V2R SDK

Overview

Voice2Report facilitates radiology reporting using advanced speech recognition technology. Radiologists only need to talk about their pathological observations; the application efficiently transforms these verbal inputs into detailed, structured radiology reports with just one click.

Installation

Install the package via npm:

npm install mtl-v2r

Usage

Importing

Import the required module:

import useV2R from 'mtl-v2r';

Initialization

Create an instance by providing your API credentials and doctor ID:

const v2rInstance = useV2R({
  apiUrl: "<API_URL>",
  email: "<EMAIL>",
  password: "<PASSWORD>",
  doctor_id: "<DOCTOR_ID>"
});

Transcribe and Generate Report

Use the transcribeAndGenerateReport method to send a complete audio recording (as a Blob) and the name of your template. It returns a JSON object with the doctor_id, transcription and an array of reports.

  const audioBlob = /* Blob containing full recorded audio */;

  const result = await v2rInstance.transcribeAndGenerateReport(audioBlob);
  console.log(result);
  
  // Example output:
  // {
  //   transcription: "<TRANSCRIPTION>"
  //   doctor_id: "<DOCTOR_ID>",
  //   reports: [ /* Array of generated report objects */ ]
  // }

Upload Template

Use the uploadTemplate method to upload a new report template. Only .doc and .docx files are supported.

  const templateFile = /* File instance (.doc or .docx) */;
  await v2rInstance.uploadTemplate(templateFile);

useV2R Parameters

| Parameter | Type | Description | Required | |-------------|--------|-----------------------------------------|--------| | apiUrl | string | The base URL of the V2R API. | Yes | | email | string | Your account email for authentication. | Yes | | password | string | Your account password for authentication.| Yes | | doctor_id | string | The unique identifier for the doctor. | Yes |