MediBrief AI

Health TechIntermediateJan 9, 2026

MediBrief AI

AI-powered clinical note summarization for healthcare professionals.

The Problem

Independent healthcare practitioners such as specialized therapists, solo medical doctors, and medical researchers often face a significant administrative burden in managing and synthesizing patient clinical notes. They routinely need to create concise summaries for various critical purposes: insurance claim submissions, patient referrals to specialists, internal case studies, or contributing to research papers. Manually extracting relevant information from lengthy, unstructured clinical notes is incredibly time-consuming, often consuming 5-10 hours per week. This equates to $200-$500 in lost billable time or reduced research productivity. Current Electronic Health Record (EHR) systems typically offer generic reporting tools that are insufficient for context-specific summarization, or they require complex integrations and extensive training, which is cost-prohibitive for solo practitioners and small clinics. The risk of human error, oversight of crucial details, or inadvertent inclusion of sensitive but irrelevant information in summaries further compounds the problem, creating compliance and quality concerns.

The Solution

MediBrief AI provides an intuitive, secure platform for healthcare professionals to generate context-specific summaries of clinical notes using advanced AI. Users can securely upload or paste patient notes (with clear guidance on anonymization and PHI handling) and then select from a library of predefined summarization contexts like 'Insurance Claim Summary,' 'Specialist Referral Brief,' or 'Research Data Extract.' The core feature is an intelligent summarization engine that processes the input and delivers a concise, relevant output tailored to the chosen context. Users can also create and save their own custom summarization contexts with specific instructions, ensuring the AI consistently extracts the exact information they need. The platform emphasizes a clean user experience, robust data security, and audit trails for each summary generated. This unique approach bypasses the complexities and costs of full EHR integration, offering a highly focused tool that saves hours weekly, improves summary accuracy, and enhances compliance by ensuring only relevant information is presented.

Tech Stack

Frontend

Next.js 14ReactTailwind CSS

Backend

Next.js API RoutesNode.js

Database

PostgreSQL (Supabase)

APIs

Stripe (Payments)Resend (Email)OpenAI APIAnthropic APIGoogle Gemini API

System Architecture

┌─────────────────┐
│     Client      │
│  (Web Browser)  │
└─────────┬───────┘
          │ HTTPS
          ▼
┌─────────────────────────────────┐
│  Next.js App (Frontend & API)   │
│          (Vercel)               │
└─────────┬───────┬───────┬───────┤
          │ HTTPS │ HTTPS │ HTTPS │ HTTPS
          ▼       ▼       ▼       ▼
┌───────────────┐ ┌─────────┐ ┌─────────┐ ┌───────────────────┐
│   Supabase    │ │  Stripe │ │  Resend │ │ AI Provider (e.g., │
│ (Auth, DB, Storage) │ (Payments)│ (Email) │ │ OpenAI/Anthropic/Gemini) │
└───────────────┘ └─────────┘ └─────────┘ └───────────────────┘

Database Schema

-- Users table handled by Supabase Auth, but adding a profile table for custom fields.
CREATE TABLE profiles (
  id UUID PRIMARY KEY REFERENCES auth.users (id) ON DELETE CASCADE,
  full_name TEXT,
  organization TEXT,
  created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
  updated_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
);
COMMENT ON TABLE profiles IS 'User profiles extending Supabase auth.users';

CREATE TABLE summarization_contexts (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  user_id UUID REFERENCES auth.users (id) ON DELETE CASCADE,
  name TEXT NOT NULL,
  prompt_template TEXT NOT NULL, -- Template for the AI prompt, e.g., "Summarize for {purpose}: {text}"
  is_custom BOOLEAN DEFAULT TRUE, -- True for user-created, false for system-provided
  created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
  updated_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
);
CREATE INDEX idx_contexts_user_id ON summarization_contexts (user_id);
COMMENT ON TABLE summarization_contexts IS 'Predefined or custom templates for AI summarization tasks.';

CREATE TABLE clinical_notes (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  user_id UUID REFERENCES auth.users (id) ON DELETE CASCADE,
  original_text TEXT NOT NULL, -- The original clinical note content (emphasize user should anonymize PHI)
  context_id UUID REFERENCES summarization_contexts (id) ON DELETE SET NULL, -- Context used for summarization
  summary_text TEXT, -- The AI-generated summary
  status TEXT DEFAULT 'pending', -- 'pending', 'completed', 'failed'
  created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
  updated_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
  deleted_at TIMESTAMP WITH TIME ZONE -- Soft delete
);
CREATE INDEX idx_notes_user_id ON clinical_notes (user_id);
CREATE INDEX idx_notes_context_id ON clinical_notes (context_id);
COMMENT ON TABLE clinical_notes IS 'Stores original notes and their generated summaries.';

CREATE TABLE user_subscriptions (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  user_id UUID UNIQUE REFERENCES auth.users (id) ON DELETE CASCADE,
  stripe_customer_id TEXT UNIQUE,
  stripe_subscription_id TEXT UNIQUE,
  plan_id TEXT, -- e.g., 'free', 'pro', 'premium'
  current_period_end TIMESTAMP WITH TIME ZONE,
  status TEXT, -- e.g., 'active', 'canceled', 'past_due'
  created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
  updated_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
);
COMMENT ON TABLE user_subscriptions IS 'Manages user subscription status and Stripe integration details.';