Clariscan

Project: Clariscan

Self-reported basis only, unverified.

The description states that Clariscan is a patient-facing web application designed to simplify complex radiology reports using OCR and locally hosted LLMs. It is presented as an MVP built for the OpenAI 2026 hackathon, with no evidence of revenue, customers or traction beyond its author’s account.

What Changed: The project was submitted as a hackathon entry, indicating initial development and conceptualization. No prior version or evolution is evidenced.

Single Most Important Open Question: Is there any evidence that the product has been used by patients or integrated into healthcare workflows, or whether it has moved beyond an MVP?

The description states that Clariscan is a patient-facing web application. It uses:

• OCR (via pytesseract) to extract text from uploaded PDFs or images of medical reports.
• Locally hosted LLMs (via Ollama) to generate simplified summaries.
• Angular.js for the frontend.

It is described as an end-to-end solution that converts complex radiology reports into structured, easy-to-understand summaries. The output includes:

• Why the scan was performed
• Key findings in plain language
• An overall summary
• A risk indicator
• A safety disclaimer

The application allows users to download or print these summaries for use during consultations.

Inference: The product is a tool that aims to improve patient understanding of medical reports, not a diagnostic tool. It is built with privacy in mind, using local LLMs and avoiding external APIs.

The description states that Clariscan was inspired by the gap between receiving a radiology report and consulting a doctor, where patients often feel uncertain and turn to unreliable online sources. The product aims to bridge this gap by offering clear, reliable, and easy-to-understand explanations.

It positions itself as:

• A patient empowerment tool
• A privacy-focused solution (no reliance on external APIs)
• An AI-powered assistant that simplifies medical jargon without replacing professional advice

Inference: The positioning is focused on patient experience and accessibility, not on replacing doctors or becoming a diagnostic platform.

The description states that Clariscan targets patients who receive radiology reports and are in the period between receiving the report and consulting a doctor. It is described as helping patients feel more informed and less anxious during this time.

It also mentions:

• A potential roadmap to include laboratory reports and other medical documents
• Expansion to multilingual summaries
• Integration with hospitals and diagnostic centers

Inference: The primary ICP appears to be individual patients, but the product may evolve toward healthcare providers or institutions in the future.

The description does not state anything about a business model or pricing. It mentions:

• A roadmap that includes secure cloud-based report history for premium users
• Potential integration with hospitals and diagnostic centers via APIs

Inference: The product may evolve toward a freemium or B2B model, but no evidence of current monetization is provided.

The description states:

• Built using Angular.js, OCR (pytesseract), Ollama-LLM, and Python
• Uses locally hosted LLMs to avoid external API reliance
• Designed with a simple and intuitive interface
• Supports PDF or image uploads
• Generates downloadable summaries

Inference: The technical stack is basic but functional for an MVP. The use of local LLMs suggests a focus on privacy, but also implies limitations in scalability or performance.

The description states:

• It was built as an MVP for the OpenAI 2026 hackathon
• It is described as a fully functional MVP
• The team size is 1 person (Mahitha S)
• No evidence of revenue, customers, or adoption beyond the author’s account

Inference: There is no evidence of traction, usage, or market validation. The product remains in an early-stage prototype phase.

The description does not mention any competitors or similar products. It does not state whether there are existing tools that attempt to simplify medical reports for patients.

Inference: No competitive landscape is evident from the description. Clariscan may be a novel idea, but its uniqueness or market presence cannot be confirmed.

• No traction or usage data: The product is described only as an MVP.
• Single-founder team: Limited capacity for rapid development or scaling.
• Privacy vs. accuracy trade-off: Using local LLMs may limit accuracy, and the risk of misinterpretation remains.
• No commercialization strategy: No pricing, monetization or B2B plans are evident.
• Unverified claims: The product’s effectiveness in reducing anxiety or improving understanding is not substantiated.

1. Has Clariscan been tested with real patients? What feedback have you received?
2. How do you plan to ensure the accuracy of AI-generated summaries without overstepping into medical diagnosis?
3. Are there any legal or regulatory concerns around using AI to interpret medical reports?
4. What is your roadmap for monetization and scaling beyond the MVP stage?
5. How do you intend to integrate with hospitals or diagnostic centers, and what are the technical challenges?