wisediag-medocr

What This Skill Does

The wisediag-medocr skill is a powerful tool designed to bridge the gap between static PDF documentation and machine-readable Markdown. Specializing in medical reports and complex layouts, it leverages the WiseDiag MedOcr API to perform high-fidelity optical character recognition. It is specifically engineered to preserve table structures and multi-column formatting, which are notoriously difficult for standard OCR tools. By running the medocr.py script, users can automate the conversion process, ensuring that medical records, diagnostic reports, and complex PDF documents are transformed into clean, searchable, and editable Markdown files without requiring manual transcription or complex API integration.

Installation

To integrate this skill into your environment, use the OpenClaw package manager. First, ensure you have the OpenClaw agent running, then execute the following command in your terminal: clawhub install openclaw/skills/skills/fmdmm/wisediag-medocr. After installation, navigate to the source directory and install the required dependencies using pip install -r requirements.txt. Finally, you must register for an API key at https://chat.wisediag.com/apiKeyManage and set it as an environment variable by running export WISEDIAG_API_KEY=your_api_key to enable authentication for the OCR services.

Use Cases

This skill is ideal for clinical research, health record digitalization, and administrative workflows. It is particularly effective for converting digitized legacy medical reports into data structures suitable for further analysis or archiving. Professionals handling high volumes of laboratory results, patient history records, or insurance documentation will find the table recognition features especially useful for maintaining data integrity during the conversion process.

Example Prompts

1. "Use MedOcr to process this lab report: /home/user/documents/blood_test_2023.pdf"
2. "I need to convert my medical history PDF. Can you run the MedOcr script on /downloads/patient_file.pdf and save it to the project folder?"
3. "Please process my latest diagnostic scan with MedOcr, use 300 DPI for better quality, and output the result to my reports directory."

Tips & Limitations

For optimal results, ensure your PDF scans are high-resolution; you can adjust the --dpi parameter when running the script if the output text is garbled. Note that this skill requires an active internet connection to communicate with the WiseDiag API. Always verify the output Markdown, as highly complex or handwritten medical notes may require occasional manual review for absolute clinical accuracy.