What “local” means in SmileCrop
Recognition, cropping, orientation, naming, and storage run on the Windows computer in the practice. SmileCrop does not upload patient photographs or names to its servers for AI processing. The practice chooses the destination folder and any backup location for original files.
What still leaves the computer
Internet access is required for the trial, licence verification, and updates. The current product documentation says a full paid licence is checked at least once every 72 hours. This is not cloud photo processing, but it is still network communication that a practice should understand.
What local processing does not solve by itself
- access control for computers and shared folders,
- device and backup encryption,
- photo retention periods,
- secure disposal or reuse of storage media and SD cards,
- the practice's own organisational and contractual obligations.
Local architecture reduces the need to transfer patient photographs, but it does not replace internal security controls or legal assessment.
SmileCrop locally processes standard orthodontic intraoral and extraoral documentation photos. It is not a general platform for storing all healthcare data.
How to verify the data flow
Start with the full security and data flow overview and the rules for patient photographs. During the trial, use a copy of a test series, configure the destination folders, and verify where both processed output and original files are stored.
A local workflow from import to output
Photos can enter through a watched SD card or folder, be processed, and be saved without a cloud AI upload. The patient and QR steps are explained in the workflow guide.
Verify the flow in your own network
Test the trial on a computer and folder structure that represents your environment.