My Documents — AI Document Q&A
Upload lab results, imaging reports, discharge summaries, or operative notes and get a plain-English breakdown powered by Gemini AI — plus a list of questions to bring to your next physician visit.
Document Q&A translates clinical language into plain English. It does not diagnose conditions, interpret results clinically, or recommend treatment changes. Always discuss your documents and findings with your physician before making any health decisions.
What You Can Upload
Accepted file formats: PDF, JPG, PNG, WebP · Maximum size: 10 MB per file
How to Use It
Find My Documents in the left sidebar under AI Tools, or go directly to /app/documents.php.
Drag and drop a PDF, JPG, or PNG onto the upload zone, or click to browse. Select the document type from the dropdown (Lab Result, Imaging Report, etc.) to help the AI calibrate its analysis.
Gemini 2.5 Pro reads your document and generates a structured analysis. This typically takes 15–30 seconds depending on document length and complexity.
Click the document card to expand the analysis. Review your plain-language summary, any flagged values, medical terms explained, and your list of questions for your next visit.
Copy or screenshot the "Questions to Bring to Your Next Visit" section. These are tailored specifically to the content of your document.
What the AI Returns
| Section | What it contains |
|---|---|
| Plain-Language Summary | 2–4 sentences explaining what the document shows overall, written for a general audience. |
| Key Findings | Up to 6 specific findings from the document, stated in plain English. |
| Values Outside Normal Range | For lab results: each abnormal value with its result, the reference range, and a plain-language explanation of what that marker measures. |
| Medical Terms Explained | Clinical terminology from the document defined in accessible language. |
| Questions for Your Doctor | 5 document-specific questions you should ask at your next visit — not generic, specific to what your document contains. |
Ensure the document is legible and not severely skewed. For multi-page lab panels, uploading the full PDF gives the AI the most context. For imaging reports, upload the written radiologist's report — not the scan image itself (the AI reads text, not diagnostic images).
Plan Limits
| Feature | Basic | Pro |
|---|---|---|
| Document uploads per month | 3 | Unlimited |
| AI analysis per upload | ✅ | ✅ |
| Saved document history | ✅ (until deleted) | ✅ Unlimited |
| Delete documents | ✅ | ✅ |
The monthly limit resets on the 1st of each calendar month. Documents already uploaded and analyzed remain accessible even after you reach the limit.
Privacy & Storage
- Documents are stored in a private, access-controlled directory on PancreaTrack servers — they are not publicly accessible.
- Files are stored per-user and can only be accessed through authenticated PHP endpoints. Direct URL access is blocked.
- Deleting a document from My Documents permanently removes both the file from disk and the analysis from the database.
- Documents are not shared with linked physicians. Only your logged pain, meal, bowel, lab, and glucose data is visible in the Physician Portal.
- Document content is sent to Google Gemini for analysis. Review PancreaTrack's Data Handling page and Google's AI data policies for full details.
Once you've discussed your document with your physician, you can delete it from My Documents if you no longer need the analysis stored in PancreaTrack. Your log data (pain, meals, etc.) is always separate and unaffected.