Controlling False Positives in Image Segmentation via Conformal Prediction
This work addresses the need for risk-aware segmentation in clinical decision-making, where over-segmentation can have serious consequences, though it is incremental as it builds on existing conformal prediction methods.
The paper tackles the problem of unreliable semantic segmentation in clinical settings by introducing a post-hoc framework that uses conformal prediction to construct confidence masks with explicit statistical guarantees on false-positive control, achieving target-level empirical validity on a polyp-segmentation benchmark.
Reliable semantic segmentation is essential for clinical decision making, yet deep models rarely provide explicit statistical guarantees on their errors. We introduce a simple post-hoc framework that constructs confidence masks with distribution-free, image-level control of false-positive predictions. Given any pretrained segmentation model, we define a nested family of shrunken masks obtained either by increasing the score threshold or by applying morphological erosion. A labeled calibration set is used to select a single shrink parameter via conformal prediction, ensuring that, for new images that are exchangeable with the calibration data, the proportion of false positives retained in the confidence mask stays below a user-specified tolerance with high probability. The method is model-agnostic, requires no retraining, and provides finite-sample guarantees regardless of the underlying predictor. Experiments on a polyp-segmentation benchmark demonstrate target-level empirical validity. Our framework enables practical, risk-aware segmentation in settings where over-segmentation can have clinical consequences. Code at https://github.com/deel-ai-papers/conseco.