Medical Keyterm Correction

Medical keyterm correction uses biomedical Named Entity Recognition (NER) combined with LLM technology to automatically identify and correct medical terminology errors in transcriptions. It detects misspellings, misheard terms, and phonetic errors in medical vocabulary, then suggests or applies corrections to ensure clinical accuracy.

How to Enable

"enable_medical_correction": "true"

Option 1: Medical Keyterm Correction During Transcription

Request:

Don’t forget to replace YOUR_API_KEY with your own secret key.
import requests

url = "https://tb2.shunyalabs.ai/v1/transcriptions"
headers = {"X-API-Key": "your-api-key"}

with open("medical_consultation.wav", "rb") as f:
    files = {"file": f}
    data = {
        "enable_medical_correction": "true"
    }

    response = requests.post(url, headers=headers, files=files, data=data)
    print(response.json())

Example Output:

{
  "success": true,
  "text": "Patient presents with symptoms of hypertension including elevated blood pressure",
  "medical_correction": {
    "entities_found": 5,
    "entities_corrected": 3,
    "corrections": [
      {
        "original": "hipertenshun",
        "corrected": "hypertension"
      },
      {
        "original": "diabetis",
        "corrected": "diabetes mellitus"
      },
      {
        "original": "metforman",
        "corrected": "metformin"
      }
    ]
  }
}

Option 2: Medical Keyterm Correction with Standalone Text

Request:

Don’t forget to replace YOUR_API_KEY with your own secret key.
import requests

url = "https://tb.shunyalabs.ai/text_intelligence"
headers = {"X-API-Key": "your-api-key"}

data = {
  "text": "Patient has hipertenshun and diabetis. Prescribed metforman for blood sugar",
}

response = requests.post(url, headers=headers, data=data)
print(response.json())

Example Output:

{
  "status": "success",
  "original_text": "Patient has hipertenshun and diabetis. Prescribed metforman for blood sugar",
  "corrected_text": "Patient has hypertension and diabetes. Prescribed metformin for blood sugar",
  "entities_corrected": 3,
  "corrections": [
    {
      "original": "hipertenshun",
      "corrected": "hypertension"
    },
    {
      "original": "diabetis",
      "corrected": "diabetes"
    },
    {
      "original": "metforman",
      "corrected": "metformin"
    }
  ]
}

Use Cases for Medical Keyterm Correction

  • Clinical Documentation: Ensure accuracy in doctor-patient consultation transcriptions
  • Medical Dictation: Correct speech-to-text errors in physician notes
  • Telemedicine: Improve accuracy of remote consultation transcriptions
  • Medical Research: Clean transcripts from clinical interviews and studies
  • Healthcare Records: Maintain quality in electronic health record (EHR) documentation
  • Medical Education: Provide accurate transcripts of lectures and case discussions