LIT (Layer-Wise Image Trajectories): In Situ Monitoring for Early Quality Prediction and Anomaly Detection in Acellular and Cell-Laden Two-Photon Polymerization
Two-photon polymerization (2PP) enables fabrication of hydrogel constructs with submicron, cell-scale resolution, but hydrogel-based bioinks are markedly more sensitive to process variability than conventional photoresists, and this sensitivity is further amplified when living cells are embedded in the resin. Post-processing evaluation, performed only after development, occurs too late to enable any corrective action. A full-factorial design of experiments across laser power and scan speed shows that fabrication outcome depends on both parameter choice and cell presence, with cells shifting and broadening the range of conditions yielding structurally sound constructs. However, substantial variability persists within each nominal condition and cannot be resolved by parameter refinement alone, indicating that outcome is governed by what occurs during each individual print rather than by the parameters set. To capture this, a layer-wise polymerization score is derived from pairwise comparisons of same-layer coaxial images, grounded in the psychophysics of relative judgment, and assembled into a Layer-wise Image Trajectory (LIT) for each print. Applied to both acellular and cell-laden formulations, LIT curves separate cleanly by post-processing outcome without any outcome label used in training, showing that fabrication quality can be predicted early in the build. Building on this signal, individual LIT curves are compared against statistical control limits derived from confirmed successful prints, enabling early detection of anomalous fabrication behavior at early-to-mid layers, well before development. To the best of the authors knowledge, this is the first application of in situ quality prediction and anomaly detection to cell-laden two-photon polymerization.