MedAssist-TestRepo

MedAssist-Pro

<!-- markdownlint-disable first-line-h1 --> <!-- markdownlint-disable html --> <!-- markdownlint-disable no-duplicate-header --> <div align="center"> <img src="figures/fig1.png" width="60%" alt="MedAssist-Pro" /> </div> <hr> <div align="center" style="line-height: 1;"> <a href="LICENSE" style="margin: 2px;"> <img alt="License" src="figures/fig2.png" style="display: inline-block; vertical-align: middle;"/> </a> </div>

1. Introduction

MedAssist-Pro is a state-of-the-art medical AI assistant designed for clinical decision support. In the latest update, MedAssist-Pro has significantly improved its diagnostic accuracy and clinical reasoning capabilities through extensive training on curated medical datasets and reinforcement learning from medical expert feedback (RLMEF).

<p align="center"> <img width="80%" src="figures/fig3.png"> </p>

Compared to previous versions, the upgraded model demonstrates remarkable improvements in handling complex medical cases. For instance, in the MedQA benchmark, the model's accuracy increased from 65% to 82.3%. The model now provides more detailed clinical reasoning chains and cites relevant medical literature.

Beyond diagnostic capabilities, this version offers improved drug interaction detection and enhanced support for ICD-10 coding.

2. Evaluation Results

Comprehensive Medical Benchmark Results

<div align="center"> </div>

Overall Performance Summary

MedAssist-Pro demonstrates strong performance across all evaluated medical benchmark categories, with particularly notable results in diagnostic accuracy and clinical safety.

3. API Access & Clinical Integration

We offer HIPAA-compliant API endpoints for clinical integration. Please contact our enterprise team for details.

4. How to Run Locally

Please refer to our clinical deployment guide for running MedAssist-Pro in your healthcare environment.

Key deployment considerations:

1. HIPAA compliance must be ensured for patient data handling.
2. Model outputs should be reviewed by qualified medical professionals.

System Prompt

We recommend using the following system prompt:

You are MedAssist-Pro, a clinical decision support AI.
Current date: {current date}.
Always recommend consulting a healthcare provider for final decisions.

Temperature

We recommend setting temperature to 0.3 for clinical applications to ensure consistent outputs.

Clinical Note Processing

For processing clinical notes, use the following template:

clinical_template = \
"""[Patient ID]: {patient_id}
[Clinical Note Begin]
{clinical_note}
[Clinical Note End]
{clinical_question}"""

5. License

This model is licensed under the Apache 2.0 License. Clinical use requires appropriate medical supervision.

6. Contact

For questions, contact us at [email protected] or file an issue on our repository.