A photo editor helps you click faster. SmileCrop takes the routine clicks off your hands. It detects the shot type, crops, rotates, names and files each image to the patient folder. No cloud step. No typing patient names. No waiting at the computer.
From the SD card to the patient folder. The real difference is not the feature list — it is who does the work.
Nine steps. All of them yours.
Every image passes through human hands. A better editor makes those steps faster. It does not make them disappear.
Two steps yours. The rest is the software.
It knows the card by its volume serial, even when Windows gives it a different drive letter. Insert it and the batch starts on its own — no dialog, no babysitting.
Processing speed is not the point when the software runs unattended. The number worth watching is how many photos still come back to a person.
In an editor, a person must place, crop, straighten and check every image. A thousand photos means a thousand trips through the manual workflow.
In the founder’s own clinic SmileCrop processed 456 photos in a single batch and flagged 33 of them — 7.2%. For every 1,000, that puts roughly 72 in the review queue — each already classified, cropped and rotated, ready for a human decision.
A flagged photo is rarely work from scratch. The software has usually guessed the type and drawn the box — it is simply below the confidence threshold. You review the suggestion, adjust it where necessary, and save. Only a genuinely unrecognised photo needs more: you pick the type from a list, the software sets the aspect ratio and rotates the shot, and you adjust the crop and the angle.
We timed exactly that case: ten photos SmileCrop actually flagged out of the 456-photo batch. Not a demo on photos that suited it — these are its own worst cases, and some needed the name corrected as well. We processed the same ten twice: once by hand in another program, once in SmileCrop's bulk editor.
Cropping alone 4:47, renaming and filing into patient folders 0:46. The whole way from photo to folder.
Cropping, renaming and filing together. About 28% of the time — and even counting only the cropping on the manual side, it is 4:47 against 1:32.
Being straight about the measurement: the manual filing had it easy. There was no tree of hundreds of existing folders to hunt through and no new folder had to be created — in a real archive that 0:46 would be longer. Flagged photos are also the awkward ones by definition, so cropping them by hand takes longer than an ordinary shot too (hence 5:33 for ten, where a routine series came out at 24 s a photo). Ten photos, one person, one sitting — treat it as a demonstration, not a study. The full procedure and its weak spots.
What the 7.2% means: the 7.2% is a flag rate observed in one clinic on one batch of 456 photos, not a measurement of crop accuracy. The unflagged photos were not systematically rechecked for errors the software might have missed. Your result will depend on how consistently your team shoots. That is why the full trial runs on your own photos: don’t accept our number — measure yours.
MDDr. Michal Hasala built it for the recurring photo workflow in his own dental practice. It is not pretending to replace every photographic tool.
SmileCrop targets one expensive problem: the routine work that repeats patient after patient. That is the job automation should remove, not merely speed up.
Run SmileCrop on your own clinical series. See how many photos it flags and how much of your real workflow it takes over. Full version for 14 days. No credit card.
Windows 10/11 · All processing stays local and patient photos are never uploaded · No credit card · SmileQR is free