Turn a photo of a hand-drawn ecomap into an editable canvas.
Snap a picture at the end of intake. Upload it. The AI reads the center, surrounding systems, and the line styles — and hands you an editable canvas you can refine.
Upload an imageWho this is for
Clinicians who still draw ecomaps by hand during intake but need a digital copy for the case file. Students who get template handouts. Anyone with paper-based assessments they want to digitize.
Social workers + case workers
Snap a picture of your intake whiteboard or paper template, get a clean digital version for the chart.
Therapists + counselors
Photograph the ecomap a client co-creates in session and have a sharable record by the time they leave.
Nursing + community health
Digitize paper assessments from home visits without re-typing every system label.
Educators + students
Convert handout templates from textbooks into editable canvases for class projects.
What the AI reads from an image
Center node
The bubble in the middle — name, type (person or family), any visible age or notes.
System nodes
Surrounding bubbles, with labels OCR-extracted and categories inferred from context.
Line styles
Thick = strong · thin = normal · dashed = weak · zig-zag = stressful · double zig-zag = stressful-strong · jagged = conflictual · broken = cut-off.
Arrowheads
Single arrow → directional energy flow · double arrow → mutual · no arrow → undirected.
Layout (auto-arranged)
Even if the photo is messy, the result is laid out in a clean Hartman ring on the canvas.
Editable output
Every node, label, edge, and category is editable. Wrong reads take seconds to correct.
Clinical Mode privacy
Photos of intake materials can contain identifying information. With Clinical Mode on, the image is sent to the AI provider once and not retained on our servers. The extracted ecomap document is encrypted at rest with AES-256-GCM before it lands in your library.
FAQ
What kind of images work?+
Clear photos, scans, or screenshots of hand-drawn or printed ecomaps. PNG, JPG, WebP, AVIF — up to 12 MB. A flat, well-lit shot reads better than a curled paper at an angle.
How long does it take?+
Five to fifteen seconds for typical inputs. Dense ecomaps with many systems and lateral edges can run up to thirty seconds.
How accurate is it?+
On clean printed ecomaps the AI reads center label, system labels, and line styles correctly the majority of the time. Hand-drawn handwriting is noisier — expect to do a quick visual pass and a 30-second sweep with the AI chat sidebar to correct any misreads.
Does the AI read line styles like strong / weak / stressful?+
Yes. The model is trained to decode thick (strong), thin (normal), dashed (weak), zig-zag (stressful), double-zig-zag (stressful-strong), jagged (conflictual), and broken (cut-off). Arrowheads map to energy-flow direction.
Is the image stored on your servers?+
In Clinical Mode the image is only sent to the AI provider once and not retained. Outside Clinical Mode the image is sent through our proxy and not stored. The resulting ecomap document is encrypted at rest with AES-256-GCM.
Free Downloadable Guides
Print-ready PDFs you can reference anytime — no sign-up required.