Visual References (Clinical Photos)¶
Source: Call 35 (3-18) Priority: Tier 3 — depends on Bawa/Lauryn providing photos KB Requirement IDs: #121-#127
The Problem¶
Patients panic about how their surgery site looks. They call the office about bruising, swelling, discoloration — things that are completely normal but look alarming. These calls are avoidable if patients have a visual reference for what "normal" looks like.
Bawa: "We probably have so many pictures in our email and like text messages from patients or from nurses."
The photos already exist scattered across communications. They just need to be curated and integrated.
Requirements (All From 3-18)¶
Normal bruising/swelling photos (KB #121)¶
Bawa: "If we could somehow include or integrate pictures... this is what swelling can look like, this is what bruising can look like."
Include photos showing the range of normal post-operative appearance.
Preempt anxiety (KB #122)¶
Bawa: "Would like preempt some of the anxiety that people have about how bad it might look after."
Show these BEFORE patients encounter the issue — proactive education, not reactive reassurance.
Worst-case-but-normal examples (KB #123)¶
Lauryn: "Show them a worst case... but that's still normal." Bawa: "Yeah, exactly."
Both confirmed: show the most alarming-looking normal case so patients can self-calibrate. "My leg looks nowhere near that — oh, okay, good."
Educational context with images (KB #124)¶
Bawa: "Bruising is normal... gravity will pull it down, it's normal to have ankle swelling."
Photos should include explanatory text: - Bruising is normal post-surgery - Gravity causes bruising and swelling to move downward - Ankle swelling is common even after knee surgery - Colors change over time (red → purple → yellow → gone)
Allergic reaction examples (KB #125 — Open)¶
Bawa: "If we gave it like fifteen pictures of allergic reactions, would the AI be able to figure out that this looks like that?"
Exploratory. Uncertain feasibility. The AI might be able to compare patient-submitted photos to reference images, but this is not confirmed.
Human oversight required (KB #126)¶
Bawa: "We're still gonna have to have eyes on that."
Even if visual comparison is implemented, concerning cases still need human review. The AI can flag for attention but not diagnose.
Infection signs (KB #127 — Open)¶
Bawa: "This is what an allergic reaction may look like. These are signs of infection... temperature is always... fevers... chills."
Include content about infection warning signs alongside visual references. Help patients distinguish between normal and concerning.
Implementation Approach¶
Phase 1: Static visual education¶
- Curate 5-10 photos from Bawa/Lauryn showing range of normal post-op appearance
- Add to protocol as educational content triggered at appropriate post-op days
- Include in AI conversation responses when patients ask about swelling/bruising
- Format: photo + explanatory text + "If it looks worse than this, call the office"
Phase 2: Interactive visual comparison (future)¶
- Patient takes photo of their surgery site
- System shows reference photo for comparison
- AI provides contextual guidance
- Always includes "if concerned, call the office" escape hatch
Phase 3: AI-assisted visual triage (exploratory, deferred)¶
- Train on curated examples of normal vs concerning
- Requires significant clinical validation
- Always human-in-the-loop for any concerning classification
3/25 Visit Plan¶
Ask Lauryn to curate photos: - 3-5 examples of normal bruising progression (day 2, day 5, day 10, day 21) - 2-3 examples of normal swelling (including ankle swelling after knee surgery) - 1-2 examples of "worst case but still normal" - If available: examples of concerning signs (infection, allergic reaction) for internal reference
Photos should be de-identified. Get verbal consent for use in the system.
Content Integration¶
When patient says: "My knee is really bruised, is that normal?"
System should respond with: 1. Reassurance: "Some bruising after surgery is completely normal" 2. Visual reference: [photo of normal bruising at similar post-op day] 3. Education: "Bruising often moves down the leg due to gravity — you might even see it at your ankle" 4. Red flags: "If you notice increasing redness, warmth, or fever, that's when to call the office" 5. Always: "If you're concerned, you can always call us"