Intelligent SEEQPI enables reconstruction of inhibitory neuronal nuclei and dry-mass mapping deep in the living mouse brain

An intelligent label-free SEEQPI framework reconstructs inhibitory-neuron nuclei deep in the living mouse brain, accurately predicting nuclear distributions and enabling dry-mass density estimation.

• Why it matters: Deep multiphoton imaging is limited by photon budgets, tissue heating, and damage, and multiplexing molecular populations typically requires extra labels that add spectral crosstalk.
• What they did: The authors paired label-free Scattering-Enabled Epi-Quantitative Phase Imaging with three-photon fluorescence, training a Feature-wise Linear Modulation U-Net with generative adversarial learning to reconstruct inhibitory-neuron nuclei.
• The result: The network maintained performance with training/validation volumes separated axially by up to 6 m, improved predictions via cross-animal training in sparsely sampled targets, and inferred nuclear maps supported dry-mass density estimation.