Maximum Likelihood Localization Maps

Maximum-likelihood localization maps (MLLM) reconstruct single-molecule localization microscopy by adding each molecule’s positional likelihood directly into an upsampled probability image, avoiding coordinate-based rendering.

• Why it matters: Coordinate-based rendering can discard the measured positional likelihood structure, and iterative localization can be slower or harder to scale.
• What they did: The method computes a non-iterative maximum-likelihood positional likelihood from measured data using a supplied, calibrated point-spread function, including local background variation and enabling 2D or 3D reconstruction.
• The result: MLLM preserves positional likelihood structure, supports any calibrated point-spread function, and can be parallelized for fast computation, enabling probability-map reconstructions without separate coordinate rendering.

AI summary of the abstract · may contain errors · not medical advice · read the paper.