LunaLog

Anonymous symptom tracking and community-driven, data-based insights for PCOS and endometriosis.

This is a graduation project prototype: users can log weekly symptom updates and lifestyle factors, explore aggregated “insight previews” (correlations + symptom heatmap), and access a simple chat assistant. The long-term goal is to support both:

• User support through anonymous sharing and evidence-style summaries.
• Real-world research data by collecting de-identified, structured logs for statistical/ML analysis.

Medical disclaimer (mandatory)

This project provides no medical advice. It displays anonymized, aggregated community analysis and descriptive statistics. Correlations shown in the UI do not imply causation and may be biased, incomplete, or confounded. For diagnosis and treatment, consult a licensed clinician.

Privacy & anonymity

The prototype supports anonymous usage. In the current version:

• An anonymous user id is created and stored locally in your browser.
• Symptom logs are stored locally (localStorage).

When you add a real backend/database, logs should be de-identified server-side and stored as an anonymized dataset.

---

Features

• Hero + quick actions

  • Mission statement: ML-powered pattern discovery for PCOS/endometriosis symptoms
  • Anonymous registration/login CTA
  • “Log today’s symptoms” CTA

• Data-driven insights dashboard (preview)

  • Correlation widgets (mock data)
  • Symptom heatmap visualization (mock data)
  • Impact counters (“experiences → statistical insights”)

• Community & support (prototype UI)

  • Moderated blog highlights (mock data)
  • Forum sneak-peek (mock data)

• Simple AI chat widget (prototype)

  • Embedded floating chat
  • Safety guardrails: refuses medical advice

• Transparency & trust

  • Prominent disclaimer + privacy section in the footer

Pages

• Home: /
• Login (anonymous): /login
• Log symptoms: /log
• Admin: /admin

Admin is gated by an admin-only login. Use:

• Email: graduationproject@ius.edu.ba
• Password: gradproj2026

---

Tech stack

• Next.js (App Router)
• React
• TypeScript
• Tailwind CSS

---

Getting started

Prerequisites

• Node.js 18+ recommended
• npm

Install

npm install

Run locally (dev)

npm run dev

Open:

• http://localhost:3000

Production build

npm run build
npm start

---

Prototype limitations (current)

• No real database: insight widgets, forum topics, and blog highlights are mocked.
• Local-only data: anonymous user id and symptom logs are stored in the browser (localStorage).
• No real moderation: admin actions are placeholders for future workflows.
• No ML pipeline yet: the UI is prepared for correlation/clustering/predictive signals but isn’t connected to training/inference.

---

Roadmap (next steps)

• Backend + DB (Flask or Node)

  • store anonymized logs (PostgreSQL or MongoDB)
  • add moderated blog + forum models

• Analytics + ML

  • baseline descriptive stats + correlation analysis
  • clustering of symptom/lifestyle profiles
  • optional predictive modeling (carefully framed, non-diagnostic)
  • scheduled retraining (and later, online learning)

• Safety & governance

  • clearer data consent UX
  • moderation tools and abuse reporting
  • bias/confounder disclosures on insights

---

Repository structure (high level)

• src/app/ — Next.js routes (/, /login, /log, /admin)
• src/components/ — UI components (navbar, chat widget, search bar, footer)
• src/lib/ — prototype session helpers and mock data

License

Add a license file if you plan to open-source this project.