GitHub Repository
Created On
Updated On
This application supports clinical reasoning processes by utilizing the Llama-2 large language model. It takes in user-inputted symptoms and generates a list of differential diagnoses and diagnostic workup suggestions. Note that this application is for experimental use only; it’s not intended for real world use and doesn’t replace medical advice.

More details

Use Cases Limitations Evidence Owner's Insight
  • Research: Researchers may employ the app to analyze how effectively an LLM can process and interpret medical data. This could lead to insights into improving natural language processing algorithms for medical applications.
  • Education: For educational purposes, the app can simulate the diagnostic reasoning process, allowing medical students to observe AI’s approach to clinical reasoning. It can stimulate discussion and critical thinking about the role of AI in medicine.
  • Not for Clinical Use: The app is experimental and is not intended for use in making real patient care decisions.
  • Response Time: Depending on latency and network speed, processing can take up to 1-2 minutes, although typically it completes within 40 seconds.
  • Service Availability: Hosted on Clarifai under the free community plan, the app may encounter service limitations once free request quotas are exceeded.
  • Data Recency: While the model has been tuned with data up to July 2023, some newer developments in medical science post-tuning may not be reflected.

Although this application isn’t peer-reviewed, the creator Benjamin Senst is a medical professional. Future updates will include feedback from initial users and healthcare professionals.

Benjamin Senst, a physician and data scientist (see for more information), created this app. His vision is to empower healthcare professionals with a tool that can assimilate and apply the latest medical information efficiently and effectively, thereby helping clinicians provide high quality care in a healthcare landscape characterized by high patient volume, complex health issues, and the necessity for prompt yet well-informed decision-making.

In the future, the owner plans to: - Fine-tune the Llama-2 LLM on public medical data not limited by security concerns. - Fine-tune it on medical records when security can be preserved. - Build additional modules beside differential diagnoses and proposals for diagnostic workup. - Scale the app to use more capable versions of Llama-2 as they become available.

Your insights and feedback on this app are greatly appreciated. If you have any thoughts, experiences, or feedback, please leave a comment below.


Warning: Not intended for clinical use. Assume outputs are unsafe and unvalidated. Use carefully.

  • Favorites: 1
  • Executions: 27

  • Natural Language Processing


A Anonymous User

Member since