Machine LearningExperimental
ImageCleanerAI
Strip burned-in annotations from medical images via AI prompting
About
Streamlit app that removes burned-in labels, measurement lines, arrows, and patient info from medical images by routing them through fal.ai generative models. Four selectable model tiers (FLUX Dev, Kontext Pro, FLUX.2 Pro, Gemini 3.1 Flash) with per-image cost estimates. Built to preprocess training datasets without manual masking, fighting shortcut learning where classifiers learn to associate markings with diagnoses rather than actual anatomy.
Status
Working prototype. Quality varies by image type; useful as a first-pass cleaner before manual review.
Stack
PythonStreamlitfal.aiFLUXGemini