{"paper":{"title":"DeepFilters: Scattering-Aware Pupil Engineering with Learned Digital Filter Reconstruction for Extended Depth of Field Microscopy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Joint optimization of a pupil filter and digital reconstruction network extends depth of field microscopy through scattering tissue.","cross_cats":["cs.CV"],"primary_cat":"physics.optics","authors_text":"Alexandra Lion, Guorong Hu, Ian Davison, Jeffrey Alido, Joseph L. Greene, Kivilcim Kili\\c{c}, Lei Tian, Qilin Deng, Ruipeng Guo, Suet YIng Chan, Tongyu Li","submitted_at":"2026-05-13T14:49:33Z","abstract_excerpt":"Extended depth of field microscopy encodes axial information into a single acquisition through engineered point spread functions, but conventional and deep optics approaches are subject to degradation in scattering tissue. We introduce DeepFilters, a scattering-aware deep optics framework that jointly optimizes a parameterized pupil filter and a digital-filter-based reconstruction network through a calibrated differentiable forward model to achieve broad generalization without retraining. Incorporating empirical scattering kernels, physics-guided regularization, and a hybrid genetic-gradient i"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"DeepFilters extends the PSF from 16 micron to >400 micron in clear media and enables signal recovery beyond 120 micron deep in biological tissues, validated across fixed brain slices and sea urchin embryos.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The calibrated differentiable forward model, incorporating empirical scattering kernels, accurately represents real tissue scattering and enables generalization without retraining across different biological samples.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DeepFilters jointly optimizes a parameterized pupil filter and digital reconstruction network via a calibrated differentiable forward model with empirical scattering kernels to extend PSF depth from 16 to over 400 microns in clear media and recover signals beyond 120 microns in tissues.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Joint optimization of a pupil filter and digital reconstruction network extends depth of field microscopy through scattering tissue.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"3af2ad05f0f10b79d23d0ac25e2bb58863ceba6d4e8fbcd493a79e61a081d1b2"},"source":{"id":"2605.13619","kind":"arxiv","version":1},"verdict":{"id":"a98c3a49-ed17-4f2a-938d-651218f3444f","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T17:50:47.945791Z","strongest_claim":"DeepFilters extends the PSF from 16 micron to >400 micron in clear media and enables signal recovery beyond 120 micron deep in biological tissues, validated across fixed brain slices and sea urchin embryos.","one_line_summary":"DeepFilters jointly optimizes a parameterized pupil filter and digital reconstruction network via a calibrated differentiable forward model with empirical scattering kernels to extend PSF depth from 16 to over 400 microns in clear media and recover signals beyond 120 microns in tissues.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The calibrated differentiable forward model, incorporating empirical scattering kernels, accurately represents real tissue scattering and enables generalization without retraining across different biological samples.","pith_extraction_headline":"Joint optimization of a pupil filter and digital reconstruction network extends depth of field microscopy through scattering tissue."},"references":{"count":16,"sample":[{"doi":"10.1038/s41377-020-00403-7","year":2020,"title":"Miniscope3D: optimized single-shot miniature 3D fluorescence microscopy,","work_id":"4860c097-7c44-4530-9e27-677207c36a15","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1038/s41551-024-01226-2","year":2024,"title":"A miniaturized mesoscope for the large-scale single-neuron-resolved imaging of neuronal activity in freely behaving mice,","work_id":"543f8048-29ae-43da-b74c-9394644eef11","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1101/2024.10.07.616920","year":2024,"title":"T-scope V4: miniaturized microscope for optogenetic tagging in freely behaving animals,","work_id":"b7843b51-4d1b-4992-a0a1-ad830327edce","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1117/1.nph.10.4.044302","year":2023,"title":"Pupil engineering for extended depth-of-field imaging in a fluorescence miniscope","work_id":"a6de71fd-b4e9-48a0-9208-cfbcd1c778ce","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1073/pnas.2013571117","year":2020,"title":"Deep learning extended depth-of-field microscope for fast and slide-free histology,","work_id":"44f285bd-338f-44d8-b188-cbaaa9947223","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":16,"snapshot_sha256":"33f4a98cf4765333b0f0df0f757322605d5c89cd133613ecd1be392ffe598222","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}