Train	2,401
Validation	300
Test	301
Total	3,002

Data format

Every row is a messages list in chat format — compatible with Unsloth, TRL SFTTrainer, LLaMA-Factory, and any OpenAI-style fine-tuning pipeline:

{
  "messages": [
    {"role": "system",    "content": "You are a ..."},
    {"role": "user",      "content": "Traditional/integrative medicine concept: Contact with and exposure to other animate mechanical forces"},
    {"role": "assistant", "content": "ICD-10-CM Code: W64
Description: Contact with and exposure to other animate mechanical forces
Note: This code is used when traditional or alternative medicine context is clinically relevant."}
  ]
}

Data source

CMS FY2026 ICD-10-CM — Z codes (health factors) + Y codes (external causes) + traditional medicine terms → https://www.cms.gov/medicare/coding-billing/icd-10-codes

All data is extracted from authoritative public sources. No LLM-generated or synthetic content.

Who should use this

AYUSH practitioners, integrative medicine hospitals, ABDM health record implementers, Indian health IT developers.

Quick start

from datasets import load_dataset

ds = load_dataset("AmareshHebbar/ayurveda-icd-sft")
print(ds["train"][0])

Fine-tuning example (Unsloth)

from unsloth import FastLanguageModel
from trl import SFTTrainer
from datasets import load_dataset

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name="unsloth/Qwen2.5-3B-Instruct",
    max_seq_length=2048,
    load_in_4bit=True,
)

dataset = load_dataset("AmareshHebbar/ayurveda-icd-sft", split="train")

trainer = SFTTrainer(
    model=model,
    tokenizer=tokenizer,
    train_dataset=dataset,
    dataset_text_field="messages",
    max_seq_length=2048,
)
trainer.train()

Related datasets in this collection

Dataset	Task	Rows
icd10-coder-sft	ICD-10-CM coding	74.7k
symptom-diagnoser-sft	Symptom → diagnosis	119k
clinical-summarizer-sft	SOAP summarization	30k
discharge-qa-sft	Discharge summary QA	30k
pmjay-classifier-sft	PM-JAY packages	11.1k
radiology-coder-sft	Radiology coding	25k
medical-ner-sft	Clinical NER	16.7k
hindi-medical-sft	Hindi medical QA	19.7k

Citation

@misc{axiomapper2026,
  author    = {Hebbar, Amaresh},
  title     = {AxisMapper: Medical AI Fine-tuning Dataset Suite},
  year      = {2026},
  publisher = {HuggingFace},
  url       = {https://huggingface.co/collections/AmareshHebbar/axiomapper-medical-ai-suite}
}

---

AxisMapper is an open-source project. Star the repo, open issues, and contribute at GitHub.